Filter.ExtractStyleBlocks

PAR1PAR1\n\nPAR2 + // ---- + // Note: This will always be the first child, since any + // previous inline element would have triggered this very + // same routine, and found the double newline. One possible + // exception would be a comment. + $token = array($this->_pStart(), $token); + } else { + // State 2.2.1:

PAR1

+ // ---- + + // State 2.2.2:

PAR1PAR1

+ // ---- + } + } + // Is the current parent a

tag? + } elseif ( + !empty($this->currentNesting) && + $this->currentNesting[count($this->currentNesting)-1]->name == 'p' + ) { + // State 3.1: ...

PAR1 + // ---- + + // State 3.2: ...

PAR1\n\nPAR2 + // ------------ + $token = array(); + $this->_splitText($text, $token); + // Abort! + } else { + // State 4.1: ...PAR1 + // ---- + + // State 4.2: ...PAR1\n\nPAR2 + // ------------ + } + } + + public function handleElement(&$token) { + // We don't have to check if we're already in a

tag for block + // tokens, because the tag would have been autoclosed by MakeWellFormed. + if ($this->allowsElement('p')) { + if (!empty($this->currentNesting)) { + if ($this->_isInline($token)) { + // State 1:

... + // --- + + // Check if this token is adjacent to the parent token + // (seek backwards until token isn't whitespace) + $i = null; + $this->backward($i, $prev); + + if (!$prev instanceof HTMLPurifier_Token_Start) { + // Token wasn't adjacent + + if ( + $prev instanceof HTMLPurifier_Token_Text && + substr($prev->data, -2) === "\n\n" + ) { + // State 1.1.4:

PAR1

\n\n + // --- + + // Quite frankly, this should be handled by splitText + $token = array($this->_pStart(), $token); + } else { + // State 1.1.1:

PAR1

+ // --- + + // State 1.1.2:

+ // --- + + // State 1.1.3:

PAR + // --- + } + + } else { + // State 1.2.1:

+ // --- + + // Lookahead to see if
is needed. + if ($this->_pLookAhead()) { + // State 1.3.1:

PAR1\n\nPAR2 + // --- + $token = array($this->_pStart(), $token); + } else { + // State 1.3.2:
PAR1
+ // --- + + // State 1.3.3:
PAR1
\n\n
+ // --- + } + } + } else { + // State 2.3: ...

+ // ----- + } + } else { + if ($this->_isInline($token)) { + // State 3.1: + // --- + // This is where the {p} tag is inserted, not reflected in + // inputTokens yet, however. + $token = array($this->_pStart(), $token); + } else { + // State 3.2:

+ // ----- + } + + $i = null; + if ($this->backward($i, $prev)) { + if ( + !$prev instanceof HTMLPurifier_Token_Text + ) { + // State 3.1.1: ...

{p} + // --- + + // State 3.2.1: ...

+ // ----- + + if (!is_array($token)) $token = array($token); + array_unshift($token, new HTMLPurifier_Token_Text("\n\n")); + } else { + // State 3.1.2: ...

\n\n{p} + // --- + + // State 3.2.2: ...
\n\n

+ // ----- + + // Note: PAR cannot occur because PAR would have been + // wrapped in

tags. + } + } + } + } else { + // State 2.2:

+ // ---- + + // State 2.4:
+ // --- + } + } + + /** + * Splits up a text in paragraph tokens and appends them + * to the result stream that will replace the original + * @param $data String text data that will be processed + * into paragraphs + * @param $result Reference to array of tokens that the + * tags will be appended onto + * @param $config Instance of HTMLPurifier_Config + * @param $context Instance of HTMLPurifier_Context + */ + private function _splitText($data, &$result) { + $raw_paragraphs = explode("\n\n", $data); + $paragraphs = array(); // without empty paragraphs + $needs_start = false; + $needs_end = false; + + $c = count($raw_paragraphs); + if ($c == 1) { + // There were no double-newlines, abort quickly. In theory this + // should never happen. + $result[] = new HTMLPurifier_Token_Text($data); + return; + } + for ($i = 0; $i < $c; $i++) { + $par = $raw_paragraphs[$i]; + if (trim($par) !== '') { + $paragraphs[] = $par; + } else { + if ($i == 0) { + // Double newline at the front + if (empty($result)) { + // The empty result indicates that the AutoParagraph + // injector did not add any start paragraph tokens. + // This means that we have been in a paragraph for + // a while, and the newline means we should start a new one. + $result[] = new HTMLPurifier_Token_End('p'); + $result[] = new HTMLPurifier_Token_Text("\n\n"); + // However, the start token should only be added if + // there is more processing to be done (i.e. there are + // real paragraphs in here). If there are none, the + // next start paragraph tag will be handled by the + // next call to the injector + $needs_start = true; + } else { + // We just started a new paragraph! + // Reinstate a double-newline for presentation's sake, since + // it was in the source code. + array_unshift($result, new HTMLPurifier_Token_Text("\n\n")); + } + } elseif ($i + 1 == $c) { + // Double newline at the end + // There should be a trailing
when we're finally done. + $needs_end = true; + } + } + } + + // Check if this was just a giant blob of whitespace. Move this earlier, + // perhaps? + if (empty($paragraphs)) { + return; + } + + // Add the start tag indicated by \n\n at the beginning of $data + if ($needs_start) { + $result[] = $this->_pStart(); + } + + // Append the paragraphs onto the result + foreach ($paragraphs as $par) { + $result[] = new HTMLPurifier_Token_Text($par); + $result[] = new HTMLPurifier_Token_End('p'); + $result[] = new HTMLPurifier_Token_Text("\n\n"); + $result[] = $this->_pStart(); + } + + // Remove trailing start token; Injector will handle this later if + // it was indeed needed. This prevents from needing to do a lookahead, + // at the cost of a lookbehind later. + array_pop($result); + + // If there is no need for an end tag, remove all of it and let + // MakeWellFormed close it later. + if (!$needs_end) { + array_pop($result); // removes \n\n + array_pop($result); // removes
+ } + + } + + /** + * Returns true if passed token is inline (and, ergo, allowed in + * paragraph tags) + */ + private function _isInline($token) { + return isset($this->htmlDefinition->info['p']->child->elements[$token->name]); + } + + /** + * Looks ahead in the token list and determines whether or not we need + * to insert a
tag. + */ + private function _pLookAhead() { + $this->current($i, $current); + if ($current instanceof HTMLPurifier_Token_Start) $nesting = 1; + else $nesting = 0; + $ok = false; + while ($this->forwardUntilEndToken($i, $current, $nesting)) { + $result = $this->_checkNeedsP($current); + if ($result !== null) { + $ok = $result; + break; + } + } + return $ok; + } + + /** + * Determines if a particular token requires an earlier inline token + * to get a paragraph. This should be used with _forwardUntilEndToken + */ + private function _checkNeedsP($current) { + if ($current instanceof HTMLPurifier_Token_Start){ + if (!$this->_isInline($current)) { + //
PAR1
+ // ---- + // Terminate early, since we hit a block element + return false; + } + } elseif ($current instanceof HTMLPurifier_Token_Text) { + if (strpos($current->data, "\n\n") !== false) { + //
PAR1PAR1\n\nPAR2 + // ---- + return true; + } else { + //
PAR1PAR1... + // ---- + } + } + return null; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Injector/DisplayLinkURI.php b/lib/htmlpurifier/library/HTMLPurifier/Injector/DisplayLinkURI.php new file mode 100644 index 00000000..9dce9bd0 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Injector/DisplayLinkURI.php @@ -0,0 +1,26 @@ +start->attr['href'])){ + $url = $token->start->attr['href']; + unset($token->start->attr['href']); + $token = array($token, new HTMLPurifier_Token_Text(" ($url)")); + } else { + // nothing to display + } + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Injector/Linkify.php b/lib/htmlpurifier/library/HTMLPurifier/Injector/Linkify.php new file mode 100644 index 00000000..296dac28 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Injector/Linkify.php @@ -0,0 +1,46 @@ + array('href')); + + public function handleText(&$token) { + if (!$this->allowsElement('a')) return; + + if (strpos($token->data, '://') === false) { + // our really quick heuristic failed, abort + // this may not work so well if we want to match things like + // "google.com", but then again, most people don't + return; + } + + // there is/are URL(s). Let's split the string: + // Note: this regex is extremely permissive + $bits = preg_split('#((?:https?|ftp)://[^\s\'"<>()]+)#S', $token->data, -1, PREG_SPLIT_DELIM_CAPTURE); + + $token = array(); + + // $i = index + // $c = count + // $l = is link + for ($i = 0, $c = count($bits), $l = false; $i < $c; $i++, $l = !$l) { + if (!$l) { + if ($bits[$i] === '') continue; + $token[] = new HTMLPurifier_Token_Text($bits[$i]); + } else { + $token[] = new HTMLPurifier_Token_Start('a', array('href' => $bits[$i])); + $token[] = new HTMLPurifier_Token_Text($bits[$i]); + $token[] = new HTMLPurifier_Token_End('a'); + } + } + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Injector/PurifierLinkify.php b/lib/htmlpurifier/library/HTMLPurifier/Injector/PurifierLinkify.php new file mode 100644 index 00000000..ad2455a9 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Injector/PurifierLinkify.php @@ -0,0 +1,45 @@ + array('href')); + + public function prepare($config, $context) { + $this->docURL = $config->get('AutoFormat.PurifierLinkify.DocURL'); + return parent::prepare($config, $context); + } + + public function handleText(&$token) { + if (!$this->allowsElement('a')) return; + if (strpos($token->data, '%') === false) return; + + $bits = preg_split('#%([a-z0-9]+\.[a-z0-9]+)#Si', $token->data, -1, PREG_SPLIT_DELIM_CAPTURE); + $token = array(); + + // $i = index + // $c = count + // $l = is link + for ($i = 0, $c = count($bits), $l = false; $i < $c; $i++, $l = !$l) { + if (!$l) { + if ($bits[$i] === '') continue; + $token[] = new HTMLPurifier_Token_Text($bits[$i]); + } else { + $token[] = new HTMLPurifier_Token_Start('a', + array('href' => str_replace('%s', $bits[$i], $this->docURL))); + $token[] = new HTMLPurifier_Token_Text('%' . $bits[$i]); + $token[] = new HTMLPurifier_Token_End('a'); + } + } + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Injector/RemoveEmpty.php b/lib/htmlpurifier/library/HTMLPurifier/Injector/RemoveEmpty.php new file mode 100644 index 00000000..638bfca0 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Injector/RemoveEmpty.php @@ -0,0 +1,51 @@ +config = $config; + $this->context = $context; + $this->removeNbsp = $config->get('AutoFormat.RemoveEmpty.RemoveNbsp'); + $this->removeNbspExceptions = $config->get('AutoFormat.RemoveEmpty.RemoveNbsp.Exceptions'); + $this->attrValidator = new HTMLPurifier_AttrValidator(); + } + + public function handleElement(&$token) { + if (!$token instanceof HTMLPurifier_Token_Start) return; + $next = false; + for ($i = $this->inputIndex + 1, $c = count($this->inputTokens); $i < $c; $i++) { + $next = $this->inputTokens[$i]; + if ($next instanceof HTMLPurifier_Token_Text) { + if ($next->is_whitespace) continue; + if ($this->removeNbsp && !isset($this->removeNbspExceptions[$token->name])) { + $plain = str_replace("\xC2\xA0", "", $next->data); + $isWsOrNbsp = $plain === '' || ctype_space($plain); + if ($isWsOrNbsp) continue; + } + } + break; + } + if (!$next || ($next instanceof HTMLPurifier_Token_End && $next->name == $token->name)) { + if ($token->name == 'colgroup') return; + $this->attrValidator->validateToken($token, $this->config, $this->context); + $token->armor['ValidateAttributes'] = true; + if (isset($token->attr['id']) || isset($token->attr['name'])) return; + $token = $i - $this->inputIndex + 1; + for ($b = $this->inputIndex - 1; $b > 0; $b--) { + $prev = $this->inputTokens[$b]; + if ($prev instanceof HTMLPurifier_Token_Text && $prev->is_whitespace) continue; + break; + } + // This is safe because we removed the token that triggered this. + $this->rewind($b - 1); + return; + } + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Injector/RemoveSpansWithoutAttributes.php b/lib/htmlpurifier/library/HTMLPurifier/Injector/RemoveSpansWithoutAttributes.php new file mode 100644 index 00000000..b2131347 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Injector/RemoveSpansWithoutAttributes.php @@ -0,0 +1,60 @@ +attrValidator = new HTMLPurifier_AttrValidator(); + $this->config = $config; + $this->context = $context; + return parent::prepare($config, $context); + } + + public function handleElement(&$token) { + if ($token->name !== 'span' || !$token instanceof HTMLPurifier_Token_Start) { + return; + } + + // We need to validate the attributes now since this doesn't normally + // happen until after MakeWellFormed. If all the attributes are removed + // the span needs to be removed too. + $this->attrValidator->validateToken($token, $this->config, $this->context); + $token->armor['ValidateAttributes'] = true; + + if (!empty($token->attr)) { + return; + } + + $nesting = 0; + $spanContentTokens = array(); + while ($this->forwardUntilEndToken($i, $current, $nesting)) {} + + if ($current instanceof HTMLPurifier_Token_End && $current->name === 'span') { + // Mark closing span tag for deletion + $current->markForDeletion = true; + // Delete open span tag + $token = false; + } + } + + public function handleEnd(&$token) { + if ($token->markForDeletion) { + $token = false; + } + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Injector/SafeObject.php b/lib/htmlpurifier/library/HTMLPurifier/Injector/SafeObject.php new file mode 100644 index 00000000..c1d8b041 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Injector/SafeObject.php @@ -0,0 +1,91 @@ + 'never', + 'allowNetworking' => 'internal', + ); + protected $allowedParam = array( + 'wmode' => true, + 'movie' => true, + 'flashvars' => true, + 'src' => true, + 'allowFullScreen' => true, // if omitted, assume to be 'false' + ); + + public function prepare($config, $context) { + parent::prepare($config, $context); + } + + public function handleElement(&$token) { + if ($token->name == 'object') { + $this->objectStack[] = $token; + $this->paramStack[] = array(); + $new = array($token); + foreach ($this->addParam as $name => $value) { + $new[] = new HTMLPurifier_Token_Empty('param', array('name' => $name, 'value' => $value)); + } + $token = $new; + } elseif ($token->name == 'param') { + $nest = count($this->currentNesting) - 1; + if ($nest >= 0 && $this->currentNesting[$nest]->name === 'object') { + $i = count($this->objectStack) - 1; + if (!isset($token->attr['name'])) { + $token = false; + return; + } + $n = $token->attr['name']; + // We need this fix because YouTube doesn't supply a data + // attribute, which we need if a type is specified. This is + // *very* Flash specific. + if (!isset($this->objectStack[$i]->attr['data']) && + ($token->attr['name'] == 'movie' || $token->attr['name'] == 'src')) { + $this->objectStack[$i]->attr['data'] = $token->attr['value']; + } + // Check if the parameter is the correct value but has not + // already been added + if ( + !isset($this->paramStack[$i][$n]) && + isset($this->addParam[$n]) && + $token->attr['name'] === $this->addParam[$n] + ) { + // keep token, and add to param stack + $this->paramStack[$i][$n] = true; + } elseif (isset($this->allowedParam[$n])) { + // keep token, don't do anything to it + // (could possibly check for duplicates here) + } else { + $token = false; + } + } else { + // not directly inside an object, DENY! + $token = false; + } + } + } + + public function handleEnd(&$token) { + // This is the WRONG way of handling the object and param stacks; + // we should be inserting them directly on the relevant object tokens + // so that the global stack handling handles it. + if ($token->name == 'object') { + array_pop($this->objectStack); + array_pop($this->paramStack); + } + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Language.php b/lib/htmlpurifier/library/HTMLPurifier/Language.php new file mode 100644 index 00000000..3e2be03b --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Language.php @@ -0,0 +1,163 @@ +config = $config; + $this->context = $context; + } + + /** + * Loads language object with necessary info from factory cache + * @note This is a lazy loader + */ + public function load() { + if ($this->_loaded) return; + $factory = HTMLPurifier_LanguageFactory::instance(); + $factory->loadLanguage($this->code); + foreach ($factory->keys as $key) { + $this->$key = $factory->cache[$this->code][$key]; + } + $this->_loaded = true; + } + + /** + * Retrieves a localised message. + * @param $key string identifier of message + * @return string localised message + */ + public function getMessage($key) { + if (!$this->_loaded) $this->load(); + if (!isset($this->messages[$key])) return "[$key]"; + return $this->messages[$key]; + } + + /** + * Retrieves a localised error name. + * @param $int integer error number, corresponding to PHP's error + * reporting + * @return string localised message + */ + public function getErrorName($int) { + if (!$this->_loaded) $this->load(); + if (!isset($this->errorNames[$int])) return "[Error: $int]"; + return $this->errorNames[$int]; + } + + /** + * Converts an array list into a string readable representation + */ + public function listify($array) { + $sep = $this->getMessage('Item separator'); + $sep_last = $this->getMessage('Item separator last'); + $ret = ''; + for ($i = 0, $c = count($array); $i < $c; $i++) { + if ($i == 0) { + } elseif ($i + 1 < $c) { + $ret .= $sep; + } else { + $ret .= $sep_last; + } + $ret .= $array[$i]; + } + return $ret; + } + + /** + * Formats a localised message with passed parameters + * @param $key string identifier of message + * @param $args Parameters to substitute in + * @return string localised message + * @todo Implement conditionals? Right now, some messages make + * reference to line numbers, but those aren't always available + */ + public function formatMessage($key, $args = array()) { + if (!$this->_loaded) $this->load(); + if (!isset($this->messages[$key])) return "[$key]"; + $raw = $this->messages[$key]; + $subst = array(); + $generator = false; + foreach ($args as $i => $value) { + if (is_object($value)) { + if ($value instanceof HTMLPurifier_Token) { + // factor this out some time + if (!$generator) $generator = $this->context->get('Generator'); + if (isset($value->name)) $subst['$'.$i.'.Name'] = $value->name; + if (isset($value->data)) $subst['$'.$i.'.Data'] = $value->data; + $subst['$'.$i.'.Compact'] = + $subst['$'.$i.'.Serialized'] = $generator->generateFromToken($value); + // a more complex algorithm for compact representation + // could be introduced for all types of tokens. This + // may need to be factored out into a dedicated class + if (!empty($value->attr)) { + $stripped_token = clone $value; + $stripped_token->attr = array(); + $subst['$'.$i.'.Compact'] = $generator->generateFromToken($stripped_token); + } + $subst['$'.$i.'.Line'] = $value->line ? $value->line : 'unknown'; + } + continue; + } elseif (is_array($value)) { + $keys = array_keys($value); + if (array_keys($keys) === $keys) { + // list + $subst['$'.$i] = $this->listify($value); + } else { + // associative array + // no $i implementation yet, sorry + $subst['$'.$i.'.Keys'] = $this->listify($keys); + $subst['$'.$i.'.Values'] = $this->listify(array_values($value)); + } + continue; + } + $subst['$' . $i] = $value; + } + return strtr($raw, $subst); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Language/classes/en-x-test.php b/lib/htmlpurifier/library/HTMLPurifier/Language/classes/en-x-test.php new file mode 100644 index 00000000..d52fcb7a --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Language/classes/en-x-test.php @@ -0,0 +1,12 @@ + 'HTML Purifier X' +); + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Language/messages/en-x-testmini.php b/lib/htmlpurifier/library/HTMLPurifier/Language/messages/en-x-testmini.php new file mode 100644 index 00000000..806c83fb --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Language/messages/en-x-testmini.php @@ -0,0 +1,12 @@ + 'HTML Purifier XNone' +); + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Language/messages/en.php b/lib/htmlpurifier/library/HTMLPurifier/Language/messages/en.php new file mode 100644 index 00000000..8d7b5736 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Language/messages/en.php @@ -0,0 +1,63 @@ + 'HTML Purifier', + +// for unit testing purposes +'LanguageFactoryTest: Pizza' => 'Pizza', +'LanguageTest: List' => '$1', +'LanguageTest: Hash' => '$1.Keys; $1.Values', + +'Item separator' => ', ', +'Item separator last' => ' and ', // non-Harvard style + +'ErrorCollector: No errors' => 'No errors detected. However, because error reporting is still incomplete, there may have been errors that the error collector was not notified of; please inspect the output HTML carefully.', +'ErrorCollector: At line' => ' at line $line', +'ErrorCollector: Incidental errors' => 'Incidental errors', + +'Lexer: Unclosed comment' => 'Unclosed comment', +'Lexer: Unescaped lt' => 'Unescaped less-than sign (<) should be <', +'Lexer: Missing gt' => 'Missing greater-than sign (>), previous less-than sign (<) should be escaped', +'Lexer: Missing attribute key' => 'Attribute declaration has no key', +'Lexer: Missing end quote' => 'Attribute declaration has no end quote', +'Lexer: Extracted body' => 'Removed document metadata tags', + +'Strategy_RemoveForeignElements: Tag transform' => '<$1> element transformed into $CurrentToken.Serialized', +'Strategy_RemoveForeignElements: Missing required attribute' => '$CurrentToken.Compact element missing required attribute $1', +'Strategy_RemoveForeignElements: Foreign element to text' => 'Unrecognized $CurrentToken.Serialized tag converted to text', +'Strategy_RemoveForeignElements: Foreign element removed' => 'Unrecognized $CurrentToken.Serialized tag removed', +'Strategy_RemoveForeignElements: Comment removed' => 'Comment containing "$CurrentToken.Data" removed', +'Strategy_RemoveForeignElements: Foreign meta element removed' => 'Unrecognized $CurrentToken.Serialized meta tag and all descendants removed', +'Strategy_RemoveForeignElements: Token removed to end' => 'Tags and text starting from $1 element where removed to end', +'Strategy_RemoveForeignElements: Trailing hyphen in comment removed' => 'Trailing hyphen(s) in comment removed', +'Strategy_RemoveForeignElements: Hyphens in comment collapsed' => 'Double hyphens in comments are not allowed, and were collapsed into single hyphens', + +'Strategy_MakeWellFormed: Unnecessary end tag removed' => 'Unnecessary $CurrentToken.Serialized tag removed', +'Strategy_MakeWellFormed: Unnecessary end tag to text' => 'Unnecessary $CurrentToken.Serialized tag converted to text', +'Strategy_MakeWellFormed: Tag auto closed' => '$1.Compact started on line $1.Line auto-closed by $CurrentToken.Compact', +'Strategy_MakeWellFormed: Tag carryover' => '$1.Compact started on line $1.Line auto-continued into $CurrentToken.Compact', +'Strategy_MakeWellFormed: Stray end tag removed' => 'Stray $CurrentToken.Serialized tag removed', +'Strategy_MakeWellFormed: Stray end tag to text' => 'Stray $CurrentToken.Serialized tag converted to text', +'Strategy_MakeWellFormed: Tag closed by element end' => '$1.Compact tag started on line $1.Line closed by end of $CurrentToken.Serialized', +'Strategy_MakeWellFormed: Tag closed by document end' => '$1.Compact tag started on line $1.Line closed by end of document', + +'Strategy_FixNesting: Node removed' => '$CurrentToken.Compact node removed', +'Strategy_FixNesting: Node excluded' => '$CurrentToken.Compact node removed due to descendant exclusion by ancestor element', +'Strategy_FixNesting: Node reorganized' => 'Contents of $CurrentToken.Compact node reorganized to enforce its content model', +'Strategy_FixNesting: Node contents removed' => 'Contents of $CurrentToken.Compact node removed', + +'AttrValidator: Attributes transformed' => 'Attributes on $CurrentToken.Compact transformed from $1.Keys to $2.Keys', +'AttrValidator: Attribute removed' => '$CurrentAttr.Name attribute on $CurrentToken.Compact removed', + +); + +$errorNames = array( + E_ERROR => 'Error', + E_WARNING => 'Warning', + E_NOTICE => 'Notice' +); + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/LanguageFactory.php b/lib/htmlpurifier/library/HTMLPurifier/LanguageFactory.php new file mode 100644 index 00000000..134ef8c7 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/LanguageFactory.php @@ -0,0 +1,198 @@ +cache[$language_code][$key] = $value + * @value array map + */ + public $cache; + + /** + * Valid keys in the HTMLPurifier_Language object. Designates which + * variables to slurp out of a message file. + * @value array list + */ + public $keys = array('fallback', 'messages', 'errorNames'); + + /** + * Instance of HTMLPurifier_AttrDef_Lang to validate language codes + * @value object HTMLPurifier_AttrDef_Lang + */ + protected $validator; + + /** + * Cached copy of dirname(__FILE__), directory of current file without + * trailing slash + * @value string filename + */ + protected $dir; + + /** + * Keys whose contents are a hash map and can be merged + * @value array lookup + */ + protected $mergeable_keys_map = array('messages' => true, 'errorNames' => true); + + /** + * Keys whose contents are a list and can be merged + * @value array lookup + */ + protected $mergeable_keys_list = array(); + + /** + * Retrieve sole instance of the factory. + * @param $prototype Optional prototype to overload sole instance with, + * or bool true to reset to default factory. + */ + public static function instance($prototype = null) { + static $instance = null; + if ($prototype !== null) { + $instance = $prototype; + } elseif ($instance === null || $prototype == true) { + $instance = new HTMLPurifier_LanguageFactory(); + $instance->setup(); + } + return $instance; + } + + /** + * Sets up the singleton, much like a constructor + * @note Prevents people from getting this outside of the singleton + */ + public function setup() { + $this->validator = new HTMLPurifier_AttrDef_Lang(); + $this->dir = HTMLPURIFIER_PREFIX . '/HTMLPurifier'; + } + + /** + * Creates a language object, handles class fallbacks + * @param $config Instance of HTMLPurifier_Config + * @param $context Instance of HTMLPurifier_Context + * @param $code Code to override configuration with. Private parameter. + */ + public function create($config, $context, $code = false) { + + // validate language code + if ($code === false) { + $code = $this->validator->validate( + $config->get('Core.Language'), $config, $context + ); + } else { + $code = $this->validator->validate($code, $config, $context); + } + if ($code === false) $code = 'en'; // malformed code becomes English + + $pcode = str_replace('-', '_', $code); // make valid PHP classname + static $depth = 0; // recursion protection + + if ($code == 'en') { + $lang = new HTMLPurifier_Language($config, $context); + } else { + $class = 'HTMLPurifier_Language_' . $pcode; + $file = $this->dir . '/Language/classes/' . $code . '.php'; + if (file_exists($file) || class_exists($class, false)) { + $lang = new $class($config, $context); + } else { + // Go fallback + $raw_fallback = $this->getFallbackFor($code); + $fallback = $raw_fallback ? $raw_fallback : 'en'; + $depth++; + $lang = $this->create($config, $context, $fallback); + if (!$raw_fallback) { + $lang->error = true; + } + $depth--; + } + } + + $lang->code = $code; + + return $lang; + + } + + /** + * Returns the fallback language for language + * @note Loads the original language into cache + * @param $code string language code + */ + public function getFallbackFor($code) { + $this->loadLanguage($code); + return $this->cache[$code]['fallback']; + } + + /** + * Loads language into the cache, handles message file and fallbacks + * @param $code string language code + */ + public function loadLanguage($code) { + static $languages_seen = array(); // recursion guard + + // abort if we've already loaded it + if (isset($this->cache[$code])) return; + + // generate filename + $filename = $this->dir . '/Language/messages/' . $code . '.php'; + + // default fallback : may be overwritten by the ensuing include + $fallback = ($code != 'en') ? 'en' : false; + + // load primary localisation + if (!file_exists($filename)) { + // skip the include: will rely solely on fallback + $filename = $this->dir . '/Language/messages/en.php'; + $cache = array(); + } else { + include $filename; + $cache = compact($this->keys); + } + + // load fallback localisation + if (!empty($fallback)) { + + // infinite recursion guard + if (isset($languages_seen[$code])) { + trigger_error('Circular fallback reference in language ' . + $code, E_USER_ERROR); + $fallback = 'en'; + } + $language_seen[$code] = true; + + // load the fallback recursively + $this->loadLanguage($fallback); + $fallback_cache = $this->cache[$fallback]; + + // merge fallback with current language + foreach ( $this->keys as $key ) { + if (isset($cache[$key]) && isset($fallback_cache[$key])) { + if (isset($this->mergeable_keys_map[$key])) { + $cache[$key] = $cache[$key] + $fallback_cache[$key]; + } elseif (isset($this->mergeable_keys_list[$key])) { + $cache[$key] = array_merge( $fallback_cache[$key], $cache[$key] ); + } + } else { + $cache[$key] = $fallback_cache[$key]; + } + } + + } + + // save to cache for later retrieval + $this->cache[$code] = $cache; + + return; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Length.php b/lib/htmlpurifier/library/HTMLPurifier/Length.php new file mode 100644 index 00000000..8d2a46b7 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Length.php @@ -0,0 +1,115 @@ + true, 'ex' => true, 'px' => true, 'in' => true, + 'cm' => true, 'mm' => true, 'pt' => true, 'pc' => true + ); + + /** + * @param number $n Magnitude + * @param string $u Unit + */ + public function __construct($n = '0', $u = false) { + $this->n = (string) $n; + $this->unit = $u !== false ? (string) $u : false; + } + + /** + * @param string $s Unit string, like '2em' or '3.4in' + * @warning Does not perform validation. + */ + static public function make($s) { + if ($s instanceof HTMLPurifier_Length) return $s; + $n_length = strspn($s, '1234567890.+-'); + $n = substr($s, 0, $n_length); + $unit = substr($s, $n_length); + if ($unit === '') $unit = false; + return new HTMLPurifier_Length($n, $unit); + } + + /** + * Validates the number and unit. + */ + protected function validate() { + // Special case: + if ($this->n === '+0' || $this->n === '-0') $this->n = '0'; + if ($this->n === '0' && $this->unit === false) return true; + if (!ctype_lower($this->unit)) $this->unit = strtolower($this->unit); + if (!isset(HTMLPurifier_Length::$allowedUnits[$this->unit])) return false; + // Hack: + $def = new HTMLPurifier_AttrDef_CSS_Number(); + $result = $def->validate($this->n, false, false); + if ($result === false) return false; + $this->n = $result; + return true; + } + + /** + * Returns string representation of number. + */ + public function toString() { + if (!$this->isValid()) return false; + return $this->n . $this->unit; + } + + /** + * Retrieves string numeric magnitude. + */ + public function getN() {return $this->n;} + + /** + * Retrieves string unit. + */ + public function getUnit() {return $this->unit;} + + /** + * Returns true if this length unit is valid. + */ + public function isValid() { + if ($this->isValid === null) $this->isValid = $this->validate(); + return $this->isValid; + } + + /** + * Compares two lengths, and returns 1 if greater, -1 if less and 0 if equal. + * @warning If both values are too large or small, this calculation will + * not work properly + */ + public function compareTo($l) { + if ($l === false) return false; + if ($l->unit !== $this->unit) { + $converter = new HTMLPurifier_UnitConverter(); + $l = $converter->convert($l, $this->unit); + if ($l === false) return false; + } + return $this->n - $l->n; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Lexer.php b/lib/htmlpurifier/library/HTMLPurifier/Lexer.php new file mode 100644 index 00000000..9bdbbbb2 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Lexer.php @@ -0,0 +1,326 @@ +get('Core.LexerImpl'); + } + + $needs_tracking = + $config->get('Core.MaintainLineNumbers') || + $config->get('Core.CollectErrors'); + + $inst = null; + if (is_object($lexer)) { + $inst = $lexer; + } else { + + if (is_null($lexer)) { do { + // auto-detection algorithm + + if ($needs_tracking) { + $lexer = 'DirectLex'; + break; + } + + if ( + class_exists('DOMDocument') && + method_exists('DOMDocument', 'loadHTML') && + !extension_loaded('domxml') + ) { + // check for DOM support, because while it's part of the + // core, it can be disabled compile time. Also, the PECL + // domxml extension overrides the default DOM, and is evil + // and nasty and we shan't bother to support it + $lexer = 'DOMLex'; + } else { + $lexer = 'DirectLex'; + } + + } while(0); } // do..while so we can break + + // instantiate recognized string names + switch ($lexer) { + case 'DOMLex': + $inst = new HTMLPurifier_Lexer_DOMLex(); + break; + case 'DirectLex': + $inst = new HTMLPurifier_Lexer_DirectLex(); + break; + case 'PH5P': + $inst = new HTMLPurifier_Lexer_PH5P(); + break; + default: + throw new HTMLPurifier_Exception("Cannot instantiate unrecognized Lexer type " . htmlspecialchars($lexer)); + } + } + + if (!$inst) throw new HTMLPurifier_Exception('No lexer was instantiated'); + + // once PHP DOM implements native line numbers, or we + // hack out something using XSLT, remove this stipulation + if ($needs_tracking && !$inst->tracksLineNumbers) { + throw new HTMLPurifier_Exception('Cannot use lexer that does not support line numbers with Core.MaintainLineNumbers or Core.CollectErrors (use DirectLex instead)'); + } + + return $inst; + + } + + // -- CONVENIENCE MEMBERS --------------------------------------------- + + public function __construct() { + $this->_entity_parser = new HTMLPurifier_EntityParser(); + } + + /** + * Most common entity to raw value conversion table for special entities. + */ + protected $_special_entity2str = + array( + '"' => '"', + '&' => '&', + '<' => '<', + '>' => '>', + ''' => "'", + ''' => "'", + ''' => "'" + ); + + /** + * Parses special entities into the proper characters. + * + * This string will translate escaped versions of the special characters + * into the correct ones. + * + * @warning + * You should be able to treat the output of this function as + * completely parsed, but that's only because all other entities should + * have been handled previously in substituteNonSpecialEntities() + * + * @param $string String character data to be parsed. + * @returns Parsed character data. + */ + public function parseData($string) { + + // following functions require at least one character + if ($string === '') return ''; + + // subtracts amps that cannot possibly be escaped + $num_amp = substr_count($string, '&') - substr_count($string, '& ') - + ($string[strlen($string)-1] === '&' ? 1 : 0); + + if (!$num_amp) return $string; // abort if no entities + $num_esc_amp = substr_count($string, '&'); + $string = strtr($string, $this->_special_entity2str); + + // code duplication for sake of optimization, see above + $num_amp_2 = substr_count($string, '&') - substr_count($string, '& ') - + ($string[strlen($string)-1] === '&' ? 1 : 0); + + if ($num_amp_2 <= $num_esc_amp) return $string; + + // hmm... now we have some uncommon entities. Use the callback. + $string = $this->_entity_parser->substituteSpecialEntities($string); + return $string; + } + + /** + * Lexes an HTML string into tokens. + * + * @param $string String HTML. + * @return HTMLPurifier_Token array representation of HTML. + */ + public function tokenizeHTML($string, $config, $context) { + trigger_error('Call to abstract class', E_USER_ERROR); + } + + /** + * Translates CDATA sections into regular sections (through escaping). + * + * @param $string HTML string to process. + * @returns HTML with CDATA sections escaped. + */ + protected static function escapeCDATA($string) { + return preg_replace_callback( + '//s', + array('HTMLPurifier_Lexer', 'CDATACallback'), + $string + ); + } + + /** + * Special CDATA case that is especially convoluted for )#si', + array($this, 'scriptCallback'), $html); + } + + $html = $this->normalize($html, $config, $context); + + $cursor = 0; // our location in the text + $inside_tag = false; // whether or not we're parsing the inside of a tag + $array = array(); // result array + + // This is also treated to mean maintain *column* numbers too + $maintain_line_numbers = $config->get('Core.MaintainLineNumbers'); + + if ($maintain_line_numbers === null) { + // automatically determine line numbering by checking + // if error collection is on + $maintain_line_numbers = $config->get('Core.CollectErrors'); + } + + if ($maintain_line_numbers) { + $current_line = 1; + $current_col = 0; + $length = strlen($html); + } else { + $current_line = false; + $current_col = false; + $length = false; + } + $context->register('CurrentLine', $current_line); + $context->register('CurrentCol', $current_col); + $nl = "\n"; + // how often to manually recalculate. This will ALWAYS be right, + // but it's pretty wasteful. Set to 0 to turn off + $synchronize_interval = $config->get('Core.DirectLexLineNumberSyncInterval'); + + $e = false; + if ($config->get('Core.CollectErrors')) { + $e =& $context->get('ErrorCollector'); + } + + // for testing synchronization + $loops = 0; + + while(++$loops) { + + // $cursor is either at the start of a token, or inside of + // a tag (i.e. there was a < immediately before it), as indicated + // by $inside_tag + + if ($maintain_line_numbers) { + + // $rcursor, however, is always at the start of a token. + $rcursor = $cursor - (int) $inside_tag; + + // Column number is cheap, so we calculate it every round. + // We're interested at the *end* of the newline string, so + // we need to add strlen($nl) == 1 to $nl_pos before subtracting it + // from our "rcursor" position. + $nl_pos = strrpos($html, $nl, $rcursor - $length); + $current_col = $rcursor - (is_bool($nl_pos) ? 0 : $nl_pos + 1); + + // recalculate lines + if ( + $synchronize_interval && // synchronization is on + $cursor > 0 && // cursor is further than zero + $loops % $synchronize_interval === 0 // time to synchronize! + ) { + $current_line = 1 + $this->substrCount($html, $nl, 0, $cursor); + } + + } + + $position_next_lt = strpos($html, '<', $cursor); + $position_next_gt = strpos($html, '>', $cursor); + + // triggers on "asdf" but not "asdf " + // special case to set up context + if ($position_next_lt === $cursor) { + $inside_tag = true; + $cursor++; + } + + if (!$inside_tag && $position_next_lt !== false) { + // We are not inside tag and there still is another tag to parse + $token = new + HTMLPurifier_Token_Text( + $this->parseData( + substr( + $html, $cursor, $position_next_lt - $cursor + ) + ) + ); + if ($maintain_line_numbers) { + $token->rawPosition($current_line, $current_col); + $current_line += $this->substrCount($html, $nl, $cursor, $position_next_lt - $cursor); + } + $array[] = $token; + $cursor = $position_next_lt + 1; + $inside_tag = true; + continue; + } elseif (!$inside_tag) { + // We are not inside tag but there are no more tags + // If we're already at the end, break + if ($cursor === strlen($html)) break; + // Create Text of rest of string + $token = new + HTMLPurifier_Token_Text( + $this->parseData( + substr( + $html, $cursor + ) + ) + ); + if ($maintain_line_numbers) $token->rawPosition($current_line, $current_col); + $array[] = $token; + break; + } elseif ($inside_tag && $position_next_gt !== false) { + // We are in tag and it is well formed + // Grab the internals of the tag + $strlen_segment = $position_next_gt - $cursor; + + if ($strlen_segment < 1) { + // there's nothing to process! + $token = new HTMLPurifier_Token_Text('<'); + $cursor++; + continue; + } + + $segment = substr($html, $cursor, $strlen_segment); + + if ($segment === false) { + // somehow, we attempted to access beyond the end of + // the string, defense-in-depth, reported by Nate Abele + break; + } + + // Check if it's a comment + if ( + substr($segment, 0, 3) === '!--' + ) { + // re-determine segment length, looking for --> + $position_comment_end = strpos($html, '-->', $cursor); + if ($position_comment_end === false) { + // uh oh, we have a comment that extends to + // infinity. Can't be helped: set comment + // end position to end of string + if ($e) $e->send(E_WARNING, 'Lexer: Unclosed comment'); + $position_comment_end = strlen($html); + $end = true; + } else { + $end = false; + } + $strlen_segment = $position_comment_end - $cursor; + $segment = substr($html, $cursor, $strlen_segment); + $token = new + HTMLPurifier_Token_Comment( + substr( + $segment, 3, $strlen_segment - 3 + ) + ); + if ($maintain_line_numbers) { + $token->rawPosition($current_line, $current_col); + $current_line += $this->substrCount($html, $nl, $cursor, $strlen_segment); + } + $array[] = $token; + $cursor = $end ? $position_comment_end : $position_comment_end + 3; + $inside_tag = false; + continue; + } + + // Check if it's an end tag + $is_end_tag = (strpos($segment,'/') === 0); + if ($is_end_tag) { + $type = substr($segment, 1); + $token = new HTMLPurifier_Token_End($type); + if ($maintain_line_numbers) { + $token->rawPosition($current_line, $current_col); + $current_line += $this->substrCount($html, $nl, $cursor, $position_next_gt - $cursor); + } + $array[] = $token; + $inside_tag = false; + $cursor = $position_next_gt + 1; + continue; + } + + // Check leading character is alnum, if not, we may + // have accidently grabbed an emoticon. Translate into + // text and go our merry way + if (!ctype_alpha($segment[0])) { + // XML: $segment[0] !== '_' && $segment[0] !== ':' + if ($e) $e->send(E_NOTICE, 'Lexer: Unescaped lt'); + $token = new HTMLPurifier_Token_Text('<'); + if ($maintain_line_numbers) { + $token->rawPosition($current_line, $current_col); + $current_line += $this->substrCount($html, $nl, $cursor, $position_next_gt - $cursor); + } + $array[] = $token; + $inside_tag = false; + continue; + } + + // Check if it is explicitly self closing, if so, remove + // trailing slash. Remember, we could have a tag like
, so + // any later token processing scripts must convert improperly + // classified EmptyTags from StartTags. + $is_self_closing = (strrpos($segment,'/') === $strlen_segment-1); + if ($is_self_closing) { + $strlen_segment--; + $segment = substr($segment, 0, $strlen_segment); + } + + // Check if there are any attributes + $position_first_space = strcspn($segment, $this->_whitespace); + + if ($position_first_space >= $strlen_segment) { + if ($is_self_closing) { + $token = new HTMLPurifier_Token_Empty($segment); + } else { + $token = new HTMLPurifier_Token_Start($segment); + } + if ($maintain_line_numbers) { + $token->rawPosition($current_line, $current_col); + $current_line += $this->substrCount($html, $nl, $cursor, $position_next_gt - $cursor); + } + $array[] = $token; + $inside_tag = false; + $cursor = $position_next_gt + 1; + continue; + } + + // Grab out all the data + $type = substr($segment, 0, $position_first_space); + $attribute_string = + trim( + substr( + $segment, $position_first_space + ) + ); + if ($attribute_string) { + $attr = $this->parseAttributeString( + $attribute_string + , $config, $context + ); + } else { + $attr = array(); + } + + if ($is_self_closing) { + $token = new HTMLPurifier_Token_Empty($type, $attr); + } else { + $token = new HTMLPurifier_Token_Start($type, $attr); + } + if ($maintain_line_numbers) { + $token->rawPosition($current_line, $current_col); + $current_line += $this->substrCount($html, $nl, $cursor, $position_next_gt - $cursor); + } + $array[] = $token; + $cursor = $position_next_gt + 1; + $inside_tag = false; + continue; + } else { + // inside tag, but there's no ending > sign + if ($e) $e->send(E_WARNING, 'Lexer: Missing gt'); + $token = new + HTMLPurifier_Token_Text( + '<' . + $this->parseData( + substr($html, $cursor) + ) + ); + if ($maintain_line_numbers) $token->rawPosition($current_line, $current_col); + // no cursor scroll? Hmm... + $array[] = $token; + break; + } + break; + } + + $context->destroy('CurrentLine'); + $context->destroy('CurrentCol'); + return $array; + } + + /** + * PHP 5.0.x compatible substr_count that implements offset and length + */ + protected function substrCount($haystack, $needle, $offset, $length) { + static $oldVersion; + if ($oldVersion === null) { + $oldVersion = version_compare(PHP_VERSION, '5.1', '<'); + } + if ($oldVersion) { + $haystack = substr($haystack, $offset, $length); + return substr_count($haystack, $needle); + } else { + return substr_count($haystack, $needle, $offset, $length); + } + } + + /** + * Takes the inside of an HTML tag and makes an assoc array of attributes. + * + * @param $string Inside of tag excluding name. + * @returns Assoc array of attributes. + */ + public function parseAttributeString($string, $config, $context) { + $string = (string) $string; // quick typecast + + if ($string == '') return array(); // no attributes + + $e = false; + if ($config->get('Core.CollectErrors')) { + $e =& $context->get('ErrorCollector'); + } + + // let's see if we can abort as quickly as possible + // one equal sign, no spaces => one attribute + $num_equal = substr_count($string, '='); + $has_space = strpos($string, ' '); + if ($num_equal === 0 && !$has_space) { + // bool attribute + return array($string => $string); + } elseif ($num_equal === 1 && !$has_space) { + // only one attribute + list($key, $quoted_value) = explode('=', $string); + $quoted_value = trim($quoted_value); + if (!$key) { + if ($e) $e->send(E_ERROR, 'Lexer: Missing attribute key'); + return array(); + } + if (!$quoted_value) return array($key => ''); + $first_char = @$quoted_value[0]; + $last_char = @$quoted_value[strlen($quoted_value)-1]; + + $same_quote = ($first_char == $last_char); + $open_quote = ($first_char == '"' || $first_char == "'"); + + if ( $same_quote && $open_quote) { + // well behaved + $value = substr($quoted_value, 1, strlen($quoted_value) - 2); + } else { + // not well behaved + if ($open_quote) { + if ($e) $e->send(E_ERROR, 'Lexer: Missing end quote'); + $value = substr($quoted_value, 1); + } else { + $value = $quoted_value; + } + } + if ($value === false) $value = ''; + return array($key => $this->parseData($value)); + } + + // setup loop environment + $array = array(); // return assoc array of attributes + $cursor = 0; // current position in string (moves forward) + $size = strlen($string); // size of the string (stays the same) + + // if we have unquoted attributes, the parser expects a terminating + // space, so let's guarantee that there's always a terminating space. + $string .= ' '; + + while(true) { + + if ($cursor >= $size) { + break; + } + + $cursor += ($value = strspn($string, $this->_whitespace, $cursor)); + // grab the key + + $key_begin = $cursor; //we're currently at the start of the key + + // scroll past all characters that are the key (not whitespace or =) + $cursor += strcspn($string, $this->_whitespace . '=', $cursor); + + $key_end = $cursor; // now at the end of the key + + $key = substr($string, $key_begin, $key_end - $key_begin); + + if (!$key) { + if ($e) $e->send(E_ERROR, 'Lexer: Missing attribute key'); + $cursor += strcspn($string, $this->_whitespace, $cursor + 1); // prevent infinite loop + continue; // empty key + } + + // scroll past all whitespace + $cursor += strspn($string, $this->_whitespace, $cursor); + + if ($cursor >= $size) { + $array[$key] = $key; + break; + } + + // if the next character is an equal sign, we've got a regular + // pair, otherwise, it's a bool attribute + $first_char = @$string[$cursor]; + + if ($first_char == '=') { + // key="value" + + $cursor++; + $cursor += strspn($string, $this->_whitespace, $cursor); + + if ($cursor === false) { + $array[$key] = ''; + break; + } + + // we might be in front of a quote right now + + $char = @$string[$cursor]; + + if ($char == '"' || $char == "'") { + // it's quoted, end bound is $char + $cursor++; + $value_begin = $cursor; + $cursor = strpos($string, $char, $cursor); + $value_end = $cursor; + } else { + // it's not quoted, end bound is whitespace + $value_begin = $cursor; + $cursor += strcspn($string, $this->_whitespace, $cursor); + $value_end = $cursor; + } + + // we reached a premature end + if ($cursor === false) { + $cursor = $size; + $value_end = $cursor; + } + + $value = substr($string, $value_begin, $value_end - $value_begin); + if ($value === false) $value = ''; + $array[$key] = $this->parseData($value); + $cursor++; + + } else { + // boolattr + if ($key !== '') { + $array[$key] = $key; + } else { + // purely theoretical + if ($e) $e->send(E_ERROR, 'Lexer: Missing attribute key'); + } + + } + } + return $array; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Lexer/PH5P.php b/lib/htmlpurifier/library/HTMLPurifier/Lexer/PH5P.php new file mode 100644 index 00000000..faf00b82 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Lexer/PH5P.php @@ -0,0 +1,3904 @@ +normalize($html, $config, $context); + $new_html = $this->wrapHTML($new_html, $config, $context); + try { + $parser = new HTML5($new_html); + $doc = $parser->save(); + } catch (DOMException $e) { + // Uh oh, it failed. Punt to DirectLex. + $lexer = new HTMLPurifier_Lexer_DirectLex(); + $context->register('PH5PError', $e); // save the error, so we can detect it + return $lexer->tokenizeHTML($html, $config, $context); // use original HTML + } + $tokens = array(); + $this->tokenizeDOM( + $doc->getElementsByTagName('html')->item(0)-> // + getElementsByTagName('body')->item(0)-> // + getElementsByTagName('div')->item(0) //
+ , $tokens); + return $tokens; + } + +} + +/* + +Copyright 2007 Jeroen van der Meer + +Permission is hereby granted, free of charge, to any person obtaining a +copy of this software and associated documentation files (the +"Software"), to deal in the Software without restriction, including +without limitation the rights to use, copy, modify, merge, publish, +distribute, sublicense, and/or sell copies of the Software, and to +permit persons to whom the Software is furnished to do so, subject to +the following conditions: + +The above copyright notice and this permission notice shall be included +in all copies or substantial portions of the Software. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS +OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF +MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. +IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY +CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, +TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE +SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. + +*/ + +class HTML5 { + private $data; + private $char; + private $EOF; + private $state; + private $tree; + private $token; + private $content_model; + private $escape = false; + private $entities = array('AElig;','AElig','AMP;','AMP','Aacute;','Aacute', + 'Acirc;','Acirc','Agrave;','Agrave','Alpha;','Aring;','Aring','Atilde;', + 'Atilde','Auml;','Auml','Beta;','COPY;','COPY','Ccedil;','Ccedil','Chi;', + 'Dagger;','Delta;','ETH;','ETH','Eacute;','Eacute','Ecirc;','Ecirc','Egrave;', + 'Egrave','Epsilon;','Eta;','Euml;','Euml','GT;','GT','Gamma;','Iacute;', + 'Iacute','Icirc;','Icirc','Igrave;','Igrave','Iota;','Iuml;','Iuml','Kappa;', + 'LT;','LT','Lambda;','Mu;','Ntilde;','Ntilde','Nu;','OElig;','Oacute;', + 'Oacute','Ocirc;','Ocirc','Ograve;','Ograve','Omega;','Omicron;','Oslash;', + 'Oslash','Otilde;','Otilde','Ouml;','Ouml','Phi;','Pi;','Prime;','Psi;', + 'QUOT;','QUOT','REG;','REG','Rho;','Scaron;','Sigma;','THORN;','THORN', + 'TRADE;','Tau;','Theta;','Uacute;','Uacute','Ucirc;','Ucirc','Ugrave;', + 'Ugrave','Upsilon;','Uuml;','Uuml','Xi;','Yacute;','Yacute','Yuml;','Zeta;', + 'aacute;','aacute','acirc;','acirc','acute;','acute','aelig;','aelig', + 'agrave;','agrave','alefsym;','alpha;','amp;','amp','and;','ang;','apos;', + 'aring;','aring','asymp;','atilde;','atilde','auml;','auml','bdquo;','beta;', + 'brvbar;','brvbar','bull;','cap;','ccedil;','ccedil','cedil;','cedil', + 'cent;','cent','chi;','circ;','clubs;','cong;','copy;','copy','crarr;', + 'cup;','curren;','curren','dArr;','dagger;','darr;','deg;','deg','delta;', + 'diams;','divide;','divide','eacute;','eacute','ecirc;','ecirc','egrave;', + 'egrave','empty;','emsp;','ensp;','epsilon;','equiv;','eta;','eth;','eth', + 'euml;','euml','euro;','exist;','fnof;','forall;','frac12;','frac12', + 'frac14;','frac14','frac34;','frac34','frasl;','gamma;','ge;','gt;','gt', + 'hArr;','harr;','hearts;','hellip;','iacute;','iacute','icirc;','icirc', + 'iexcl;','iexcl','igrave;','igrave','image;','infin;','int;','iota;', + 'iquest;','iquest','isin;','iuml;','iuml','kappa;','lArr;','lambda;','lang;', + 'laquo;','laquo','larr;','lceil;','ldquo;','le;','lfloor;','lowast;','loz;', + 'lrm;','lsaquo;','lsquo;','lt;','lt','macr;','macr','mdash;','micro;','micro', + 'middot;','middot','minus;','mu;','nabla;','nbsp;','nbsp','ndash;','ne;', + 'ni;','not;','not','notin;','nsub;','ntilde;','ntilde','nu;','oacute;', + 'oacute','ocirc;','ocirc','oelig;','ograve;','ograve','oline;','omega;', + 'omicron;','oplus;','or;','ordf;','ordf','ordm;','ordm','oslash;','oslash', + 'otilde;','otilde','otimes;','ouml;','ouml','para;','para','part;','permil;', + 'perp;','phi;','pi;','piv;','plusmn;','plusmn','pound;','pound','prime;', + 'prod;','prop;','psi;','quot;','quot','rArr;','radic;','rang;','raquo;', + 'raquo','rarr;','rceil;','rdquo;','real;','reg;','reg','rfloor;','rho;', + 'rlm;','rsaquo;','rsquo;','sbquo;','scaron;','sdot;','sect;','sect','shy;', + 'shy','sigma;','sigmaf;','sim;','spades;','sub;','sube;','sum;','sup1;', + 'sup1','sup2;','sup2','sup3;','sup3','sup;','supe;','szlig;','szlig','tau;', + 'there4;','theta;','thetasym;','thinsp;','thorn;','thorn','tilde;','times;', + 'times','trade;','uArr;','uacute;','uacute','uarr;','ucirc;','ucirc', + 'ugrave;','ugrave','uml;','uml','upsih;','upsilon;','uuml;','uuml','weierp;', + 'xi;','yacute;','yacute','yen;','yen','yuml;','yuml','zeta;','zwj;','zwnj;'); + + const PCDATA = 0; + const RCDATA = 1; + const CDATA = 2; + const PLAINTEXT = 3; + + const DOCTYPE = 0; + const STARTTAG = 1; + const ENDTAG = 2; + const COMMENT = 3; + const CHARACTR = 4; + const EOF = 5; + + public function __construct($data) { + + $this->data = $data; + $this->char = -1; + $this->EOF = strlen($data); + $this->tree = new HTML5TreeConstructer; + $this->content_model = self::PCDATA; + + $this->state = 'data'; + + while($this->state !== null) { + $this->{$this->state.'State'}(); + } + } + + public function save() { + return $this->tree->save(); + } + + private function char() { + return ($this->char < $this->EOF) + ? $this->data[$this->char] + : false; + } + + private function character($s, $l = 0) { + if($s + $l < $this->EOF) { + if($l === 0) { + return $this->data[$s]; + } else { + return substr($this->data, $s, $l); + } + } + } + + private function characters($char_class, $start) { + return preg_replace('#^(['.$char_class.']+).*#s', '\\1', substr($this->data, $start)); + } + + private function dataState() { + // Consume the next input character + $this->char++; + $char = $this->char(); + + if($char === '&' && ($this->content_model === self::PCDATA || $this->content_model === self::RCDATA)) { + /* U+0026 AMPERSAND (&) + When the content model flag is set to one of the PCDATA or RCDATA + states: switch to the entity data state. Otherwise: treat it as per + the "anything else" entry below. */ + $this->state = 'entityData'; + + } elseif($char === '-') { + /* If the content model flag is set to either the RCDATA state or + the CDATA state, and the escape flag is false, and there are at + least three characters before this one in the input stream, and the + last four characters in the input stream, including this one, are + U+003C LESS-THAN SIGN, U+0021 EXCLAMATION MARK, U+002D HYPHEN-MINUS, + and U+002D HYPHEN-MINUS (""), + set the escape flag to false. */ + if(($this->content_model === self::RCDATA || + $this->content_model === self::CDATA) && $this->escape === true && + $this->character($this->char, 3) === '-->') { + $this->escape = false; + } + + /* In any case, emit the input character as a character token. + Stay in the data state. */ + $this->emitToken(array( + 'type' => self::CHARACTR, + 'data' => $char + )); + + } elseif($this->char === $this->EOF) { + /* EOF + Emit an end-of-file token. */ + $this->EOF(); + + } elseif($this->content_model === self::PLAINTEXT) { + /* When the content model flag is set to the PLAINTEXT state + THIS DIFFERS GREATLY FROM THE SPEC: Get the remaining characters of + the text and emit it as a character token. */ + $this->emitToken(array( + 'type' => self::CHARACTR, + 'data' => substr($this->data, $this->char) + )); + + $this->EOF(); + + } else { + /* Anything else + THIS DIFFERS GREATLY FROM THE SPEC: Get as many character that + otherwise would also be treated as a character token and emit it + as a single character token. Stay in the data state. */ + $len = strcspn($this->data, '<&', $this->char); + $char = substr($this->data, $this->char, $len); + $this->char += $len - 1; + + $this->emitToken(array( + 'type' => self::CHARACTR, + 'data' => $char + )); + + $this->state = 'data'; + } + } + + private function entityDataState() { + // Attempt to consume an entity. + $entity = $this->entity(); + + // If nothing is returned, emit a U+0026 AMPERSAND character token. + // Otherwise, emit the character token that was returned. + $char = (!$entity) ? '&' : $entity; + $this->emitToken(array( + 'type' => self::CHARACTR, + 'data' => $char + )); + + // Finally, switch to the data state. + $this->state = 'data'; + } + + private function tagOpenState() { + switch($this->content_model) { + case self::RCDATA: + case self::CDATA: + /* If the next input character is a U+002F SOLIDUS (/) character, + consume it and switch to the close tag open state. If the next + input character is not a U+002F SOLIDUS (/) character, emit a + U+003C LESS-THAN SIGN character token and switch to the data + state to process the next input character. */ + if($this->character($this->char + 1) === '/') { + $this->char++; + $this->state = 'closeTagOpen'; + + } else { + $this->emitToken(array( + 'type' => self::CHARACTR, + 'data' => '<' + )); + + $this->state = 'data'; + } + break; + + case self::PCDATA: + // If the content model flag is set to the PCDATA state + // Consume the next input character: + $this->char++; + $char = $this->char(); + + if($char === '!') { + /* U+0021 EXCLAMATION MARK (!) + Switch to the markup declaration open state. */ + $this->state = 'markupDeclarationOpen'; + + } elseif($char === '/') { + /* U+002F SOLIDUS (/) + Switch to the close tag open state. */ + $this->state = 'closeTagOpen'; + + } elseif(preg_match('/^[A-Za-z]$/', $char)) { + /* U+0041 LATIN LETTER A through to U+005A LATIN LETTER Z + Create a new start tag token, set its tag name to the lowercase + version of the input character (add 0x0020 to the character's code + point), then switch to the tag name state. (Don't emit the token + yet; further details will be filled in before it is emitted.) */ + $this->token = array( + 'name' => strtolower($char), + 'type' => self::STARTTAG, + 'attr' => array() + ); + + $this->state = 'tagName'; + + } elseif($char === '>') { + /* U+003E GREATER-THAN SIGN (>) + Parse error. Emit a U+003C LESS-THAN SIGN character token and a + U+003E GREATER-THAN SIGN character token. Switch to the data state. */ + $this->emitToken(array( + 'type' => self::CHARACTR, + 'data' => '<>' + )); + + $this->state = 'data'; + + } elseif($char === '?') { + /* U+003F QUESTION MARK (?) + Parse error. Switch to the bogus comment state. */ + $this->state = 'bogusComment'; + + } else { + /* Anything else + Parse error. Emit a U+003C LESS-THAN SIGN character token and + reconsume the current input character in the data state. */ + $this->emitToken(array( + 'type' => self::CHARACTR, + 'data' => '<' + )); + + $this->char--; + $this->state = 'data'; + } + break; + } + } + + private function closeTagOpenState() { + $next_node = strtolower($this->characters('A-Za-z', $this->char + 1)); + $the_same = count($this->tree->stack) > 0 && $next_node === end($this->tree->stack)->nodeName; + + if(($this->content_model === self::RCDATA || $this->content_model === self::CDATA) && + (!$the_same || ($the_same && (!preg_match('/[\t\n\x0b\x0c >\/]/', + $this->character($this->char + 1 + strlen($next_node))) || $this->EOF === $this->char)))) { + /* If the content model flag is set to the RCDATA or CDATA states then + examine the next few characters. If they do not match the tag name of + the last start tag token emitted (case insensitively), or if they do but + they are not immediately followed by one of the following characters: + * U+0009 CHARACTER TABULATION + * U+000A LINE FEED (LF) + * U+000B LINE TABULATION + * U+000C FORM FEED (FF) + * U+0020 SPACE + * U+003E GREATER-THAN SIGN (>) + * U+002F SOLIDUS (/) + * EOF + ...then there is a parse error. Emit a U+003C LESS-THAN SIGN character + token, a U+002F SOLIDUS character token, and switch to the data state + to process the next input character. */ + $this->emitToken(array( + 'type' => self::CHARACTR, + 'data' => 'state = 'data'; + + } else { + /* Otherwise, if the content model flag is set to the PCDATA state, + or if the next few characters do match that tag name, consume the + next input character: */ + $this->char++; + $char = $this->char(); + + if(preg_match('/^[A-Za-z]$/', $char)) { + /* U+0041 LATIN LETTER A through to U+005A LATIN LETTER Z + Create a new end tag token, set its tag name to the lowercase version + of the input character (add 0x0020 to the character's code point), then + switch to the tag name state. (Don't emit the token yet; further details + will be filled in before it is emitted.) */ + $this->token = array( + 'name' => strtolower($char), + 'type' => self::ENDTAG + ); + + $this->state = 'tagName'; + + } elseif($char === '>') { + /* U+003E GREATER-THAN SIGN (>) + Parse error. Switch to the data state. */ + $this->state = 'data'; + + } elseif($this->char === $this->EOF) { + /* EOF + Parse error. Emit a U+003C LESS-THAN SIGN character token and a U+002F + SOLIDUS character token. Reconsume the EOF character in the data state. */ + $this->emitToken(array( + 'type' => self::CHARACTR, + 'data' => 'char--; + $this->state = 'data'; + + } else { + /* Parse error. Switch to the bogus comment state. */ + $this->state = 'bogusComment'; + } + } + } + + private function tagNameState() { + // Consume the next input character: + $this->char++; + $char = $this->character($this->char); + + if(preg_match('/^[\t\n\x0b\x0c ]$/', $char)) { + /* U+0009 CHARACTER TABULATION + U+000A LINE FEED (LF) + U+000B LINE TABULATION + U+000C FORM FEED (FF) + U+0020 SPACE + Switch to the before attribute name state. */ + $this->state = 'beforeAttributeName'; + + } elseif($char === '>') { + /* U+003E GREATER-THAN SIGN (>) + Emit the current tag token. Switch to the data state. */ + $this->emitToken($this->token); + $this->state = 'data'; + + } elseif($this->char === $this->EOF) { + /* EOF + Parse error. Emit the current tag token. Reconsume the EOF + character in the data state. */ + $this->emitToken($this->token); + + $this->char--; + $this->state = 'data'; + + } elseif($char === '/') { + /* U+002F SOLIDUS (/) + Parse error unless this is a permitted slash. Switch to the before + attribute name state. */ + $this->state = 'beforeAttributeName'; + + } else { + /* Anything else + Append the current input character to the current tag token's tag name. + Stay in the tag name state. */ + $this->token['name'] .= strtolower($char); + $this->state = 'tagName'; + } + } + + private function beforeAttributeNameState() { + // Consume the next input character: + $this->char++; + $char = $this->character($this->char); + + if(preg_match('/^[\t\n\x0b\x0c ]$/', $char)) { + /* U+0009 CHARACTER TABULATION + U+000A LINE FEED (LF) + U+000B LINE TABULATION + U+000C FORM FEED (FF) + U+0020 SPACE + Stay in the before attribute name state. */ + $this->state = 'beforeAttributeName'; + + } elseif($char === '>') { + /* U+003E GREATER-THAN SIGN (>) + Emit the current tag token. Switch to the data state. */ + $this->emitToken($this->token); + $this->state = 'data'; + + } elseif($char === '/') { + /* U+002F SOLIDUS (/) + Parse error unless this is a permitted slash. Stay in the before + attribute name state. */ + $this->state = 'beforeAttributeName'; + + } elseif($this->char === $this->EOF) { + /* EOF + Parse error. Emit the current tag token. Reconsume the EOF + character in the data state. */ + $this->emitToken($this->token); + + $this->char--; + $this->state = 'data'; + + } else { + /* Anything else + Start a new attribute in the current tag token. Set that attribute's + name to the current input character, and its value to the empty string. + Switch to the attribute name state. */ + $this->token['attr'][] = array( + 'name' => strtolower($char), + 'value' => null + ); + + $this->state = 'attributeName'; + } + } + + private function attributeNameState() { + // Consume the next input character: + $this->char++; + $char = $this->character($this->char); + + if(preg_match('/^[\t\n\x0b\x0c ]$/', $char)) { + /* U+0009 CHARACTER TABULATION + U+000A LINE FEED (LF) + U+000B LINE TABULATION + U+000C FORM FEED (FF) + U+0020 SPACE + Stay in the before attribute name state. */ + $this->state = 'afterAttributeName'; + + } elseif($char === '=') { + /* U+003D EQUALS SIGN (=) + Switch to the before attribute value state. */ + $this->state = 'beforeAttributeValue'; + + } elseif($char === '>') { + /* U+003E GREATER-THAN SIGN (>) + Emit the current tag token. Switch to the data state. */ + $this->emitToken($this->token); + $this->state = 'data'; + + } elseif($char === '/' && $this->character($this->char + 1) !== '>') { + /* U+002F SOLIDUS (/) + Parse error unless this is a permitted slash. Switch to the before + attribute name state. */ + $this->state = 'beforeAttributeName'; + + } elseif($this->char === $this->EOF) { + /* EOF + Parse error. Emit the current tag token. Reconsume the EOF + character in the data state. */ + $this->emitToken($this->token); + + $this->char--; + $this->state = 'data'; + + } else { + /* Anything else + Append the current input character to the current attribute's name. + Stay in the attribute name state. */ + $last = count($this->token['attr']) - 1; + $this->token['attr'][$last]['name'] .= strtolower($char); + + $this->state = 'attributeName'; + } + } + + private function afterAttributeNameState() { + // Consume the next input character: + $this->char++; + $char = $this->character($this->char); + + if(preg_match('/^[\t\n\x0b\x0c ]$/', $char)) { + /* U+0009 CHARACTER TABULATION + U+000A LINE FEED (LF) + U+000B LINE TABULATION + U+000C FORM FEED (FF) + U+0020 SPACE + Stay in the after attribute name state. */ + $this->state = 'afterAttributeName'; + + } elseif($char === '=') { + /* U+003D EQUALS SIGN (=) + Switch to the before attribute value state. */ + $this->state = 'beforeAttributeValue'; + + } elseif($char === '>') { + /* U+003E GREATER-THAN SIGN (>) + Emit the current tag token. Switch to the data state. */ + $this->emitToken($this->token); + $this->state = 'data'; + + } elseif($char === '/' && $this->character($this->char + 1) !== '>') { + /* U+002F SOLIDUS (/) + Parse error unless this is a permitted slash. Switch to the + before attribute name state. */ + $this->state = 'beforeAttributeName'; + + } elseif($this->char === $this->EOF) { + /* EOF + Parse error. Emit the current tag token. Reconsume the EOF + character in the data state. */ + $this->emitToken($this->token); + + $this->char--; + $this->state = 'data'; + + } else { + /* Anything else + Start a new attribute in the current tag token. Set that attribute's + name to the current input character, and its value to the empty string. + Switch to the attribute name state. */ + $this->token['attr'][] = array( + 'name' => strtolower($char), + 'value' => null + ); + + $this->state = 'attributeName'; + } + } + + private function beforeAttributeValueState() { + // Consume the next input character: + $this->char++; + $char = $this->character($this->char); + + if(preg_match('/^[\t\n\x0b\x0c ]$/', $char)) { + /* U+0009 CHARACTER TABULATION + U+000A LINE FEED (LF) + U+000B LINE TABULATION + U+000C FORM FEED (FF) + U+0020 SPACE + Stay in the before attribute value state. */ + $this->state = 'beforeAttributeValue'; + + } elseif($char === '"') { + /* U+0022 QUOTATION MARK (") + Switch to the attribute value (double-quoted) state. */ + $this->state = 'attributeValueDoubleQuoted'; + + } elseif($char === '&') { + /* U+0026 AMPERSAND (&) + Switch to the attribute value (unquoted) state and reconsume + this input character. */ + $this->char--; + $this->state = 'attributeValueUnquoted'; + + } elseif($char === '\'') { + /* U+0027 APOSTROPHE (') + Switch to the attribute value (single-quoted) state. */ + $this->state = 'attributeValueSingleQuoted'; + + } elseif($char === '>') { + /* U+003E GREATER-THAN SIGN (>) + Emit the current tag token. Switch to the data state. */ + $this->emitToken($this->token); + $this->state = 'data'; + + } else { + /* Anything else + Append the current input character to the current attribute's value. + Switch to the attribute value (unquoted) state. */ + $last = count($this->token['attr']) - 1; + $this->token['attr'][$last]['value'] .= $char; + + $this->state = 'attributeValueUnquoted'; + } + } + + private function attributeValueDoubleQuotedState() { + // Consume the next input character: + $this->char++; + $char = $this->character($this->char); + + if($char === '"') { + /* U+0022 QUOTATION MARK (") + Switch to the before attribute name state. */ + $this->state = 'beforeAttributeName'; + + } elseif($char === '&') { + /* U+0026 AMPERSAND (&) + Switch to the entity in attribute value state. */ + $this->entityInAttributeValueState('double'); + + } elseif($this->char === $this->EOF) { + /* EOF + Parse error. Emit the current tag token. Reconsume the character + in the data state. */ + $this->emitToken($this->token); + + $this->char--; + $this->state = 'data'; + + } else { + /* Anything else + Append the current input character to the current attribute's value. + Stay in the attribute value (double-quoted) state. */ + $last = count($this->token['attr']) - 1; + $this->token['attr'][$last]['value'] .= $char; + + $this->state = 'attributeValueDoubleQuoted'; + } + } + + private function attributeValueSingleQuotedState() { + // Consume the next input character: + $this->char++; + $char = $this->character($this->char); + + if($char === '\'') { + /* U+0022 QUOTATION MARK (') + Switch to the before attribute name state. */ + $this->state = 'beforeAttributeName'; + + } elseif($char === '&') { + /* U+0026 AMPERSAND (&) + Switch to the entity in attribute value state. */ + $this->entityInAttributeValueState('single'); + + } elseif($this->char === $this->EOF) { + /* EOF + Parse error. Emit the current tag token. Reconsume the character + in the data state. */ + $this->emitToken($this->token); + + $this->char--; + $this->state = 'data'; + + } else { + /* Anything else + Append the current input character to the current attribute's value. + Stay in the attribute value (single-quoted) state. */ + $last = count($this->token['attr']) - 1; + $this->token['attr'][$last]['value'] .= $char; + + $this->state = 'attributeValueSingleQuoted'; + } + } + + private function attributeValueUnquotedState() { + // Consume the next input character: + $this->char++; + $char = $this->character($this->char); + + if(preg_match('/^[\t\n\x0b\x0c ]$/', $char)) { + /* U+0009 CHARACTER TABULATION + U+000A LINE FEED (LF) + U+000B LINE TABULATION + U+000C FORM FEED (FF) + U+0020 SPACE + Switch to the before attribute name state. */ + $this->state = 'beforeAttributeName'; + + } elseif($char === '&') { + /* U+0026 AMPERSAND (&) + Switch to the entity in attribute value state. */ + $this->entityInAttributeValueState(); + + } elseif($char === '>') { + /* U+003E GREATER-THAN SIGN (>) + Emit the current tag token. Switch to the data state. */ + $this->emitToken($this->token); + $this->state = 'data'; + + } else { + /* Anything else + Append the current input character to the current attribute's value. + Stay in the attribute value (unquoted) state. */ + $last = count($this->token['attr']) - 1; + $this->token['attr'][$last]['value'] .= $char; + + $this->state = 'attributeValueUnquoted'; + } + } + + private function entityInAttributeValueState() { + // Attempt to consume an entity. + $entity = $this->entity(); + + // If nothing is returned, append a U+0026 AMPERSAND character to the + // current attribute's value. Otherwise, emit the character token that + // was returned. + $char = (!$entity) + ? '&' + : $entity; + + $last = count($this->token['attr']) - 1; + $this->token['attr'][$last]['value'] .= $char; + } + + private function bogusCommentState() { + /* Consume every character up to the first U+003E GREATER-THAN SIGN + character (>) or the end of the file (EOF), whichever comes first. Emit + a comment token whose data is the concatenation of all the characters + starting from and including the character that caused the state machine + to switch into the bogus comment state, up to and including the last + consumed character before the U+003E character, if any, or up to the + end of the file otherwise. (If the comment was started by the end of + the file (EOF), the token is empty.) */ + $data = $this->characters('^>', $this->char); + $this->emitToken(array( + 'data' => $data, + 'type' => self::COMMENT + )); + + $this->char += strlen($data); + + /* Switch to the data state. */ + $this->state = 'data'; + + /* If the end of the file was reached, reconsume the EOF character. */ + if($this->char === $this->EOF) { + $this->char = $this->EOF - 1; + } + } + + private function markupDeclarationOpenState() { + /* If the next two characters are both U+002D HYPHEN-MINUS (-) + characters, consume those two characters, create a comment token whose + data is the empty string, and switch to the comment state. */ + if($this->character($this->char + 1, 2) === '--') { + $this->char += 2; + $this->state = 'comment'; + $this->token = array( + 'data' => null, + 'type' => self::COMMENT + ); + + /* Otherwise if the next seven chacacters are a case-insensitive match + for the word "DOCTYPE", then consume those characters and switch to the + DOCTYPE state. */ + } elseif(strtolower($this->character($this->char + 1, 7)) === 'doctype') { + $this->char += 7; + $this->state = 'doctype'; + + /* Otherwise, is is a parse error. Switch to the bogus comment state. + The next character that is consumed, if any, is the first character + that will be in the comment. */ + } else { + $this->char++; + $this->state = 'bogusComment'; + } + } + + private function commentState() { + /* Consume the next input character: */ + $this->char++; + $char = $this->char(); + + /* U+002D HYPHEN-MINUS (-) */ + if($char === '-') { + /* Switch to the comment dash state */ + $this->state = 'commentDash'; + + /* EOF */ + } elseif($this->char === $this->EOF) { + /* Parse error. Emit the comment token. Reconsume the EOF character + in the data state. */ + $this->emitToken($this->token); + $this->char--; + $this->state = 'data'; + + /* Anything else */ + } else { + /* Append the input character to the comment token's data. Stay in + the comment state. */ + $this->token['data'] .= $char; + } + } + + private function commentDashState() { + /* Consume the next input character: */ + $this->char++; + $char = $this->char(); + + /* U+002D HYPHEN-MINUS (-) */ + if($char === '-') { + /* Switch to the comment end state */ + $this->state = 'commentEnd'; + + /* EOF */ + } elseif($this->char === $this->EOF) { + /* Parse error. Emit the comment token. Reconsume the EOF character + in the data state. */ + $this->emitToken($this->token); + $this->char--; + $this->state = 'data'; + + /* Anything else */ + } else { + /* Append a U+002D HYPHEN-MINUS (-) character and the input + character to the comment token's data. Switch to the comment state. */ + $this->token['data'] .= '-'.$char; + $this->state = 'comment'; + } + } + + private function commentEndState() { + /* Consume the next input character: */ + $this->char++; + $char = $this->char(); + + if($char === '>') { + $this->emitToken($this->token); + $this->state = 'data'; + + } elseif($char === '-') { + $this->token['data'] .= '-'; + + } elseif($this->char === $this->EOF) { + $this->emitToken($this->token); + $this->char--; + $this->state = 'data'; + + } else { + $this->token['data'] .= '--'.$char; + $this->state = 'comment'; + } + } + + private function doctypeState() { + /* Consume the next input character: */ + $this->char++; + $char = $this->char(); + + if(preg_match('/^[\t\n\x0b\x0c ]$/', $char)) { + $this->state = 'beforeDoctypeName'; + + } else { + $this->char--; + $this->state = 'beforeDoctypeName'; + } + } + + private function beforeDoctypeNameState() { + /* Consume the next input character: */ + $this->char++; + $char = $this->char(); + + if(preg_match('/^[\t\n\x0b\x0c ]$/', $char)) { + // Stay in the before DOCTYPE name state. + + } elseif(preg_match('/^[a-z]$/', $char)) { + $this->token = array( + 'name' => strtoupper($char), + 'type' => self::DOCTYPE, + 'error' => true + ); + + $this->state = 'doctypeName'; + + } elseif($char === '>') { + $this->emitToken(array( + 'name' => null, + 'type' => self::DOCTYPE, + 'error' => true + )); + + $this->state = 'data'; + + } elseif($this->char === $this->EOF) { + $this->emitToken(array( + 'name' => null, + 'type' => self::DOCTYPE, + 'error' => true + )); + + $this->char--; + $this->state = 'data'; + + } else { + $this->token = array( + 'name' => $char, + 'type' => self::DOCTYPE, + 'error' => true + ); + + $this->state = 'doctypeName'; + } + } + + private function doctypeNameState() { + /* Consume the next input character: */ + $this->char++; + $char = $this->char(); + + if(preg_match('/^[\t\n\x0b\x0c ]$/', $char)) { + $this->state = 'AfterDoctypeName'; + + } elseif($char === '>') { + $this->emitToken($this->token); + $this->state = 'data'; + + } elseif(preg_match('/^[a-z]$/', $char)) { + $this->token['name'] .= strtoupper($char); + + } elseif($this->char === $this->EOF) { + $this->emitToken($this->token); + $this->char--; + $this->state = 'data'; + + } else { + $this->token['name'] .= $char; + } + + $this->token['error'] = ($this->token['name'] === 'HTML') + ? false + : true; + } + + private function afterDoctypeNameState() { + /* Consume the next input character: */ + $this->char++; + $char = $this->char(); + + if(preg_match('/^[\t\n\x0b\x0c ]$/', $char)) { + // Stay in the DOCTYPE name state. + + } elseif($char === '>') { + $this->emitToken($this->token); + $this->state = 'data'; + + } elseif($this->char === $this->EOF) { + $this->emitToken($this->token); + $this->char--; + $this->state = 'data'; + + } else { + $this->token['error'] = true; + $this->state = 'bogusDoctype'; + } + } + + private function bogusDoctypeState() { + /* Consume the next input character: */ + $this->char++; + $char = $this->char(); + + if($char === '>') { + $this->emitToken($this->token); + $this->state = 'data'; + + } elseif($this->char === $this->EOF) { + $this->emitToken($this->token); + $this->char--; + $this->state = 'data'; + + } else { + // Stay in the bogus DOCTYPE state. + } + } + + private function entity() { + $start = $this->char; + + // This section defines how to consume an entity. This definition is + // used when parsing entities in text and in attributes. + + // The behaviour depends on the identity of the next character (the + // one immediately after the U+0026 AMPERSAND character): + + switch($this->character($this->char + 1)) { + // U+0023 NUMBER SIGN (#) + case '#': + + // The behaviour further depends on the character after the + // U+0023 NUMBER SIGN: + switch($this->character($this->char + 1)) { + // U+0078 LATIN SMALL LETTER X + // U+0058 LATIN CAPITAL LETTER X + case 'x': + case 'X': + // Follow the steps below, but using the range of + // characters U+0030 DIGIT ZERO through to U+0039 DIGIT + // NINE, U+0061 LATIN SMALL LETTER A through to U+0066 + // LATIN SMALL LETTER F, and U+0041 LATIN CAPITAL LETTER + // A, through to U+0046 LATIN CAPITAL LETTER F (in other + // words, 0-9, A-F, a-f). + $char = 1; + $char_class = '0-9A-Fa-f'; + break; + + // Anything else + default: + // Follow the steps below, but using the range of + // characters U+0030 DIGIT ZERO through to U+0039 DIGIT + // NINE (i.e. just 0-9). + $char = 0; + $char_class = '0-9'; + break; + } + + // Consume as many characters as match the range of characters + // given above. + $this->char++; + $e_name = $this->characters($char_class, $this->char + $char + 1); + $entity = $this->character($start, $this->char); + $cond = strlen($e_name) > 0; + + // The rest of the parsing happens bellow. + break; + + // Anything else + default: + // Consume the maximum number of characters possible, with the + // consumed characters case-sensitively matching one of the + // identifiers in the first column of the entities table. + $e_name = $this->characters('0-9A-Za-z;', $this->char + 1); + $len = strlen($e_name); + + for($c = 1; $c <= $len; $c++) { + $id = substr($e_name, 0, $c); + $this->char++; + + if(in_array($id, $this->entities)) { + if ($e_name[$c-1] !== ';') { + if ($c < $len && $e_name[$c] == ';') { + $this->char++; // consume extra semicolon + } + } + $entity = $id; + break; + } + } + + $cond = isset($entity); + // The rest of the parsing happens bellow. + break; + } + + if(!$cond) { + // If no match can be made, then this is a parse error. No + // characters are consumed, and nothing is returned. + $this->char = $start; + return false; + } + + // Return a character token for the character corresponding to the + // entity name (as given by the second column of the entities table). + return html_entity_decode('&'.$entity.';', ENT_QUOTES, 'UTF-8'); + } + + private function emitToken($token) { + $emit = $this->tree->emitToken($token); + + if(is_int($emit)) { + $this->content_model = $emit; + + } elseif($token['type'] === self::ENDTAG) { + $this->content_model = self::PCDATA; + } + } + + private function EOF() { + $this->state = null; + $this->tree->emitToken(array( + 'type' => self::EOF + )); + } +} + +class HTML5TreeConstructer { + public $stack = array(); + + private $phase; + private $mode; + private $dom; + private $foster_parent = null; + private $a_formatting = array(); + + private $head_pointer = null; + private $form_pointer = null; + + private $scoping = array('button','caption','html','marquee','object','table','td','th'); + private $formatting = array('a','b','big','em','font','i','nobr','s','small','strike','strong','tt','u'); + private $special = array('address','area','base','basefont','bgsound', + 'blockquote','body','br','center','col','colgroup','dd','dir','div','dl', + 'dt','embed','fieldset','form','frame','frameset','h1','h2','h3','h4','h5', + 'h6','head','hr','iframe','image','img','input','isindex','li','link', + 'listing','menu','meta','noembed','noframes','noscript','ol','optgroup', + 'option','p','param','plaintext','pre','script','select','spacer','style', + 'tbody','textarea','tfoot','thead','title','tr','ul','wbr'); + + // The different phases. + const INIT_PHASE = 0; + const ROOT_PHASE = 1; + const MAIN_PHASE = 2; + const END_PHASE = 3; + + // The different insertion modes for the main phase. + const BEFOR_HEAD = 0; + const IN_HEAD = 1; + const AFTER_HEAD = 2; + const IN_BODY = 3; + const IN_TABLE = 4; + const IN_CAPTION = 5; + const IN_CGROUP = 6; + const IN_TBODY = 7; + const IN_ROW = 8; + const IN_CELL = 9; + const IN_SELECT = 10; + const AFTER_BODY = 11; + const IN_FRAME = 12; + const AFTR_FRAME = 13; + + // The different types of elements. + const SPECIAL = 0; + const SCOPING = 1; + const FORMATTING = 2; + const PHRASING = 3; + + const MARKER = 0; + + public function __construct() { + $this->phase = self::INIT_PHASE; + $this->mode = self::BEFOR_HEAD; + $this->dom = new DOMDocument; + + $this->dom->encoding = 'UTF-8'; + $this->dom->preserveWhiteSpace = true; + $this->dom->substituteEntities = true; + $this->dom->strictErrorChecking = false; + } + + // Process tag tokens + public function emitToken($token) { + switch($this->phase) { + case self::INIT_PHASE: return $this->initPhase($token); break; + case self::ROOT_PHASE: return $this->rootElementPhase($token); break; + case self::MAIN_PHASE: return $this->mainPhase($token); break; + case self::END_PHASE : return $this->trailingEndPhase($token); break; + } + } + + private function initPhase($token) { + /* Initially, the tree construction stage must handle each token + emitted from the tokenisation stage as follows: */ + + /* A DOCTYPE token that is marked as being in error + A comment token + A start tag token + An end tag token + A character token that is not one of one of U+0009 CHARACTER TABULATION, + U+000A LINE FEED (LF), U+000B LINE TABULATION, U+000C FORM FEED (FF), + or U+0020 SPACE + An end-of-file token */ + if((isset($token['error']) && $token['error']) || + $token['type'] === HTML5::COMMENT || + $token['type'] === HTML5::STARTTAG || + $token['type'] === HTML5::ENDTAG || + $token['type'] === HTML5::EOF || + ($token['type'] === HTML5::CHARACTR && isset($token['data']) && + !preg_match('/^[\t\n\x0b\x0c ]+$/', $token['data']))) { + /* This specification does not define how to handle this case. In + particular, user agents may ignore the entirety of this specification + altogether for such documents, and instead invoke special parse modes + with a greater emphasis on backwards compatibility. */ + + $this->phase = self::ROOT_PHASE; + return $this->rootElementPhase($token); + + /* A DOCTYPE token marked as being correct */ + } elseif(isset($token['error']) && !$token['error']) { + /* Append a DocumentType node to the Document node, with the name + attribute set to the name given in the DOCTYPE token (which will be + "HTML"), and the other attributes specific to DocumentType objects + set to null, empty lists, or the empty string as appropriate. */ + $doctype = new DOMDocumentType(null, null, 'HTML'); + + /* Then, switch to the root element phase of the tree construction + stage. */ + $this->phase = self::ROOT_PHASE; + + /* A character token that is one of one of U+0009 CHARACTER TABULATION, + U+000A LINE FEED (LF), U+000B LINE TABULATION, U+000C FORM FEED (FF), + or U+0020 SPACE */ + } elseif(isset($token['data']) && preg_match('/^[\t\n\x0b\x0c ]+$/', + $token['data'])) { + /* Append that character to the Document node. */ + $text = $this->dom->createTextNode($token['data']); + $this->dom->appendChild($text); + } + } + + private function rootElementPhase($token) { + /* After the initial phase, as each token is emitted from the tokenisation + stage, it must be processed as described in this section. */ + + /* A DOCTYPE token */ + if($token['type'] === HTML5::DOCTYPE) { + // Parse error. Ignore the token. + + /* A comment token */ + } elseif($token['type'] === HTML5::COMMENT) { + /* Append a Comment node to the Document object with the data + attribute set to the data given in the comment token. */ + $comment = $this->dom->createComment($token['data']); + $this->dom->appendChild($comment); + + /* A character token that is one of one of U+0009 CHARACTER TABULATION, + U+000A LINE FEED (LF), U+000B LINE TABULATION, U+000C FORM FEED (FF), + or U+0020 SPACE */ + } elseif($token['type'] === HTML5::CHARACTR && + preg_match('/^[\t\n\x0b\x0c ]+$/', $token['data'])) { + /* Append that character to the Document node. */ + $text = $this->dom->createTextNode($token['data']); + $this->dom->appendChild($text); + + /* A character token that is not one of U+0009 CHARACTER TABULATION, + U+000A LINE FEED (LF), U+000B LINE TABULATION, U+000C FORM FEED + (FF), or U+0020 SPACE + A start tag token + An end tag token + An end-of-file token */ + } elseif(($token['type'] === HTML5::CHARACTR && + !preg_match('/^[\t\n\x0b\x0c ]+$/', $token['data'])) || + $token['type'] === HTML5::STARTTAG || + $token['type'] === HTML5::ENDTAG || + $token['type'] === HTML5::EOF) { + /* Create an HTMLElement node with the tag name html, in the HTML + namespace. Append it to the Document object. Switch to the main + phase and reprocess the current token. */ + $html = $this->dom->createElement('html'); + $this->dom->appendChild($html); + $this->stack[] = $html; + + $this->phase = self::MAIN_PHASE; + return $this->mainPhase($token); + } + } + + private function mainPhase($token) { + /* Tokens in the main phase must be handled as follows: */ + + /* A DOCTYPE token */ + if($token['type'] === HTML5::DOCTYPE) { + // Parse error. Ignore the token. + + /* A start tag token with the tag name "html" */ + } elseif($token['type'] === HTML5::STARTTAG && $token['name'] === 'html') { + /* If this start tag token was not the first start tag token, then + it is a parse error. */ + + /* For each attribute on the token, check to see if the attribute + is already present on the top element of the stack of open elements. + If it is not, add the attribute and its corresponding value to that + element. */ + foreach($token['attr'] as $attr) { + if(!$this->stack[0]->hasAttribute($attr['name'])) { + $this->stack[0]->setAttribute($attr['name'], $attr['value']); + } + } + + /* An end-of-file token */ + } elseif($token['type'] === HTML5::EOF) { + /* Generate implied end tags. */ + $this->generateImpliedEndTags(); + + /* Anything else. */ + } else { + /* Depends on the insertion mode: */ + switch($this->mode) { + case self::BEFOR_HEAD: return $this->beforeHead($token); break; + case self::IN_HEAD: return $this->inHead($token); break; + case self::AFTER_HEAD: return $this->afterHead($token); break; + case self::IN_BODY: return $this->inBody($token); break; + case self::IN_TABLE: return $this->inTable($token); break; + case self::IN_CAPTION: return $this->inCaption($token); break; + case self::IN_CGROUP: return $this->inColumnGroup($token); break; + case self::IN_TBODY: return $this->inTableBody($token); break; + case self::IN_ROW: return $this->inRow($token); break; + case self::IN_CELL: return $this->inCell($token); break; + case self::IN_SELECT: return $this->inSelect($token); break; + case self::AFTER_BODY: return $this->afterBody($token); break; + case self::IN_FRAME: return $this->inFrameset($token); break; + case self::AFTR_FRAME: return $this->afterFrameset($token); break; + case self::END_PHASE: return $this->trailingEndPhase($token); break; + } + } + } + + private function beforeHead($token) { + /* Handle the token as follows: */ + + /* A character token that is one of one of U+0009 CHARACTER TABULATION, + U+000A LINE FEED (LF), U+000B LINE TABULATION, U+000C FORM FEED (FF), + or U+0020 SPACE */ + if($token['type'] === HTML5::CHARACTR && + preg_match('/^[\t\n\x0b\x0c ]+$/', $token['data'])) { + /* Append the character to the current node. */ + $this->insertText($token['data']); + + /* A comment token */ + } elseif($token['type'] === HTML5::COMMENT) { + /* Append a Comment node to the current node with the data attribute + set to the data given in the comment token. */ + $this->insertComment($token['data']); + + /* A start tag token with the tag name "head" */ + } elseif($token['type'] === HTML5::STARTTAG && $token['name'] === 'head') { + /* Create an element for the token, append the new element to the + current node and push it onto the stack of open elements. */ + $element = $this->insertElement($token); + + /* Set the head element pointer to this new element node. */ + $this->head_pointer = $element; + + /* Change the insertion mode to "in head". */ + $this->mode = self::IN_HEAD; + + /* A start tag token whose tag name is one of: "base", "link", "meta", + "script", "style", "title". Or an end tag with the tag name "html". + Or a character token that is not one of U+0009 CHARACTER TABULATION, + U+000A LINE FEED (LF), U+000B LINE TABULATION, U+000C FORM FEED (FF), + or U+0020 SPACE. Or any other start tag token */ + } elseif($token['type'] === HTML5::STARTTAG || + ($token['type'] === HTML5::ENDTAG && $token['name'] === 'html') || + ($token['type'] === HTML5::CHARACTR && !preg_match('/^[\t\n\x0b\x0c ]$/', + $token['data']))) { + /* Act as if a start tag token with the tag name "head" and no + attributes had been seen, then reprocess the current token. */ + $this->beforeHead(array( + 'name' => 'head', + 'type' => HTML5::STARTTAG, + 'attr' => array() + )); + + return $this->inHead($token); + + /* Any other end tag */ + } elseif($token['type'] === HTML5::ENDTAG) { + /* Parse error. Ignore the token. */ + } + } + + private function inHead($token) { + /* Handle the token as follows: */ + + /* A character token that is one of one of U+0009 CHARACTER TABULATION, + U+000A LINE FEED (LF), U+000B LINE TABULATION, U+000C FORM FEED (FF), + or U+0020 SPACE. + + THIS DIFFERS FROM THE SPEC: If the current node is either a title, style + or script element, append the character to the current node regardless + of its content. */ + if(($token['type'] === HTML5::CHARACTR && + preg_match('/^[\t\n\x0b\x0c ]+$/', $token['data'])) || ( + $token['type'] === HTML5::CHARACTR && in_array(end($this->stack)->nodeName, + array('title', 'style', 'script')))) { + /* Append the character to the current node. */ + $this->insertText($token['data']); + + /* A comment token */ + } elseif($token['type'] === HTML5::COMMENT) { + /* Append a Comment node to the current node with the data attribute + set to the data given in the comment token. */ + $this->insertComment($token['data']); + + } elseif($token['type'] === HTML5::ENDTAG && + in_array($token['name'], array('title', 'style', 'script'))) { + array_pop($this->stack); + return HTML5::PCDATA; + + /* A start tag with the tag name "title" */ + } elseif($token['type'] === HTML5::STARTTAG && $token['name'] === 'title') { + /* Create an element for the token and append the new element to the + node pointed to by the head element pointer, or, if that is null + (innerHTML case), to the current node. */ + if($this->head_pointer !== null) { + $element = $this->insertElement($token, false); + $this->head_pointer->appendChild($element); + + } else { + $element = $this->insertElement($token); + } + + /* Switch the tokeniser's content model flag to the RCDATA state. */ + return HTML5::RCDATA; + + /* A start tag with the tag name "style" */ + } elseif($token['type'] === HTML5::STARTTAG && $token['name'] === 'style') { + /* Create an element for the token and append the new element to the + node pointed to by the head element pointer, or, if that is null + (innerHTML case), to the current node. */ + if($this->head_pointer !== null) { + $element = $this->insertElement($token, false); + $this->head_pointer->appendChild($element); + + } else { + $this->insertElement($token); + } + + /* Switch the tokeniser's content model flag to the CDATA state. */ + return HTML5::CDATA; + + /* A start tag with the tag name "script" */ + } elseif($token['type'] === HTML5::STARTTAG && $token['name'] === 'script') { + /* Create an element for the token. */ + $element = $this->insertElement($token, false); + $this->head_pointer->appendChild($element); + + /* Switch the tokeniser's content model flag to the CDATA state. */ + return HTML5::CDATA; + + /* A start tag with the tag name "base", "link", or "meta" */ + } elseif($token['type'] === HTML5::STARTTAG && in_array($token['name'], + array('base', 'link', 'meta'))) { + /* Create an element for the token and append the new element to the + node pointed to by the head element pointer, or, if that is null + (innerHTML case), to the current node. */ + if($this->head_pointer !== null) { + $element = $this->insertElement($token, false); + $this->head_pointer->appendChild($element); + array_pop($this->stack); + + } else { + $this->insertElement($token); + } + + /* An end tag with the tag name "head" */ + } elseif($token['type'] === HTML5::ENDTAG && $token['name'] === 'head') { + /* If the current node is a head element, pop the current node off + the stack of open elements. */ + if($this->head_pointer->isSameNode(end($this->stack))) { + array_pop($this->stack); + + /* Otherwise, this is a parse error. */ + } else { + // k + } + + /* Change the insertion mode to "after head". */ + $this->mode = self::AFTER_HEAD; + + /* A start tag with the tag name "head" or an end tag except "html". */ + } elseif(($token['type'] === HTML5::STARTTAG && $token['name'] === 'head') || + ($token['type'] === HTML5::ENDTAG && $token['name'] !== 'html')) { + // Parse error. Ignore the token. + + /* Anything else */ + } else { + /* If the current node is a head element, act as if an end tag + token with the tag name "head" had been seen. */ + if($this->head_pointer->isSameNode(end($this->stack))) { + $this->inHead(array( + 'name' => 'head', + 'type' => HTML5::ENDTAG + )); + + /* Otherwise, change the insertion mode to "after head". */ + } else { + $this->mode = self::AFTER_HEAD; + } + + /* Then, reprocess the current token. */ + return $this->afterHead($token); + } + } + + private function afterHead($token) { + /* Handle the token as follows: */ + + /* A character token that is one of one of U+0009 CHARACTER TABULATION, + U+000A LINE FEED (LF), U+000B LINE TABULATION, U+000C FORM FEED (FF), + or U+0020 SPACE */ + if($token['type'] === HTML5::CHARACTR && + preg_match('/^[\t\n\x0b\x0c ]+$/', $token['data'])) { + /* Append the character to the current node. */ + $this->insertText($token['data']); + + /* A comment token */ + } elseif($token['type'] === HTML5::COMMENT) { + /* Append a Comment node to the current node with the data attribute + set to the data given in the comment token. */ + $this->insertComment($token['data']); + + /* A start tag token with the tag name "body" */ + } elseif($token['type'] === HTML5::STARTTAG && $token['name'] === 'body') { + /* Insert a body element for the token. */ + $this->insertElement($token); + + /* Change the insertion mode to "in body". */ + $this->mode = self::IN_BODY; + + /* A start tag token with the tag name "frameset" */ + } elseif($token['type'] === HTML5::STARTTAG && $token['name'] === 'frameset') { + /* Insert a frameset element for the token. */ + $this->insertElement($token); + + /* Change the insertion mode to "in frameset". */ + $this->mode = self::IN_FRAME; + + /* A start tag token whose tag name is one of: "base", "link", "meta", + "script", "style", "title" */ + } elseif($token['type'] === HTML5::STARTTAG && in_array($token['name'], + array('base', 'link', 'meta', 'script', 'style', 'title'))) { + /* Parse error. Switch the insertion mode back to "in head" and + reprocess the token. */ + $this->mode = self::IN_HEAD; + return $this->inHead($token); + + /* Anything else */ + } else { + /* Act as if a start tag token with the tag name "body" and no + attributes had been seen, and then reprocess the current token. */ + $this->afterHead(array( + 'name' => 'body', + 'type' => HTML5::STARTTAG, + 'attr' => array() + )); + + return $this->inBody($token); + } + } + + private function inBody($token) { + /* Handle the token as follows: */ + + switch($token['type']) { + /* A character token */ + case HTML5::CHARACTR: + /* Reconstruct the active formatting elements, if any. */ + $this->reconstructActiveFormattingElements(); + + /* Append the token's character to the current node. */ + $this->insertText($token['data']); + break; + + /* A comment token */ + case HTML5::COMMENT: + /* Append a Comment node to the current node with the data + attribute set to the data given in the comment token. */ + $this->insertComment($token['data']); + break; + + case HTML5::STARTTAG: + switch($token['name']) { + /* A start tag token whose tag name is one of: "script", + "style" */ + case 'script': case 'style': + /* Process the token as if the insertion mode had been "in + head". */ + return $this->inHead($token); + break; + + /* A start tag token whose tag name is one of: "base", "link", + "meta", "title" */ + case 'base': case 'link': case 'meta': case 'title': + /* Parse error. Process the token as if the insertion mode + had been "in head". */ + return $this->inHead($token); + break; + + /* A start tag token with the tag name "body" */ + case 'body': + /* Parse error. If the second element on the stack of open + elements is not a body element, or, if the stack of open + elements has only one node on it, then ignore the token. + (innerHTML case) */ + if(count($this->stack) === 1 || $this->stack[1]->nodeName !== 'body') { + // Ignore + + /* Otherwise, for each attribute on the token, check to see + if the attribute is already present on the body element (the + second element) on the stack of open elements. If it is not, + add the attribute and its corresponding value to that + element. */ + } else { + foreach($token['attr'] as $attr) { + if(!$this->stack[1]->hasAttribute($attr['name'])) { + $this->stack[1]->setAttribute($attr['name'], $attr['value']); + } + } + } + break; + + /* A start tag whose tag name is one of: "address", + "blockquote", "center", "dir", "div", "dl", "fieldset", + "listing", "menu", "ol", "p", "ul" */ + case 'address': case 'blockquote': case 'center': case 'dir': + case 'div': case 'dl': case 'fieldset': case 'listing': + case 'menu': case 'ol': case 'p': case 'ul': + /* If the stack of open elements has a p element in scope, + then act as if an end tag with the tag name p had been + seen. */ + if($this->elementInScope('p')) { + $this->emitToken(array( + 'name' => 'p', + 'type' => HTML5::ENDTAG + )); + } + + /* Insert an HTML element for the token. */ + $this->insertElement($token); + break; + + /* A start tag whose tag name is "form" */ + case 'form': + /* If the form element pointer is not null, ignore the + token with a parse error. */ + if($this->form_pointer !== null) { + // Ignore. + + /* Otherwise: */ + } else { + /* If the stack of open elements has a p element in + scope, then act as if an end tag with the tag name p + had been seen. */ + if($this->elementInScope('p')) { + $this->emitToken(array( + 'name' => 'p', + 'type' => HTML5::ENDTAG + )); + } + + /* Insert an HTML element for the token, and set the + form element pointer to point to the element created. */ + $element = $this->insertElement($token); + $this->form_pointer = $element; + } + break; + + /* A start tag whose tag name is "li", "dd" or "dt" */ + case 'li': case 'dd': case 'dt': + /* If the stack of open elements has a p element in scope, + then act as if an end tag with the tag name p had been + seen. */ + if($this->elementInScope('p')) { + $this->emitToken(array( + 'name' => 'p', + 'type' => HTML5::ENDTAG + )); + } + + $stack_length = count($this->stack) - 1; + + for($n = $stack_length; 0 <= $n; $n--) { + /* 1. Initialise node to be the current node (the + bottommost node of the stack). */ + $stop = false; + $node = $this->stack[$n]; + $cat = $this->getElementCategory($node->tagName); + + /* 2. If node is an li, dd or dt element, then pop all + the nodes from the current node up to node, including + node, then stop this algorithm. */ + if($token['name'] === $node->tagName || ($token['name'] !== 'li' + && ($node->tagName === 'dd' || $node->tagName === 'dt'))) { + for($x = $stack_length; $x >= $n ; $x--) { + array_pop($this->stack); + } + + break; + } + + /* 3. If node is not in the formatting category, and is + not in the phrasing category, and is not an address or + div element, then stop this algorithm. */ + if($cat !== self::FORMATTING && $cat !== self::PHRASING && + $node->tagName !== 'address' && $node->tagName !== 'div') { + break; + } + } + + /* Finally, insert an HTML element with the same tag + name as the token's. */ + $this->insertElement($token); + break; + + /* A start tag token whose tag name is "plaintext" */ + case 'plaintext': + /* If the stack of open elements has a p element in scope, + then act as if an end tag with the tag name p had been + seen. */ + if($this->elementInScope('p')) { + $this->emitToken(array( + 'name' => 'p', + 'type' => HTML5::ENDTAG + )); + } + + /* Insert an HTML element for the token. */ + $this->insertElement($token); + + return HTML5::PLAINTEXT; + break; + + /* A start tag whose tag name is one of: "h1", "h2", "h3", "h4", + "h5", "h6" */ + case 'h1': case 'h2': case 'h3': case 'h4': case 'h5': case 'h6': + /* If the stack of open elements has a p element in scope, + then act as if an end tag with the tag name p had been seen. */ + if($this->elementInScope('p')) { + $this->emitToken(array( + 'name' => 'p', + 'type' => HTML5::ENDTAG + )); + } + + /* If the stack of open elements has in scope an element whose + tag name is one of "h1", "h2", "h3", "h4", "h5", or "h6", then + this is a parse error; pop elements from the stack until an + element with one of those tag names has been popped from the + stack. */ + while($this->elementInScope(array('h1', 'h2', 'h3', 'h4', 'h5', 'h6'))) { + array_pop($this->stack); + } + + /* Insert an HTML element for the token. */ + $this->insertElement($token); + break; + + /* A start tag whose tag name is "a" */ + case 'a': + /* If the list of active formatting elements contains + an element whose tag name is "a" between the end of the + list and the last marker on the list (or the start of + the list if there is no marker on the list), then this + is a parse error; act as if an end tag with the tag name + "a" had been seen, then remove that element from the list + of active formatting elements and the stack of open + elements if the end tag didn't already remove it (it + might not have if the element is not in table scope). */ + $leng = count($this->a_formatting); + + for($n = $leng - 1; $n >= 0; $n--) { + if($this->a_formatting[$n] === self::MARKER) { + break; + + } elseif($this->a_formatting[$n]->nodeName === 'a') { + $this->emitToken(array( + 'name' => 'a', + 'type' => HTML5::ENDTAG + )); + break; + } + } + + /* Reconstruct the active formatting elements, if any. */ + $this->reconstructActiveFormattingElements(); + + /* Insert an HTML element for the token. */ + $el = $this->insertElement($token); + + /* Add that element to the list of active formatting + elements. */ + $this->a_formatting[] = $el; + break; + + /* A start tag whose tag name is one of: "b", "big", "em", "font", + "i", "nobr", "s", "small", "strike", "strong", "tt", "u" */ + case 'b': case 'big': case 'em': case 'font': case 'i': + case 'nobr': case 's': case 'small': case 'strike': + case 'strong': case 'tt': case 'u': + /* Reconstruct the active formatting elements, if any. */ + $this->reconstructActiveFormattingElements(); + + /* Insert an HTML element for the token. */ + $el = $this->insertElement($token); + + /* Add that element to the list of active formatting + elements. */ + $this->a_formatting[] = $el; + break; + + /* A start tag token whose tag name is "button" */ + case 'button': + /* If the stack of open elements has a button element in scope, + then this is a parse error; act as if an end tag with the tag + name "button" had been seen, then reprocess the token. (We don't + do that. Unnecessary.) */ + if($this->elementInScope('button')) { + $this->inBody(array( + 'name' => 'button', + 'type' => HTML5::ENDTAG + )); + } + + /* Reconstruct the active formatting elements, if any. */ + $this->reconstructActiveFormattingElements(); + + /* Insert an HTML element for the token. */ + $this->insertElement($token); + + /* Insert a marker at the end of the list of active + formatting elements. */ + $this->a_formatting[] = self::MARKER; + break; + + /* A start tag token whose tag name is one of: "marquee", "object" */ + case 'marquee': case 'object': + /* Reconstruct the active formatting elements, if any. */ + $this->reconstructActiveFormattingElements(); + + /* Insert an HTML element for the token. */ + $this->insertElement($token); + + /* Insert a marker at the end of the list of active + formatting elements. */ + $this->a_formatting[] = self::MARKER; + break; + + /* A start tag token whose tag name is "xmp" */ + case 'xmp': + /* Reconstruct the active formatting elements, if any. */ + $this->reconstructActiveFormattingElements(); + + /* Insert an HTML element for the token. */ + $this->insertElement($token); + + /* Switch the content model flag to the CDATA state. */ + return HTML5::CDATA; + break; + + /* A start tag whose tag name is "table" */ + case 'table': + /* If the stack of open elements has a p element in scope, + then act as if an end tag with the tag name p had been seen. */ + if($this->elementInScope('p')) { + $this->emitToken(array( + 'name' => 'p', + 'type' => HTML5::ENDTAG + )); + } + + /* Insert an HTML element for the token. */ + $this->insertElement($token); + + /* Change the insertion mode to "in table". */ + $this->mode = self::IN_TABLE; + break; + + /* A start tag whose tag name is one of: "area", "basefont", + "bgsound", "br", "embed", "img", "param", "spacer", "wbr" */ + case 'area': case 'basefont': case 'bgsound': case 'br': + case 'embed': case 'img': case 'param': case 'spacer': + case 'wbr': + /* Reconstruct the active formatting elements, if any. */ + $this->reconstructActiveFormattingElements(); + + /* Insert an HTML element for the token. */ + $this->insertElement($token); + + /* Immediately pop the current node off the stack of open elements. */ + array_pop($this->stack); + break; + + /* A start tag whose tag name is "hr" */ + case 'hr': + /* If the stack of open elements has a p element in scope, + then act as if an end tag with the tag name p had been seen. */ + if($this->elementInScope('p')) { + $this->emitToken(array( + 'name' => 'p', + 'type' => HTML5::ENDTAG + )); + } + + /* Insert an HTML element for the token. */ + $this->insertElement($token); + + /* Immediately pop the current node off the stack of open elements. */ + array_pop($this->stack); + break; + + /* A start tag whose tag name is "image" */ + case 'image': + /* Parse error. Change the token's tag name to "img" and + reprocess it. (Don't ask.) */ + $token['name'] = 'img'; + return $this->inBody($token); + break; + + /* A start tag whose tag name is "input" */ + case 'input': + /* Reconstruct the active formatting elements, if any. */ + $this->reconstructActiveFormattingElements(); + + /* Insert an input element for the token. */ + $element = $this->insertElement($token, false); + + /* If the form element pointer is not null, then associate the + input element with the form element pointed to by the form + element pointer. */ + $this->form_pointer !== null + ? $this->form_pointer->appendChild($element) + : end($this->stack)->appendChild($element); + + /* Pop that input element off the stack of open elements. */ + array_pop($this->stack); + break; + + /* A start tag whose tag name is "isindex" */ + case 'isindex': + /* Parse error. */ + // w/e + + /* If the form element pointer is not null, + then ignore the token. */ + if($this->form_pointer === null) { + /* Act as if a start tag token with the tag name "form" had + been seen. */ + $this->inBody(array( + 'name' => 'body', + 'type' => HTML5::STARTTAG, + 'attr' => array() + )); + + /* Act as if a start tag token with the tag name "hr" had + been seen. */ + $this->inBody(array( + 'name' => 'hr', + 'type' => HTML5::STARTTAG, + 'attr' => array() + )); + + /* Act as if a start tag token with the tag name "p" had + been seen. */ + $this->inBody(array( + 'name' => 'p', + 'type' => HTML5::STARTTAG, + 'attr' => array() + )); + + /* Act as if a start tag token with the tag name "label" + had been seen. */ + $this->inBody(array( + 'name' => 'label', + 'type' => HTML5::STARTTAG, + 'attr' => array() + )); + + /* Act as if a stream of character tokens had been seen. */ + $this->insertText('This is a searchable index. '. + 'Insert your search keywords here: '); + + /* Act as if a start tag token with the tag name "input" + had been seen, with all the attributes from the "isindex" + token, except with the "name" attribute set to the value + "isindex" (ignoring any explicit "name" attribute). */ + $attr = $token['attr']; + $attr[] = array('name' => 'name', 'value' => 'isindex'); + + $this->inBody(array( + 'name' => 'input', + 'type' => HTML5::STARTTAG, + 'attr' => $attr + )); + + /* Act as if a stream of character tokens had been seen + (see below for what they should say). */ + $this->insertText('This is a searchable index. '. + 'Insert your search keywords here: '); + + /* Act as if an end tag token with the tag name "label" + had been seen. */ + $this->inBody(array( + 'name' => 'label', + 'type' => HTML5::ENDTAG + )); + + /* Act as if an end tag token with the tag name "p" had + been seen. */ + $this->inBody(array( + 'name' => 'p', + 'type' => HTML5::ENDTAG + )); + + /* Act as if a start tag token with the tag name "hr" had + been seen. */ + $this->inBody(array( + 'name' => 'hr', + 'type' => HTML5::ENDTAG + )); + + /* Act as if an end tag token with the tag name "form" had + been seen. */ + $this->inBody(array( + 'name' => 'form', + 'type' => HTML5::ENDTAG + )); + } + break; + + /* A start tag whose tag name is "textarea" */ + case 'textarea': + $this->insertElement($token); + + /* Switch the tokeniser's content model flag to the + RCDATA state. */ + return HTML5::RCDATA; + break; + + /* A start tag whose tag name is one of: "iframe", "noembed", + "noframes" */ + case 'iframe': case 'noembed': case 'noframes': + $this->insertElement($token); + + /* Switch the tokeniser's content model flag to the CDATA state. */ + return HTML5::CDATA; + break; + + /* A start tag whose tag name is "select" */ + case 'select': + /* Reconstruct the active formatting elements, if any. */ + $this->reconstructActiveFormattingElements(); + + /* Insert an HTML element for the token. */ + $this->insertElement($token); + + /* Change the insertion mode to "in select". */ + $this->mode = self::IN_SELECT; + break; + + /* A start or end tag whose tag name is one of: "caption", "col", + "colgroup", "frame", "frameset", "head", "option", "optgroup", + "tbody", "td", "tfoot", "th", "thead", "tr". */ + case 'caption': case 'col': case 'colgroup': case 'frame': + case 'frameset': case 'head': case 'option': case 'optgroup': + case 'tbody': case 'td': case 'tfoot': case 'th': case 'thead': + case 'tr': + // Parse error. Ignore the token. + break; + + /* A start or end tag whose tag name is one of: "event-source", + "section", "nav", "article", "aside", "header", "footer", + "datagrid", "command" */ + case 'event-source': case 'section': case 'nav': case 'article': + case 'aside': case 'header': case 'footer': case 'datagrid': + case 'command': + // Work in progress! + break; + + /* A start tag token not covered by the previous entries */ + default: + /* Reconstruct the active formatting elements, if any. */ + $this->reconstructActiveFormattingElements(); + + $this->insertElement($token, true, true); + break; + } + break; + + case HTML5::ENDTAG: + switch($token['name']) { + /* An end tag with the tag name "body" */ + case 'body': + /* If the second element in the stack of open elements is + not a body element, this is a parse error. Ignore the token. + (innerHTML case) */ + if(count($this->stack) < 2 || $this->stack[1]->nodeName !== 'body') { + // Ignore. + + /* If the current node is not the body element, then this + is a parse error. */ + } elseif(end($this->stack)->nodeName !== 'body') { + // Parse error. + } + + /* Change the insertion mode to "after body". */ + $this->mode = self::AFTER_BODY; + break; + + /* An end tag with the tag name "html" */ + case 'html': + /* Act as if an end tag with tag name "body" had been seen, + then, if that token wasn't ignored, reprocess the current + token. */ + $this->inBody(array( + 'name' => 'body', + 'type' => HTML5::ENDTAG + )); + + return $this->afterBody($token); + break; + + /* An end tag whose tag name is one of: "address", "blockquote", + "center", "dir", "div", "dl", "fieldset", "listing", "menu", + "ol", "pre", "ul" */ + case 'address': case 'blockquote': case 'center': case 'dir': + case 'div': case 'dl': case 'fieldset': case 'listing': + case 'menu': case 'ol': case 'pre': case 'ul': + /* If the stack of open elements has an element in scope + with the same tag name as that of the token, then generate + implied end tags. */ + if($this->elementInScope($token['name'])) { + $this->generateImpliedEndTags(); + + /* Now, if the current node is not an element with + the same tag name as that of the token, then this + is a parse error. */ + // w/e + + /* If the stack of open elements has an element in + scope with the same tag name as that of the token, + then pop elements from this stack until an element + with that tag name has been popped from the stack. */ + for($n = count($this->stack) - 1; $n >= 0; $n--) { + if($this->stack[$n]->nodeName === $token['name']) { + $n = -1; + } + + array_pop($this->stack); + } + } + break; + + /* An end tag whose tag name is "form" */ + case 'form': + /* If the stack of open elements has an element in scope + with the same tag name as that of the token, then generate + implied end tags. */ + if($this->elementInScope($token['name'])) { + $this->generateImpliedEndTags(); + + } + + if(end($this->stack)->nodeName !== $token['name']) { + /* Now, if the current node is not an element with the + same tag name as that of the token, then this is a parse + error. */ + // w/e + + } else { + /* Otherwise, if the current node is an element with + the same tag name as that of the token pop that element + from the stack. */ + array_pop($this->stack); + } + + /* In any case, set the form element pointer to null. */ + $this->form_pointer = null; + break; + + /* An end tag whose tag name is "p" */ + case 'p': + /* If the stack of open elements has a p element in scope, + then generate implied end tags, except for p elements. */ + if($this->elementInScope('p')) { + $this->generateImpliedEndTags(array('p')); + + /* If the current node is not a p element, then this is + a parse error. */ + // k + + /* If the stack of open elements has a p element in + scope, then pop elements from this stack until the stack + no longer has a p element in scope. */ + for($n = count($this->stack) - 1; $n >= 0; $n--) { + if($this->elementInScope('p')) { + array_pop($this->stack); + + } else { + break; + } + } + } + break; + + /* An end tag whose tag name is "dd", "dt", or "li" */ + case 'dd': case 'dt': case 'li': + /* If the stack of open elements has an element in scope + whose tag name matches the tag name of the token, then + generate implied end tags, except for elements with the + same tag name as the token. */ + if($this->elementInScope($token['name'])) { + $this->generateImpliedEndTags(array($token['name'])); + + /* If the current node is not an element with the same + tag name as the token, then this is a parse error. */ + // w/e + + /* If the stack of open elements has an element in scope + whose tag name matches the tag name of the token, then + pop elements from this stack until an element with that + tag name has been popped from the stack. */ + for($n = count($this->stack) - 1; $n >= 0; $n--) { + if($this->stack[$n]->nodeName === $token['name']) { + $n = -1; + } + + array_pop($this->stack); + } + } + break; + + /* An end tag whose tag name is one of: "h1", "h2", "h3", "h4", + "h5", "h6" */ + case 'h1': case 'h2': case 'h3': case 'h4': case 'h5': case 'h6': + $elements = array('h1', 'h2', 'h3', 'h4', 'h5', 'h6'); + + /* If the stack of open elements has in scope an element whose + tag name is one of "h1", "h2", "h3", "h4", "h5", or "h6", then + generate implied end tags. */ + if($this->elementInScope($elements)) { + $this->generateImpliedEndTags(); + + /* Now, if the current node is not an element with the same + tag name as that of the token, then this is a parse error. */ + // w/e + + /* If the stack of open elements has in scope an element + whose tag name is one of "h1", "h2", "h3", "h4", "h5", or + "h6", then pop elements from the stack until an element + with one of those tag names has been popped from the stack. */ + while($this->elementInScope($elements)) { + array_pop($this->stack); + } + } + break; + + /* An end tag whose tag name is one of: "a", "b", "big", "em", + "font", "i", "nobr", "s", "small", "strike", "strong", "tt", "u" */ + case 'a': case 'b': case 'big': case 'em': case 'font': + case 'i': case 'nobr': case 's': case 'small': case 'strike': + case 'strong': case 'tt': case 'u': + /* 1. Let the formatting element be the last element in + the list of active formatting elements that: + * is between the end of the list and the last scope + marker in the list, if any, or the start of the list + otherwise, and + * has the same tag name as the token. + */ + while(true) { + for($a = count($this->a_formatting) - 1; $a >= 0; $a--) { + if($this->a_formatting[$a] === self::MARKER) { + break; + + } elseif($this->a_formatting[$a]->tagName === $token['name']) { + $formatting_element = $this->a_formatting[$a]; + $in_stack = in_array($formatting_element, $this->stack, true); + $fe_af_pos = $a; + break; + } + } + + /* If there is no such node, or, if that node is + also in the stack of open elements but the element + is not in scope, then this is a parse error. Abort + these steps. The token is ignored. */ + if(!isset($formatting_element) || ($in_stack && + !$this->elementInScope($token['name']))) { + break; + + /* Otherwise, if there is such a node, but that node + is not in the stack of open elements, then this is a + parse error; remove the element from the list, and + abort these steps. */ + } elseif(isset($formatting_element) && !$in_stack) { + unset($this->a_formatting[$fe_af_pos]); + $this->a_formatting = array_merge($this->a_formatting); + break; + } + + /* 2. Let the furthest block be the topmost node in the + stack of open elements that is lower in the stack + than the formatting element, and is not an element in + the phrasing or formatting categories. There might + not be one. */ + $fe_s_pos = array_search($formatting_element, $this->stack, true); + $length = count($this->stack); + + for($s = $fe_s_pos + 1; $s < $length; $s++) { + $category = $this->getElementCategory($this->stack[$s]->nodeName); + + if($category !== self::PHRASING && $category !== self::FORMATTING) { + $furthest_block = $this->stack[$s]; + } + } + + /* 3. If there is no furthest block, then the UA must + skip the subsequent steps and instead just pop all + the nodes from the bottom of the stack of open + elements, from the current node up to the formatting + element, and remove the formatting element from the + list of active formatting elements. */ + if(!isset($furthest_block)) { + for($n = $length - 1; $n >= $fe_s_pos; $n--) { + array_pop($this->stack); + } + + unset($this->a_formatting[$fe_af_pos]); + $this->a_formatting = array_merge($this->a_formatting); + break; + } + + /* 4. Let the common ancestor be the element + immediately above the formatting element in the stack + of open elements. */ + $common_ancestor = $this->stack[$fe_s_pos - 1]; + + /* 5. If the furthest block has a parent node, then + remove the furthest block from its parent node. */ + if($furthest_block->parentNode !== null) { + $furthest_block->parentNode->removeChild($furthest_block); + } + + /* 6. Let a bookmark note the position of the + formatting element in the list of active formatting + elements relative to the elements on either side + of it in the list. */ + $bookmark = $fe_af_pos; + + /* 7. Let node and last node be the furthest block. + Follow these steps: */ + $node = $furthest_block; + $last_node = $furthest_block; + + while(true) { + for($n = array_search($node, $this->stack, true) - 1; $n >= 0; $n--) { + /* 7.1 Let node be the element immediately + prior to node in the stack of open elements. */ + $node = $this->stack[$n]; + + /* 7.2 If node is not in the list of active + formatting elements, then remove node from + the stack of open elements and then go back + to step 1. */ + if(!in_array($node, $this->a_formatting, true)) { + unset($this->stack[$n]); + $this->stack = array_merge($this->stack); + + } else { + break; + } + } + + /* 7.3 Otherwise, if node is the formatting + element, then go to the next step in the overall + algorithm. */ + if($node === $formatting_element) { + break; + + /* 7.4 Otherwise, if last node is the furthest + block, then move the aforementioned bookmark to + be immediately after the node in the list of + active formatting elements. */ + } elseif($last_node === $furthest_block) { + $bookmark = array_search($node, $this->a_formatting, true) + 1; + } + + /* 7.5 If node has any children, perform a + shallow clone of node, replace the entry for + node in the list of active formatting elements + with an entry for the clone, replace the entry + for node in the stack of open elements with an + entry for the clone, and let node be the clone. */ + if($node->hasChildNodes()) { + $clone = $node->cloneNode(); + $s_pos = array_search($node, $this->stack, true); + $a_pos = array_search($node, $this->a_formatting, true); + + $this->stack[$s_pos] = $clone; + $this->a_formatting[$a_pos] = $clone; + $node = $clone; + } + + /* 7.6 Insert last node into node, first removing + it from its previous parent node if any. */ + if($last_node->parentNode !== null) { + $last_node->parentNode->removeChild($last_node); + } + + $node->appendChild($last_node); + + /* 7.7 Let last node be node. */ + $last_node = $node; + } + + /* 8. Insert whatever last node ended up being in + the previous step into the common ancestor node, + first removing it from its previous parent node if + any. */ + if($last_node->parentNode !== null) { + $last_node->parentNode->removeChild($last_node); + } + + $common_ancestor->appendChild($last_node); + + /* 9. Perform a shallow clone of the formatting + element. */ + $clone = $formatting_element->cloneNode(); + + /* 10. Take all of the child nodes of the furthest + block and append them to the clone created in the + last step. */ + while($furthest_block->hasChildNodes()) { + $child = $furthest_block->firstChild; + $furthest_block->removeChild($child); + $clone->appendChild($child); + } + + /* 11. Append that clone to the furthest block. */ + $furthest_block->appendChild($clone); + + /* 12. Remove the formatting element from the list + of active formatting elements, and insert the clone + into the list of active formatting elements at the + position of the aforementioned bookmark. */ + $fe_af_pos = array_search($formatting_element, $this->a_formatting, true); + unset($this->a_formatting[$fe_af_pos]); + $this->a_formatting = array_merge($this->a_formatting); + + $af_part1 = array_slice($this->a_formatting, 0, $bookmark - 1); + $af_part2 = array_slice($this->a_formatting, $bookmark, count($this->a_formatting)); + $this->a_formatting = array_merge($af_part1, array($clone), $af_part2); + + /* 13. Remove the formatting element from the stack + of open elements, and insert the clone into the stack + of open elements immediately after (i.e. in a more + deeply nested position than) the position of the + furthest block in that stack. */ + $fe_s_pos = array_search($formatting_element, $this->stack, true); + $fb_s_pos = array_search($furthest_block, $this->stack, true); + unset($this->stack[$fe_s_pos]); + + $s_part1 = array_slice($this->stack, 0, $fb_s_pos); + $s_part2 = array_slice($this->stack, $fb_s_pos + 1, count($this->stack)); + $this->stack = array_merge($s_part1, array($clone), $s_part2); + + /* 14. Jump back to step 1 in this series of steps. */ + unset($formatting_element, $fe_af_pos, $fe_s_pos, $furthest_block); + } + break; + + /* An end tag token whose tag name is one of: "button", + "marquee", "object" */ + case 'button': case 'marquee': case 'object': + /* If the stack of open elements has an element in scope whose + tag name matches the tag name of the token, then generate implied + tags. */ + if($this->elementInScope($token['name'])) { + $this->generateImpliedEndTags(); + + /* Now, if the current node is not an element with the same + tag name as the token, then this is a parse error. */ + // k + + /* Now, if the stack of open elements has an element in scope + whose tag name matches the tag name of the token, then pop + elements from the stack until that element has been popped from + the stack, and clear the list of active formatting elements up + to the last marker. */ + for($n = count($this->stack) - 1; $n >= 0; $n--) { + if($this->stack[$n]->nodeName === $token['name']) { + $n = -1; + } + + array_pop($this->stack); + } + + $marker = end(array_keys($this->a_formatting, self::MARKER, true)); + + for($n = count($this->a_formatting) - 1; $n > $marker; $n--) { + array_pop($this->a_formatting); + } + } + break; + + /* Or an end tag whose tag name is one of: "area", "basefont", + "bgsound", "br", "embed", "hr", "iframe", "image", "img", + "input", "isindex", "noembed", "noframes", "param", "select", + "spacer", "table", "textarea", "wbr" */ + case 'area': case 'basefont': case 'bgsound': case 'br': + case 'embed': case 'hr': case 'iframe': case 'image': + case 'img': case 'input': case 'isindex': case 'noembed': + case 'noframes': case 'param': case 'select': case 'spacer': + case 'table': case 'textarea': case 'wbr': + // Parse error. Ignore the token. + break; + + /* An end tag token not covered by the previous entries */ + default: + for($n = count($this->stack) - 1; $n >= 0; $n--) { + /* Initialise node to be the current node (the bottommost + node of the stack). */ + $node = end($this->stack); + + /* If node has the same tag name as the end tag token, + then: */ + if($token['name'] === $node->nodeName) { + /* Generate implied end tags. */ + $this->generateImpliedEndTags(); + + /* If the tag name of the end tag token does not + match the tag name of the current node, this is a + parse error. */ + // k + + /* Pop all the nodes from the current node up to + node, including node, then stop this algorithm. */ + for($x = count($this->stack) - $n; $x >= $n; $x--) { + array_pop($this->stack); + } + + } else { + $category = $this->getElementCategory($node); + + if($category !== self::SPECIAL && $category !== self::SCOPING) { + /* Otherwise, if node is in neither the formatting + category nor the phrasing category, then this is a + parse error. Stop this algorithm. The end tag token + is ignored. */ + return false; + } + } + } + break; + } + break; + } + } + + private function inTable($token) { + $clear = array('html', 'table'); + + /* A character token that is one of one of U+0009 CHARACTER TABULATION, + U+000A LINE FEED (LF), U+000B LINE TABULATION, U+000C FORM FEED (FF), + or U+0020 SPACE */ + if($token['type'] === HTML5::CHARACTR && + preg_match('/^[\t\n\x0b\x0c ]+$/', $token['data'])) { + /* Append the character to the current node. */ + $text = $this->dom->createTextNode($token['data']); + end($this->stack)->appendChild($text); + + /* A comment token */ + } elseif($token['type'] === HTML5::COMMENT) { + /* Append a Comment node to the current node with the data + attribute set to the data given in the comment token. */ + $comment = $this->dom->createComment($token['data']); + end($this->stack)->appendChild($comment); + + /* A start tag whose tag name is "caption" */ + } elseif($token['type'] === HTML5::STARTTAG && + $token['name'] === 'caption') { + /* Clear the stack back to a table context. */ + $this->clearStackToTableContext($clear); + + /* Insert a marker at the end of the list of active + formatting elements. */ + $this->a_formatting[] = self::MARKER; + + /* Insert an HTML element for the token, then switch the + insertion mode to "in caption". */ + $this->insertElement($token); + $this->mode = self::IN_CAPTION; + + /* A start tag whose tag name is "colgroup" */ + } elseif($token['type'] === HTML5::STARTTAG && + $token['name'] === 'colgroup') { + /* Clear the stack back to a table context. */ + $this->clearStackToTableContext($clear); + + /* Insert an HTML element for the token, then switch the + insertion mode to "in column group". */ + $this->insertElement($token); + $this->mode = self::IN_CGROUP; + + /* A start tag whose tag name is "col" */ + } elseif($token['type'] === HTML5::STARTTAG && + $token['name'] === 'col') { + $this->inTable(array( + 'name' => 'colgroup', + 'type' => HTML5::STARTTAG, + 'attr' => array() + )); + + $this->inColumnGroup($token); + + /* A start tag whose tag name is one of: "tbody", "tfoot", "thead" */ + } elseif($token['type'] === HTML5::STARTTAG && in_array($token['name'], + array('tbody', 'tfoot', 'thead'))) { + /* Clear the stack back to a table context. */ + $this->clearStackToTableContext($clear); + + /* Insert an HTML element for the token, then switch the insertion + mode to "in table body". */ + $this->insertElement($token); + $this->mode = self::IN_TBODY; + + /* A start tag whose tag name is one of: "td", "th", "tr" */ + } elseif($token['type'] === HTML5::STARTTAG && + in_array($token['name'], array('td', 'th', 'tr'))) { + /* Act as if a start tag token with the tag name "tbody" had been + seen, then reprocess the current token. */ + $this->inTable(array( + 'name' => 'tbody', + 'type' => HTML5::STARTTAG, + 'attr' => array() + )); + + return $this->inTableBody($token); + + /* A start tag whose tag name is "table" */ + } elseif($token['type'] === HTML5::STARTTAG && + $token['name'] === 'table') { + /* Parse error. Act as if an end tag token with the tag name "table" + had been seen, then, if that token wasn't ignored, reprocess the + current token. */ + $this->inTable(array( + 'name' => 'table', + 'type' => HTML5::ENDTAG + )); + + return $this->mainPhase($token); + + /* An end tag whose tag name is "table" */ + } elseif($token['type'] === HTML5::ENDTAG && + $token['name'] === 'table') { + /* If the stack of open elements does not have an element in table + scope with the same tag name as the token, this is a parse error. + Ignore the token. (innerHTML case) */ + if(!$this->elementInScope($token['name'], true)) { + return false; + + /* Otherwise: */ + } else { + /* Generate implied end tags. */ + $this->generateImpliedEndTags(); + + /* Now, if the current node is not a table element, then this + is a parse error. */ + // w/e + + /* Pop elements from this stack until a table element has been + popped from the stack. */ + while(true) { + $current = end($this->stack)->nodeName; + array_pop($this->stack); + + if($current === 'table') { + break; + } + } + + /* Reset the insertion mode appropriately. */ + $this->resetInsertionMode(); + } + + /* An end tag whose tag name is one of: "body", "caption", "col", + "colgroup", "html", "tbody", "td", "tfoot", "th", "thead", "tr" */ + } elseif($token['type'] === HTML5::ENDTAG && in_array($token['name'], + array('body', 'caption', 'col', 'colgroup', 'html', 'tbody', 'td', + 'tfoot', 'th', 'thead', 'tr'))) { + // Parse error. Ignore the token. + + /* Anything else */ + } else { + /* Parse error. Process the token as if the insertion mode was "in + body", with the following exception: */ + + /* If the current node is a table, tbody, tfoot, thead, or tr + element, then, whenever a node would be inserted into the current + node, it must instead be inserted into the foster parent element. */ + if(in_array(end($this->stack)->nodeName, + array('table', 'tbody', 'tfoot', 'thead', 'tr'))) { + /* The foster parent element is the parent element of the last + table element in the stack of open elements, if there is a + table element and it has such a parent element. If there is no + table element in the stack of open elements (innerHTML case), + then the foster parent element is the first element in the + stack of open elements (the html element). Otherwise, if there + is a table element in the stack of open elements, but the last + table element in the stack of open elements has no parent, or + its parent node is not an element, then the foster parent + element is the element before the last table element in the + stack of open elements. */ + for($n = count($this->stack) - 1; $n >= 0; $n--) { + if($this->stack[$n]->nodeName === 'table') { + $table = $this->stack[$n]; + break; + } + } + + if(isset($table) && $table->parentNode !== null) { + $this->foster_parent = $table->parentNode; + + } elseif(!isset($table)) { + $this->foster_parent = $this->stack[0]; + + } elseif(isset($table) && ($table->parentNode === null || + $table->parentNode->nodeType !== XML_ELEMENT_NODE)) { + $this->foster_parent = $this->stack[$n - 1]; + } + } + + $this->inBody($token); + } + } + + private function inCaption($token) { + /* An end tag whose tag name is "caption" */ + if($token['type'] === HTML5::ENDTAG && $token['name'] === 'caption') { + /* If the stack of open elements does not have an element in table + scope with the same tag name as the token, this is a parse error. + Ignore the token. (innerHTML case) */ + if(!$this->elementInScope($token['name'], true)) { + // Ignore + + /* Otherwise: */ + } else { + /* Generate implied end tags. */ + $this->generateImpliedEndTags(); + + /* Now, if the current node is not a caption element, then this + is a parse error. */ + // w/e + + /* Pop elements from this stack until a caption element has + been popped from the stack. */ + while(true) { + $node = end($this->stack)->nodeName; + array_pop($this->stack); + + if($node === 'caption') { + break; + } + } + + /* Clear the list of active formatting elements up to the last + marker. */ + $this->clearTheActiveFormattingElementsUpToTheLastMarker(); + + /* Switch the insertion mode to "in table". */ + $this->mode = self::IN_TABLE; + } + + /* A start tag whose tag name is one of: "caption", "col", "colgroup", + "tbody", "td", "tfoot", "th", "thead", "tr", or an end tag whose tag + name is "table" */ + } elseif(($token['type'] === HTML5::STARTTAG && in_array($token['name'], + array('caption', 'col', 'colgroup', 'tbody', 'td', 'tfoot', 'th', + 'thead', 'tr'))) || ($token['type'] === HTML5::ENDTAG && + $token['name'] === 'table')) { + /* Parse error. Act as if an end tag with the tag name "caption" + had been seen, then, if that token wasn't ignored, reprocess the + current token. */ + $this->inCaption(array( + 'name' => 'caption', + 'type' => HTML5::ENDTAG + )); + + return $this->inTable($token); + + /* An end tag whose tag name is one of: "body", "col", "colgroup", + "html", "tbody", "td", "tfoot", "th", "thead", "tr" */ + } elseif($token['type'] === HTML5::ENDTAG && in_array($token['name'], + array('body', 'col', 'colgroup', 'html', 'tbody', 'tfoot', 'th', + 'thead', 'tr'))) { + // Parse error. Ignore the token. + + /* Anything else */ + } else { + /* Process the token as if the insertion mode was "in body". */ + $this->inBody($token); + } + } + + private function inColumnGroup($token) { + /* A character token that is one of one of U+0009 CHARACTER TABULATION, + U+000A LINE FEED (LF), U+000B LINE TABULATION, U+000C FORM FEED (FF), + or U+0020 SPACE */ + if($token['type'] === HTML5::CHARACTR && + preg_match('/^[\t\n\x0b\x0c ]+$/', $token['data'])) { + /* Append the character to the current node. */ + $text = $this->dom->createTextNode($token['data']); + end($this->stack)->appendChild($text); + + /* A comment token */ + } elseif($token['type'] === HTML5::COMMENT) { + /* Append a Comment node to the current node with the data + attribute set to the data given in the comment token. */ + $comment = $this->dom->createComment($token['data']); + end($this->stack)->appendChild($comment); + + /* A start tag whose tag name is "col" */ + } elseif($token['type'] === HTML5::STARTTAG && $token['name'] === 'col') { + /* Insert a col element for the token. Immediately pop the current + node off the stack of open elements. */ + $this->insertElement($token); + array_pop($this->stack); + + /* An end tag whose tag name is "colgroup" */ + } elseif($token['type'] === HTML5::ENDTAG && + $token['name'] === 'colgroup') { + /* If the current node is the root html element, then this is a + parse error, ignore the token. (innerHTML case) */ + if(end($this->stack)->nodeName === 'html') { + // Ignore + + /* Otherwise, pop the current node (which will be a colgroup + element) from the stack of open elements. Switch the insertion + mode to "in table". */ + } else { + array_pop($this->stack); + $this->mode = self::IN_TABLE; + } + + /* An end tag whose tag name is "col" */ + } elseif($token['type'] === HTML5::ENDTAG && $token['name'] === 'col') { + /* Parse error. Ignore the token. */ + + /* Anything else */ + } else { + /* Act as if an end tag with the tag name "colgroup" had been seen, + and then, if that token wasn't ignored, reprocess the current token. */ + $this->inColumnGroup(array( + 'name' => 'colgroup', + 'type' => HTML5::ENDTAG + )); + + return $this->inTable($token); + } + } + + private function inTableBody($token) { + $clear = array('tbody', 'tfoot', 'thead', 'html'); + + /* A start tag whose tag name is "tr" */ + if($token['type'] === HTML5::STARTTAG && $token['name'] === 'tr') { + /* Clear the stack back to a table body context. */ + $this->clearStackToTableContext($clear); + + /* Insert a tr element for the token, then switch the insertion + mode to "in row". */ + $this->insertElement($token); + $this->mode = self::IN_ROW; + + /* A start tag whose tag name is one of: "th", "td" */ + } elseif($token['type'] === HTML5::STARTTAG && + ($token['name'] === 'th' || $token['name'] === 'td')) { + /* Parse error. Act as if a start tag with the tag name "tr" had + been seen, then reprocess the current token. */ + $this->inTableBody(array( + 'name' => 'tr', + 'type' => HTML5::STARTTAG, + 'attr' => array() + )); + + return $this->inRow($token); + + /* An end tag whose tag name is one of: "tbody", "tfoot", "thead" */ + } elseif($token['type'] === HTML5::ENDTAG && + in_array($token['name'], array('tbody', 'tfoot', 'thead'))) { + /* If the stack of open elements does not have an element in table + scope with the same tag name as the token, this is a parse error. + Ignore the token. */ + if(!$this->elementInScope($token['name'], true)) { + // Ignore + + /* Otherwise: */ + } else { + /* Clear the stack back to a table body context. */ + $this->clearStackToTableContext($clear); + + /* Pop the current node from the stack of open elements. Switch + the insertion mode to "in table". */ + array_pop($this->stack); + $this->mode = self::IN_TABLE; + } + + /* A start tag whose tag name is one of: "caption", "col", "colgroup", + "tbody", "tfoot", "thead", or an end tag whose tag name is "table" */ + } elseif(($token['type'] === HTML5::STARTTAG && in_array($token['name'], + array('caption', 'col', 'colgroup', 'tbody', 'tfoor', 'thead'))) || + ($token['type'] === HTML5::STARTTAG && $token['name'] === 'table')) { + /* If the stack of open elements does not have a tbody, thead, or + tfoot element in table scope, this is a parse error. Ignore the + token. (innerHTML case) */ + if(!$this->elementInScope(array('tbody', 'thead', 'tfoot'), true)) { + // Ignore. + + /* Otherwise: */ + } else { + /* Clear the stack back to a table body context. */ + $this->clearStackToTableContext($clear); + + /* Act as if an end tag with the same tag name as the current + node ("tbody", "tfoot", or "thead") had been seen, then + reprocess the current token. */ + $this->inTableBody(array( + 'name' => end($this->stack)->nodeName, + 'type' => HTML5::ENDTAG + )); + + return $this->mainPhase($token); + } + + /* An end tag whose tag name is one of: "body", "caption", "col", + "colgroup", "html", "td", "th", "tr" */ + } elseif($token['type'] === HTML5::ENDTAG && in_array($token['name'], + array('body', 'caption', 'col', 'colgroup', 'html', 'td', 'th', 'tr'))) { + /* Parse error. Ignore the token. */ + + /* Anything else */ + } else { + /* Process the token as if the insertion mode was "in table". */ + $this->inTable($token); + } + } + + private function inRow($token) { + $clear = array('tr', 'html'); + + /* A start tag whose tag name is one of: "th", "td" */ + if($token['type'] === HTML5::STARTTAG && + ($token['name'] === 'th' || $token['name'] === 'td')) { + /* Clear the stack back to a table row context. */ + $this->clearStackToTableContext($clear); + + /* Insert an HTML element for the token, then switch the insertion + mode to "in cell". */ + $this->insertElement($token); + $this->mode = self::IN_CELL; + + /* Insert a marker at the end of the list of active formatting + elements. */ + $this->a_formatting[] = self::MARKER; + + /* An end tag whose tag name is "tr" */ + } elseif($token['type'] === HTML5::ENDTAG && $token['name'] === 'tr') { + /* If the stack of open elements does not have an element in table + scope with the same tag name as the token, this is a parse error. + Ignore the token. (innerHTML case) */ + if(!$this->elementInScope($token['name'], true)) { + // Ignore. + + /* Otherwise: */ + } else { + /* Clear the stack back to a table row context. */ + $this->clearStackToTableContext($clear); + + /* Pop the current node (which will be a tr element) from the + stack of open elements. Switch the insertion mode to "in table + body". */ + array_pop($this->stack); + $this->mode = self::IN_TBODY; + } + + /* A start tag whose tag name is one of: "caption", "col", "colgroup", + "tbody", "tfoot", "thead", "tr" or an end tag whose tag name is "table" */ + } elseif($token['type'] === HTML5::STARTTAG && in_array($token['name'], + array('caption', 'col', 'colgroup', 'tbody', 'tfoot', 'thead', 'tr'))) { + /* Act as if an end tag with the tag name "tr" had been seen, then, + if that token wasn't ignored, reprocess the current token. */ + $this->inRow(array( + 'name' => 'tr', + 'type' => HTML5::ENDTAG + )); + + return $this->inCell($token); + + /* An end tag whose tag name is one of: "tbody", "tfoot", "thead" */ + } elseif($token['type'] === HTML5::ENDTAG && + in_array($token['name'], array('tbody', 'tfoot', 'thead'))) { + /* If the stack of open elements does not have an element in table + scope with the same tag name as the token, this is a parse error. + Ignore the token. */ + if(!$this->elementInScope($token['name'], true)) { + // Ignore. + + /* Otherwise: */ + } else { + /* Otherwise, act as if an end tag with the tag name "tr" had + been seen, then reprocess the current token. */ + $this->inRow(array( + 'name' => 'tr', + 'type' => HTML5::ENDTAG + )); + + return $this->inCell($token); + } + + /* An end tag whose tag name is one of: "body", "caption", "col", + "colgroup", "html", "td", "th" */ + } elseif($token['type'] === HTML5::ENDTAG && in_array($token['name'], + array('body', 'caption', 'col', 'colgroup', 'html', 'td', 'th', 'tr'))) { + /* Parse error. Ignore the token. */ + + /* Anything else */ + } else { + /* Process the token as if the insertion mode was "in table". */ + $this->inTable($token); + } + } + + private function inCell($token) { + /* An end tag whose tag name is one of: "td", "th" */ + if($token['type'] === HTML5::ENDTAG && + ($token['name'] === 'td' || $token['name'] === 'th')) { + /* If the stack of open elements does not have an element in table + scope with the same tag name as that of the token, then this is a + parse error and the token must be ignored. */ + if(!$this->elementInScope($token['name'], true)) { + // Ignore. + + /* Otherwise: */ + } else { + /* Generate implied end tags, except for elements with the same + tag name as the token. */ + $this->generateImpliedEndTags(array($token['name'])); + + /* Now, if the current node is not an element with the same tag + name as the token, then this is a parse error. */ + // k + + /* Pop elements from this stack until an element with the same + tag name as the token has been popped from the stack. */ + while(true) { + $node = end($this->stack)->nodeName; + array_pop($this->stack); + + if($node === $token['name']) { + break; + } + } + + /* Clear the list of active formatting elements up to the last + marker. */ + $this->clearTheActiveFormattingElementsUpToTheLastMarker(); + + /* Switch the insertion mode to "in row". (The current node + will be a tr element at this point.) */ + $this->mode = self::IN_ROW; + } + + /* A start tag whose tag name is one of: "caption", "col", "colgroup", + "tbody", "td", "tfoot", "th", "thead", "tr" */ + } elseif($token['type'] === HTML5::STARTTAG && in_array($token['name'], + array('caption', 'col', 'colgroup', 'tbody', 'td', 'tfoot', 'th', + 'thead', 'tr'))) { + /* If the stack of open elements does not have a td or th element + in table scope, then this is a parse error; ignore the token. + (innerHTML case) */ + if(!$this->elementInScope(array('td', 'th'), true)) { + // Ignore. + + /* Otherwise, close the cell (see below) and reprocess the current + token. */ + } else { + $this->closeCell(); + return $this->inRow($token); + } + + /* A start tag whose tag name is one of: "caption", "col", "colgroup", + "tbody", "td", "tfoot", "th", "thead", "tr" */ + } elseif($token['type'] === HTML5::STARTTAG && in_array($token['name'], + array('caption', 'col', 'colgroup', 'tbody', 'td', 'tfoot', 'th', + 'thead', 'tr'))) { + /* If the stack of open elements does not have a td or th element + in table scope, then this is a parse error; ignore the token. + (innerHTML case) */ + if(!$this->elementInScope(array('td', 'th'), true)) { + // Ignore. + + /* Otherwise, close the cell (see below) and reprocess the current + token. */ + } else { + $this->closeCell(); + return $this->inRow($token); + } + + /* An end tag whose tag name is one of: "body", "caption", "col", + "colgroup", "html" */ + } elseif($token['type'] === HTML5::ENDTAG && in_array($token['name'], + array('body', 'caption', 'col', 'colgroup', 'html'))) { + /* Parse error. Ignore the token. */ + + /* An end tag whose tag name is one of: "table", "tbody", "tfoot", + "thead", "tr" */ + } elseif($token['type'] === HTML5::ENDTAG && in_array($token['name'], + array('table', 'tbody', 'tfoot', 'thead', 'tr'))) { + /* If the stack of open elements does not have an element in table + scope with the same tag name as that of the token (which can only + happen for "tbody", "tfoot" and "thead", or, in the innerHTML case), + then this is a parse error and the token must be ignored. */ + if(!$this->elementInScope($token['name'], true)) { + // Ignore. + + /* Otherwise, close the cell (see below) and reprocess the current + token. */ + } else { + $this->closeCell(); + return $this->inRow($token); + } + + /* Anything else */ + } else { + /* Process the token as if the insertion mode was "in body". */ + $this->inBody($token); + } + } + + private function inSelect($token) { + /* Handle the token as follows: */ + + /* A character token */ + if($token['type'] === HTML5::CHARACTR) { + /* Append the token's character to the current node. */ + $this->insertText($token['data']); + + /* A comment token */ + } elseif($token['type'] === HTML5::COMMENT) { + /* Append a Comment node to the current node with the data + attribute set to the data given in the comment token. */ + $this->insertComment($token['data']); + + /* A start tag token whose tag name is "option" */ + } elseif($token['type'] === HTML5::STARTTAG && + $token['name'] === 'option') { + /* If the current node is an option element, act as if an end tag + with the tag name "option" had been seen. */ + if(end($this->stack)->nodeName === 'option') { + $this->inSelect(array( + 'name' => 'option', + 'type' => HTML5::ENDTAG + )); + } + + /* Insert an HTML element for the token. */ + $this->insertElement($token); + + /* A start tag token whose tag name is "optgroup" */ + } elseif($token['type'] === HTML5::STARTTAG && + $token['name'] === 'optgroup') { + /* If the current node is an option element, act as if an end tag + with the tag name "option" had been seen. */ + if(end($this->stack)->nodeName === 'option') { + $this->inSelect(array( + 'name' => 'option', + 'type' => HTML5::ENDTAG + )); + } + + /* If the current node is an optgroup element, act as if an end tag + with the tag name "optgroup" had been seen. */ + if(end($this->stack)->nodeName === 'optgroup') { + $this->inSelect(array( + 'name' => 'optgroup', + 'type' => HTML5::ENDTAG + )); + } + + /* Insert an HTML element for the token. */ + $this->insertElement($token); + + /* An end tag token whose tag name is "optgroup" */ + } elseif($token['type'] === HTML5::ENDTAG && + $token['name'] === 'optgroup') { + /* First, if the current node is an option element, and the node + immediately before it in the stack of open elements is an optgroup + element, then act as if an end tag with the tag name "option" had + been seen. */ + $elements_in_stack = count($this->stack); + + if($this->stack[$elements_in_stack - 1]->nodeName === 'option' && + $this->stack[$elements_in_stack - 2]->nodeName === 'optgroup') { + $this->inSelect(array( + 'name' => 'option', + 'type' => HTML5::ENDTAG + )); + } + + /* If the current node is an optgroup element, then pop that node + from the stack of open elements. Otherwise, this is a parse error, + ignore the token. */ + if($this->stack[$elements_in_stack - 1] === 'optgroup') { + array_pop($this->stack); + } + + /* An end tag token whose tag name is "option" */ + } elseif($token['type'] === HTML5::ENDTAG && + $token['name'] === 'option') { + /* If the current node is an option element, then pop that node + from the stack of open elements. Otherwise, this is a parse error, + ignore the token. */ + if(end($this->stack)->nodeName === 'option') { + array_pop($this->stack); + } + + /* An end tag whose tag name is "select" */ + } elseif($token['type'] === HTML5::ENDTAG && + $token['name'] === 'select') { + /* If the stack of open elements does not have an element in table + scope with the same tag name as the token, this is a parse error. + Ignore the token. (innerHTML case) */ + if(!$this->elementInScope($token['name'], true)) { + // w/e + + /* Otherwise: */ + } else { + /* Pop elements from the stack of open elements until a select + element has been popped from the stack. */ + while(true) { + $current = end($this->stack)->nodeName; + array_pop($this->stack); + + if($current === 'select') { + break; + } + } + + /* Reset the insertion mode appropriately. */ + $this->resetInsertionMode(); + } + + /* A start tag whose tag name is "select" */ + } elseif($token['name'] === 'select' && + $token['type'] === HTML5::STARTTAG) { + /* Parse error. Act as if the token had been an end tag with the + tag name "select" instead. */ + $this->inSelect(array( + 'name' => 'select', + 'type' => HTML5::ENDTAG + )); + + /* An end tag whose tag name is one of: "caption", "table", "tbody", + "tfoot", "thead", "tr", "td", "th" */ + } elseif(in_array($token['name'], array('caption', 'table', 'tbody', + 'tfoot', 'thead', 'tr', 'td', 'th')) && $token['type'] === HTML5::ENDTAG) { + /* Parse error. */ + // w/e + + /* If the stack of open elements has an element in table scope with + the same tag name as that of the token, then act as if an end tag + with the tag name "select" had been seen, and reprocess the token. + Otherwise, ignore the token. */ + if($this->elementInScope($token['name'], true)) { + $this->inSelect(array( + 'name' => 'select', + 'type' => HTML5::ENDTAG + )); + + $this->mainPhase($token); + } + + /* Anything else */ + } else { + /* Parse error. Ignore the token. */ + } + } + + private function afterBody($token) { + /* Handle the token as follows: */ + + /* A character token that is one of one of U+0009 CHARACTER TABULATION, + U+000A LINE FEED (LF), U+000B LINE TABULATION, U+000C FORM FEED (FF), + or U+0020 SPACE */ + if($token['type'] === HTML5::CHARACTR && + preg_match('/^[\t\n\x0b\x0c ]+$/', $token['data'])) { + /* Process the token as it would be processed if the insertion mode + was "in body". */ + $this->inBody($token); + + /* A comment token */ + } elseif($token['type'] === HTML5::COMMENT) { + /* Append a Comment node to the first element in the stack of open + elements (the html element), with the data attribute set to the + data given in the comment token. */ + $comment = $this->dom->createComment($token['data']); + $this->stack[0]->appendChild($comment); + + /* An end tag with the tag name "html" */ + } elseif($token['type'] === HTML5::ENDTAG && $token['name'] === 'html') { + /* If the parser was originally created in order to handle the + setting of an element's innerHTML attribute, this is a parse error; + ignore the token. (The element will be an html element in this + case.) (innerHTML case) */ + + /* Otherwise, switch to the trailing end phase. */ + $this->phase = self::END_PHASE; + + /* Anything else */ + } else { + /* Parse error. Set the insertion mode to "in body" and reprocess + the token. */ + $this->mode = self::IN_BODY; + return $this->inBody($token); + } + } + + private function inFrameset($token) { + /* Handle the token as follows: */ + + /* A character token that is one of one of U+0009 CHARACTER TABULATION, + U+000A LINE FEED (LF), U+000B LINE TABULATION, U+000C FORM FEED (FF), + U+000D CARRIAGE RETURN (CR), or U+0020 SPACE */ + if($token['type'] === HTML5::CHARACTR && + preg_match('/^[\t\n\x0b\x0c ]+$/', $token['data'])) { + /* Append the character to the current node. */ + $this->insertText($token['data']); + + /* A comment token */ + } elseif($token['type'] === HTML5::COMMENT) { + /* Append a Comment node to the current node with the data + attribute set to the data given in the comment token. */ + $this->insertComment($token['data']); + + /* A start tag with the tag name "frameset" */ + } elseif($token['name'] === 'frameset' && + $token['type'] === HTML5::STARTTAG) { + $this->insertElement($token); + + /* An end tag with the tag name "frameset" */ + } elseif($token['name'] === 'frameset' && + $token['type'] === HTML5::ENDTAG) { + /* If the current node is the root html element, then this is a + parse error; ignore the token. (innerHTML case) */ + if(end($this->stack)->nodeName === 'html') { + // Ignore + + } else { + /* Otherwise, pop the current node from the stack of open + elements. */ + array_pop($this->stack); + + /* If the parser was not originally created in order to handle + the setting of an element's innerHTML attribute (innerHTML case), + and the current node is no longer a frameset element, then change + the insertion mode to "after frameset". */ + $this->mode = self::AFTR_FRAME; + } + + /* A start tag with the tag name "frame" */ + } elseif($token['name'] === 'frame' && + $token['type'] === HTML5::STARTTAG) { + /* Insert an HTML element for the token. */ + $this->insertElement($token); + + /* Immediately pop the current node off the stack of open elements. */ + array_pop($this->stack); + + /* A start tag with the tag name "noframes" */ + } elseif($token['name'] === 'noframes' && + $token['type'] === HTML5::STARTTAG) { + /* Process the token as if the insertion mode had been "in body". */ + $this->inBody($token); + + /* Anything else */ + } else { + /* Parse error. Ignore the token. */ + } + } + + private function afterFrameset($token) { + /* Handle the token as follows: */ + + /* A character token that is one of one of U+0009 CHARACTER TABULATION, + U+000A LINE FEED (LF), U+000B LINE TABULATION, U+000C FORM FEED (FF), + U+000D CARRIAGE RETURN (CR), or U+0020 SPACE */ + if($token['type'] === HTML5::CHARACTR && + preg_match('/^[\t\n\x0b\x0c ]+$/', $token['data'])) { + /* Append the character to the current node. */ + $this->insertText($token['data']); + + /* A comment token */ + } elseif($token['type'] === HTML5::COMMENT) { + /* Append a Comment node to the current node with the data + attribute set to the data given in the comment token. */ + $this->insertComment($token['data']); + + /* An end tag with the tag name "html" */ + } elseif($token['name'] === 'html' && + $token['type'] === HTML5::ENDTAG) { + /* Switch to the trailing end phase. */ + $this->phase = self::END_PHASE; + + /* A start tag with the tag name "noframes" */ + } elseif($token['name'] === 'noframes' && + $token['type'] === HTML5::STARTTAG) { + /* Process the token as if the insertion mode had been "in body". */ + $this->inBody($token); + + /* Anything else */ + } else { + /* Parse error. Ignore the token. */ + } + } + + private function trailingEndPhase($token) { + /* After the main phase, as each token is emitted from the tokenisation + stage, it must be processed as described in this section. */ + + /* A DOCTYPE token */ + if($token['type'] === HTML5::DOCTYPE) { + // Parse error. Ignore the token. + + /* A comment token */ + } elseif($token['type'] === HTML5::COMMENT) { + /* Append a Comment node to the Document object with the data + attribute set to the data given in the comment token. */ + $comment = $this->dom->createComment($token['data']); + $this->dom->appendChild($comment); + + /* A character token that is one of one of U+0009 CHARACTER TABULATION, + U+000A LINE FEED (LF), U+000B LINE TABULATION, U+000C FORM FEED (FF), + or U+0020 SPACE */ + } elseif($token['type'] === HTML5::CHARACTR && + preg_match('/^[\t\n\x0b\x0c ]+$/', $token['data'])) { + /* Process the token as it would be processed in the main phase. */ + $this->mainPhase($token); + + /* A character token that is not one of U+0009 CHARACTER TABULATION, + U+000A LINE FEED (LF), U+000B LINE TABULATION, U+000C FORM FEED (FF), + or U+0020 SPACE. Or a start tag token. Or an end tag token. */ + } elseif(($token['type'] === HTML5::CHARACTR && + preg_match('/^[\t\n\x0b\x0c ]+$/', $token['data'])) || + $token['type'] === HTML5::STARTTAG || $token['type'] === HTML5::ENDTAG) { + /* Parse error. Switch back to the main phase and reprocess the + token. */ + $this->phase = self::MAIN_PHASE; + return $this->mainPhase($token); + + /* An end-of-file token */ + } elseif($token['type'] === HTML5::EOF) { + /* OMG DONE!! */ + } + } + + private function insertElement($token, $append = true, $check = false) { + // Proprietary workaround for libxml2's limitations with tag names + if ($check) { + // Slightly modified HTML5 tag-name modification, + // removing anything that's not an ASCII letter, digit, or hyphen + $token['name'] = preg_replace('/[^a-z0-9-]/i', '', $token['name']); + // Remove leading hyphens and numbers + $token['name'] = ltrim($token['name'], '-0..9'); + // In theory, this should ever be needed, but just in case + if ($token['name'] === '') $token['name'] = 'span'; // arbitrary generic choice + } + + $el = $this->dom->createElement($token['name']); + + foreach($token['attr'] as $attr) { + if(!$el->hasAttribute($attr['name'])) { + $el->setAttribute($attr['name'], $attr['value']); + } + } + + $this->appendToRealParent($el); + $this->stack[] = $el; + + return $el; + } + + private function insertText($data) { + $text = $this->dom->createTextNode($data); + $this->appendToRealParent($text); + } + + private function insertComment($data) { + $comment = $this->dom->createComment($data); + $this->appendToRealParent($comment); + } + + private function appendToRealParent($node) { + if($this->foster_parent === null) { + end($this->stack)->appendChild($node); + + } elseif($this->foster_parent !== null) { + /* If the foster parent element is the parent element of the + last table element in the stack of open elements, then the new + node must be inserted immediately before the last table element + in the stack of open elements in the foster parent element; + otherwise, the new node must be appended to the foster parent + element. */ + for($n = count($this->stack) - 1; $n >= 0; $n--) { + if($this->stack[$n]->nodeName === 'table' && + $this->stack[$n]->parentNode !== null) { + $table = $this->stack[$n]; + break; + } + } + + if(isset($table) && $this->foster_parent->isSameNode($table->parentNode)) + $this->foster_parent->insertBefore($node, $table); + else + $this->foster_parent->appendChild($node); + + $this->foster_parent = null; + } + } + + private function elementInScope($el, $table = false) { + if(is_array($el)) { + foreach($el as $element) { + if($this->elementInScope($element, $table)) { + return true; + } + } + + return false; + } + + $leng = count($this->stack); + + for($n = 0; $n < $leng; $n++) { + /* 1. Initialise node to be the current node (the bottommost node of + the stack). */ + $node = $this->stack[$leng - 1 - $n]; + + if($node->tagName === $el) { + /* 2. If node is the target node, terminate in a match state. */ + return true; + + } elseif($node->tagName === 'table') { + /* 3. Otherwise, if node is a table element, terminate in a failure + state. */ + return false; + + } elseif($table === true && in_array($node->tagName, array('caption', 'td', + 'th', 'button', 'marquee', 'object'))) { + /* 4. Otherwise, if the algorithm is the "has an element in scope" + variant (rather than the "has an element in table scope" variant), + and node is one of the following, terminate in a failure state. */ + return false; + + } elseif($node === $node->ownerDocument->documentElement) { + /* 5. Otherwise, if node is an html element (root element), terminate + in a failure state. (This can only happen if the node is the topmost + node of the stack of open elements, and prevents the next step from + being invoked if there are no more elements in the stack.) */ + return false; + } + + /* Otherwise, set node to the previous entry in the stack of open + elements and return to step 2. (This will never fail, since the loop + will always terminate in the previous step if the top of the stack + is reached.) */ + } + } + + private function reconstructActiveFormattingElements() { + /* 1. If there are no entries in the list of active formatting elements, + then there is nothing to reconstruct; stop this algorithm. */ + $formatting_elements = count($this->a_formatting); + + if($formatting_elements === 0) { + return false; + } + + /* 3. Let entry be the last (most recently added) element in the list + of active formatting elements. */ + $entry = end($this->a_formatting); + + /* 2. If the last (most recently added) entry in the list of active + formatting elements is a marker, or if it is an element that is in the + stack of open elements, then there is nothing to reconstruct; stop this + algorithm. */ + if($entry === self::MARKER || in_array($entry, $this->stack, true)) { + return false; + } + + for($a = $formatting_elements - 1; $a >= 0; true) { + /* 4. If there are no entries before entry in the list of active + formatting elements, then jump to step 8. */ + if($a === 0) { + $step_seven = false; + break; + } + + /* 5. Let entry be the entry one earlier than entry in the list of + active formatting elements. */ + $a--; + $entry = $this->a_formatting[$a]; + + /* 6. If entry is neither a marker nor an element that is also in + thetack of open elements, go to step 4. */ + if($entry === self::MARKER || in_array($entry, $this->stack, true)) { + break; + } + } + + while(true) { + /* 7. Let entry be the element one later than entry in the list of + active formatting elements. */ + if(isset($step_seven) && $step_seven === true) { + $a++; + $entry = $this->a_formatting[$a]; + } + + /* 8. Perform a shallow clone of the element entry to obtain clone. */ + $clone = $entry->cloneNode(); + + /* 9. Append clone to the current node and push it onto the stack + of open elements so that it is the new current node. */ + end($this->stack)->appendChild($clone); + $this->stack[] = $clone; + + /* 10. Replace the entry for entry in the list with an entry for + clone. */ + $this->a_formatting[$a] = $clone; + + /* 11. If the entry for clone in the list of active formatting + elements is not the last entry in the list, return to step 7. */ + if(end($this->a_formatting) !== $clone) { + $step_seven = true; + } else { + break; + } + } + } + + private function clearTheActiveFormattingElementsUpToTheLastMarker() { + /* When the steps below require the UA to clear the list of active + formatting elements up to the last marker, the UA must perform the + following steps: */ + + while(true) { + /* 1. Let entry be the last (most recently added) entry in the list + of active formatting elements. */ + $entry = end($this->a_formatting); + + /* 2. Remove entry from the list of active formatting elements. */ + array_pop($this->a_formatting); + + /* 3. If entry was a marker, then stop the algorithm at this point. + The list has been cleared up to the last marker. */ + if($entry === self::MARKER) { + break; + } + } + } + + private function generateImpliedEndTags($exclude = array()) { + /* When the steps below require the UA to generate implied end tags, + then, if the current node is a dd element, a dt element, an li element, + a p element, a td element, a th element, or a tr element, the UA must + act as if an end tag with the respective tag name had been seen and + then generate implied end tags again. */ + $node = end($this->stack); + $elements = array_diff(array('dd', 'dt', 'li', 'p', 'td', 'th', 'tr'), $exclude); + + while(in_array(end($this->stack)->nodeName, $elements)) { + array_pop($this->stack); + } + } + + private function getElementCategory($node) { + $name = $node->tagName; + if(in_array($name, $this->special)) + return self::SPECIAL; + + elseif(in_array($name, $this->scoping)) + return self::SCOPING; + + elseif(in_array($name, $this->formatting)) + return self::FORMATTING; + + else + return self::PHRASING; + } + + private function clearStackToTableContext($elements) { + /* When the steps above require the UA to clear the stack back to a + table context, it means that the UA must, while the current node is not + a table element or an html element, pop elements from the stack of open + elements. If this causes any elements to be popped from the stack, then + this is a parse error. */ + while(true) { + $node = end($this->stack)->nodeName; + + if(in_array($node, $elements)) { + break; + } else { + array_pop($this->stack); + } + } + } + + private function resetInsertionMode() { + /* 1. Let last be false. */ + $last = false; + $leng = count($this->stack); + + for($n = $leng - 1; $n >= 0; $n--) { + /* 2. Let node be the last node in the stack of open elements. */ + $node = $this->stack[$n]; + + /* 3. If node is the first node in the stack of open elements, then + set last to true. If the element whose innerHTML attribute is being + set is neither a td element nor a th element, then set node to the + element whose innerHTML attribute is being set. (innerHTML case) */ + if($this->stack[0]->isSameNode($node)) { + $last = true; + } + + /* 4. If node is a select element, then switch the insertion mode to + "in select" and abort these steps. (innerHTML case) */ + if($node->nodeName === 'select') { + $this->mode = self::IN_SELECT; + break; + + /* 5. If node is a td or th element, then switch the insertion mode + to "in cell" and abort these steps. */ + } elseif($node->nodeName === 'td' || $node->nodeName === 'th') { + $this->mode = self::IN_CELL; + break; + + /* 6. If node is a tr element, then switch the insertion mode to + "in row" and abort these steps. */ + } elseif($node->nodeName === 'tr') { + $this->mode = self::IN_ROW; + break; + + /* 7. If node is a tbody, thead, or tfoot element, then switch the + insertion mode to "in table body" and abort these steps. */ + } elseif(in_array($node->nodeName, array('tbody', 'thead', 'tfoot'))) { + $this->mode = self::IN_TBODY; + break; + + /* 8. If node is a caption element, then switch the insertion mode + to "in caption" and abort these steps. */ + } elseif($node->nodeName === 'caption') { + $this->mode = self::IN_CAPTION; + break; + + /* 9. If node is a colgroup element, then switch the insertion mode + to "in column group" and abort these steps. (innerHTML case) */ + } elseif($node->nodeName === 'colgroup') { + $this->mode = self::IN_CGROUP; + break; + + /* 10. If node is a table element, then switch the insertion mode + to "in table" and abort these steps. */ + } elseif($node->nodeName === 'table') { + $this->mode = self::IN_TABLE; + break; + + /* 11. If node is a head element, then switch the insertion mode + to "in body" ("in body"! not "in head"!) and abort these steps. + (innerHTML case) */ + } elseif($node->nodeName === 'head') { + $this->mode = self::IN_BODY; + break; + + /* 12. If node is a body element, then switch the insertion mode to + "in body" and abort these steps. */ + } elseif($node->nodeName === 'body') { + $this->mode = self::IN_BODY; + break; + + /* 13. If node is a frameset element, then switch the insertion + mode to "in frameset" and abort these steps. (innerHTML case) */ + } elseif($node->nodeName === 'frameset') { + $this->mode = self::IN_FRAME; + break; + + /* 14. If node is an html element, then: if the head element + pointer is null, switch the insertion mode to "before head", + otherwise, switch the insertion mode to "after head". In either + case, abort these steps. (innerHTML case) */ + } elseif($node->nodeName === 'html') { + $this->mode = ($this->head_pointer === null) + ? self::BEFOR_HEAD + : self::AFTER_HEAD; + + break; + + /* 15. If last is true, then set the insertion mode to "in body" + and abort these steps. (innerHTML case) */ + } elseif($last) { + $this->mode = self::IN_BODY; + break; + } + } + } + + private function closeCell() { + /* If the stack of open elements has a td or th element in table scope, + then act as if an end tag token with that tag name had been seen. */ + foreach(array('td', 'th') as $cell) { + if($this->elementInScope($cell, true)) { + $this->inCell(array( + 'name' => $cell, + 'type' => HTML5::ENDTAG + )); + + break; + } + } + } + + public function save() { + return $this->dom; + } +} +?> diff --git a/lib/htmlpurifier/library/HTMLPurifier/PercentEncoder.php b/lib/htmlpurifier/library/HTMLPurifier/PercentEncoder.php new file mode 100644 index 00000000..a43c44f4 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/PercentEncoder.php @@ -0,0 +1,98 @@ +preserve[$i] = true; // digits + for ($i = 65; $i <= 90; $i++) $this->preserve[$i] = true; // upper-case + for ($i = 97; $i <= 122; $i++) $this->preserve[$i] = true; // lower-case + $this->preserve[45] = true; // Dash - + $this->preserve[46] = true; // Period . + $this->preserve[95] = true; // Underscore _ + $this->preserve[126]= true; // Tilde ~ + + // extra letters not to escape + if ($preserve !== false) { + for ($i = 0, $c = strlen($preserve); $i < $c; $i++) { + $this->preserve[ord($preserve[$i])] = true; + } + } + } + + /** + * Our replacement for urlencode, it encodes all non-reserved characters, + * as well as any extra characters that were instructed to be preserved. + * @note + * Assumes that the string has already been normalized, making any + * and all percent escape sequences valid. Percents will not be + * re-escaped, regardless of their status in $preserve + * @param $string String to be encoded + * @return Encoded string. + */ + public function encode($string) { + $ret = ''; + for ($i = 0, $c = strlen($string); $i < $c; $i++) { + if ($string[$i] !== '%' && !isset($this->preserve[$int = ord($string[$i])]) ) { + $ret .= '%' . sprintf('%02X', $int); + } else { + $ret .= $string[$i]; + } + } + return $ret; + } + + /** + * Fix up percent-encoding by decoding unreserved characters and normalizing. + * @warning This function is affected by $preserve, even though the + * usual desired behavior is for this not to preserve those + * characters. Be careful when reusing instances of PercentEncoder! + * @param $string String to normalize + */ + public function normalize($string) { + if ($string == '') return ''; + $parts = explode('%', $string); + $ret = array_shift($parts); + foreach ($parts as $part) { + $length = strlen($part); + if ($length < 2) { + $ret .= '%25' . $part; + continue; + } + $encoding = substr($part, 0, 2); + $text = substr($part, 2); + if (!ctype_xdigit($encoding)) { + $ret .= '%25' . $part; + continue; + } + $int = hexdec($encoding); + if (isset($this->preserve[$int])) { + $ret .= chr($int) . $text; + continue; + } + $encoding = strtoupper($encoding); + $ret .= '%' . $encoding . $text; + } + return $ret; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Printer.php b/lib/htmlpurifier/library/HTMLPurifier/Printer.php new file mode 100644 index 00000000..e7eb82e8 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Printer.php @@ -0,0 +1,176 @@ +getAll(); + $context = new HTMLPurifier_Context(); + $this->generator = new HTMLPurifier_Generator($config, $context); + } + + /** + * Main function that renders object or aspect of that object + * @note Parameters vary depending on printer + */ + // function render() {} + + /** + * Returns a start tag + * @param $tag Tag name + * @param $attr Attribute array + */ + protected function start($tag, $attr = array()) { + return $this->generator->generateFromToken( + new HTMLPurifier_Token_Start($tag, $attr ? $attr : array()) + ); + } + + /** + * Returns an end teg + * @param $tag Tag name + */ + protected function end($tag) { + return $this->generator->generateFromToken( + new HTMLPurifier_Token_End($tag) + ); + } + + /** + * Prints a complete element with content inside + * @param $tag Tag name + * @param $contents Element contents + * @param $attr Tag attributes + * @param $escape Bool whether or not to escape contents + */ + protected function element($tag, $contents, $attr = array(), $escape = true) { + return $this->start($tag, $attr) . + ($escape ? $this->escape($contents) : $contents) . + $this->end($tag); + } + + protected function elementEmpty($tag, $attr = array()) { + return $this->generator->generateFromToken( + new HTMLPurifier_Token_Empty($tag, $attr) + ); + } + + protected function text($text) { + return $this->generator->generateFromToken( + new HTMLPurifier_Token_Text($text) + ); + } + + /** + * Prints a simple key/value row in a table. + * @param $name Key + * @param $value Value + */ + protected function row($name, $value) { + if (is_bool($value)) $value = $value ? 'On' : 'Off'; + return + $this->start('tr') . "\n" . + $this->element('th', $name) . "\n" . + $this->element('td', $value) . "\n" . + $this->end('tr') + ; + } + + /** + * Escapes a string for HTML output. + * @param $string String to escape + */ + protected function escape($string) { + $string = HTMLPurifier_Encoder::cleanUTF8($string); + $string = htmlspecialchars($string, ENT_COMPAT, 'UTF-8'); + return $string; + } + + /** + * Takes a list of strings and turns them into a single list + * @param $array List of strings + * @param $polite Bool whether or not to add an end before the last + */ + protected function listify($array, $polite = false) { + if (empty($array)) return 'None'; + $ret = ''; + $i = count($array); + foreach ($array as $value) { + $i--; + $ret .= $value; + if ($i > 0 && !($polite && $i == 1)) $ret .= ', '; + if ($polite && $i == 1) $ret .= 'and '; + } + return $ret; + } + + /** + * Retrieves the class of an object without prefixes, as well as metadata + * @param $obj Object to determine class of + * @param $prefix Further prefix to remove + */ + protected function getClass($obj, $sec_prefix = '') { + static $five = null; + if ($five === null) $five = version_compare(PHP_VERSION, '5', '>='); + $prefix = 'HTMLPurifier_' . $sec_prefix; + if (!$five) $prefix = strtolower($prefix); + $class = str_replace($prefix, '', get_class($obj)); + $lclass = strtolower($class); + $class .= '('; + switch ($lclass) { + case 'enum': + $values = array(); + foreach ($obj->valid_values as $value => $bool) { + $values[] = $value; + } + $class .= implode(', ', $values); + break; + case 'css_composite': + $values = array(); + foreach ($obj->defs as $def) { + $values[] = $this->getClass($def, $sec_prefix); + } + $class .= implode(', ', $values); + break; + case 'css_multiple': + $class .= $this->getClass($obj->single, $sec_prefix) . ', '; + $class .= $obj->max; + break; + case 'css_denyelementdecorator': + $class .= $this->getClass($obj->def, $sec_prefix) . ', '; + $class .= $obj->element; + break; + case 'css_importantdecorator': + $class .= $this->getClass($obj->def, $sec_prefix); + if ($obj->allow) $class .= ', !important'; + break; + } + $class .= ')'; + return $class; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Printer/CSSDefinition.php b/lib/htmlpurifier/library/HTMLPurifier/Printer/CSSDefinition.php new file mode 100644 index 00000000..81f98659 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Printer/CSSDefinition.php @@ -0,0 +1,38 @@ +def = $config->getCSSDefinition(); + $ret = ''; + + $ret .= $this->start('div', array('class' => 'HTMLPurifier_Printer')); + $ret .= $this->start('table'); + + $ret .= $this->element('caption', 'Properties ($info)'); + + $ret .= $this->start('thead'); + $ret .= $this->start('tr'); + $ret .= $this->element('th', 'Property', array('class' => 'heavy')); + $ret .= $this->element('th', 'Definition', array('class' => 'heavy', 'style' => 'width:auto;')); + $ret .= $this->end('tr'); + $ret .= $this->end('thead'); + + ksort($this->def->info); + foreach ($this->def->info as $property => $obj) { + $name = $this->getClass($obj, 'AttrDef_'); + $ret .= $this->row($property, $name); + } + + $ret .= $this->end('table'); + $ret .= $this->end('div'); + + return $ret; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Printer/ConfigForm.css b/lib/htmlpurifier/library/HTMLPurifier/Printer/ConfigForm.css new file mode 100644 index 00000000..3ff1a88a --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Printer/ConfigForm.css @@ -0,0 +1,10 @@ + +.hp-config {} + +.hp-config tbody th {text-align:right; padding-right:0.5em;} +.hp-config thead, .hp-config .namespace {background:#3C578C; color:#FFF;} +.hp-config .namespace th {text-align:center;} +.hp-config .verbose {display:none;} +.hp-config .controls {text-align:center;} + +/* vim: et sw=4 sts=4 */ diff --git a/lib/htmlpurifier/library/HTMLPurifier/Printer/ConfigForm.js b/lib/htmlpurifier/library/HTMLPurifier/Printer/ConfigForm.js new file mode 100644 index 00000000..cba00c9b --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Printer/ConfigForm.js @@ -0,0 +1,5 @@ +function toggleWriteability(id_of_patient, checked) { + document.getElementById(id_of_patient).disabled = checked; +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Printer/ConfigForm.php b/lib/htmlpurifier/library/HTMLPurifier/Printer/ConfigForm.php new file mode 100644 index 00000000..02aa6568 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Printer/ConfigForm.php @@ -0,0 +1,368 @@ +docURL = $doc_url; + $this->name = $name; + $this->compress = $compress; + // initialize sub-printers + $this->fields[0] = new HTMLPurifier_Printer_ConfigForm_default(); + $this->fields[HTMLPurifier_VarParser::BOOL] = new HTMLPurifier_Printer_ConfigForm_bool(); + } + + /** + * Sets default column and row size for textareas in sub-printers + * @param $cols Integer columns of textarea, null to use default + * @param $rows Integer rows of textarea, null to use default + */ + public function setTextareaDimensions($cols = null, $rows = null) { + if ($cols) $this->fields['default']->cols = $cols; + if ($rows) $this->fields['default']->rows = $rows; + } + + /** + * Retrieves styling, in case it is not accessible by webserver + */ + public static function getCSS() { + return file_get_contents(HTMLPURIFIER_PREFIX . '/HTMLPurifier/Printer/ConfigForm.css'); + } + + /** + * Retrieves JavaScript, in case it is not accessible by webserver + */ + public static function getJavaScript() { + return file_get_contents(HTMLPURIFIER_PREFIX . '/HTMLPurifier/Printer/ConfigForm.js'); + } + + /** + * Returns HTML output for a configuration form + * @param $config Configuration object of current form state, or an array + * where [0] has an HTML namespace and [1] is being rendered. + * @param $allowed Optional namespace(s) and directives to restrict form to. + */ + public function render($config, $allowed = true, $render_controls = true) { + if (is_array($config) && isset($config[0])) { + $gen_config = $config[0]; + $config = $config[1]; + } else { + $gen_config = $config; + } + + $this->config = $config; + $this->genConfig = $gen_config; + $this->prepareGenerator($gen_config); + + $allowed = HTMLPurifier_Config::getAllowedDirectivesForForm($allowed, $config->def); + $all = array(); + foreach ($allowed as $key) { + list($ns, $directive) = $key; + $all[$ns][$directive] = $config->get($ns .'.'. $directive); + } + + $ret = ''; + $ret .= $this->start('table', array('class' => 'hp-config')); + $ret .= $this->start('thead'); + $ret .= $this->start('tr'); + $ret .= $this->element('th', 'Directive', array('class' => 'hp-directive')); + $ret .= $this->element('th', 'Value', array('class' => 'hp-value')); + $ret .= $this->end('tr'); + $ret .= $this->end('thead'); + foreach ($all as $ns => $directives) { + $ret .= $this->renderNamespace($ns, $directives); + } + if ($render_controls) { + $ret .= $this->start('tbody'); + $ret .= $this->start('tr'); + $ret .= $this->start('td', array('colspan' => 2, 'class' => 'controls')); + $ret .= $this->elementEmpty('input', array('type' => 'submit', 'value' => 'Submit')); + $ret .= '[Reset]'; + $ret .= $this->end('td'); + $ret .= $this->end('tr'); + $ret .= $this->end('tbody'); + } + $ret .= $this->end('table'); + return $ret; + } + + /** + * Renders a single namespace + * @param $ns String namespace name + * @param $directive Associative array of directives to values + */ + protected function renderNamespace($ns, $directives) { + $ret = ''; + $ret .= $this->start('tbody', array('class' => 'namespace')); + $ret .= $this->start('tr'); + $ret .= $this->element('th', $ns, array('colspan' => 2)); + $ret .= $this->end('tr'); + $ret .= $this->end('tbody'); + $ret .= $this->start('tbody'); + foreach ($directives as $directive => $value) { + $ret .= $this->start('tr'); + $ret .= $this->start('th'); + if ($this->docURL) { + $url = str_replace('%s', urlencode("$ns.$directive"), $this->docURL); + $ret .= $this->start('a', array('href' => $url)); + } + $attr = array('for' => "{$this->name}:$ns.$directive"); + + // crop directive name if it's too long + if (!$this->compress || (strlen($directive) < $this->compress)) { + $directive_disp = $directive; + } else { + $directive_disp = substr($directive, 0, $this->compress - 2) . '...'; + $attr['title'] = $directive; + } + + $ret .= $this->element( + 'label', + $directive_disp, + // component printers must create an element with this id + $attr + ); + if ($this->docURL) $ret .= $this->end('a'); + $ret .= $this->end('th'); + + $ret .= $this->start('td'); + $def = $this->config->def->info["$ns.$directive"]; + if (is_int($def)) { + $allow_null = $def < 0; + $type = abs($def); + } else { + $type = $def->type; + $allow_null = isset($def->allow_null); + } + if (!isset($this->fields[$type])) $type = 0; // default + $type_obj = $this->fields[$type]; + if ($allow_null) { + $type_obj = new HTMLPurifier_Printer_ConfigForm_NullDecorator($type_obj); + } + $ret .= $type_obj->render($ns, $directive, $value, $this->name, array($this->genConfig, $this->config)); + $ret .= $this->end('td'); + $ret .= $this->end('tr'); + } + $ret .= $this->end('tbody'); + return $ret; + } + +} + +/** + * Printer decorator for directives that accept null + */ +class HTMLPurifier_Printer_ConfigForm_NullDecorator extends HTMLPurifier_Printer { + /** + * Printer being decorated + */ + protected $obj; + /** + * @param $obj Printer to decorate + */ + public function __construct($obj) { + parent::__construct(); + $this->obj = $obj; + } + public function render($ns, $directive, $value, $name, $config) { + if (is_array($config) && isset($config[0])) { + $gen_config = $config[0]; + $config = $config[1]; + } else { + $gen_config = $config; + } + $this->prepareGenerator($gen_config); + + $ret = ''; + $ret .= $this->start('label', array('for' => "$name:Null_$ns.$directive")); + $ret .= $this->element('span', "$ns.$directive:", array('class' => 'verbose')); + $ret .= $this->text(' Null/Disabled'); + $ret .= $this->end('label'); + $attr = array( + 'type' => 'checkbox', + 'value' => '1', + 'class' => 'null-toggle', + 'name' => "$name"."[Null_$ns.$directive]", + 'id' => "$name:Null_$ns.$directive", + 'onclick' => "toggleWriteability('$name:$ns.$directive',checked)" // INLINE JAVASCRIPT!!!! + ); + if ($this->obj instanceof HTMLPurifier_Printer_ConfigForm_bool) { + // modify inline javascript slightly + $attr['onclick'] = "toggleWriteability('$name:Yes_$ns.$directive',checked);toggleWriteability('$name:No_$ns.$directive',checked)"; + } + if ($value === null) $attr['checked'] = 'checked'; + $ret .= $this->elementEmpty('input', $attr); + $ret .= $this->text(' or '); + $ret .= $this->elementEmpty('br'); + $ret .= $this->obj->render($ns, $directive, $value, $name, array($gen_config, $config)); + return $ret; + } +} + +/** + * Swiss-army knife configuration form field printer + */ +class HTMLPurifier_Printer_ConfigForm_default extends HTMLPurifier_Printer { + public $cols = 18; + public $rows = 5; + public function render($ns, $directive, $value, $name, $config) { + if (is_array($config) && isset($config[0])) { + $gen_config = $config[0]; + $config = $config[1]; + } else { + $gen_config = $config; + } + $this->prepareGenerator($gen_config); + // this should probably be split up a little + $ret = ''; + $def = $config->def->info["$ns.$directive"]; + if (is_int($def)) { + $type = abs($def); + } else { + $type = $def->type; + } + if (is_array($value)) { + switch ($type) { + case HTMLPurifier_VarParser::LOOKUP: + $array = $value; + $value = array(); + foreach ($array as $val => $b) { + $value[] = $val; + } + case HTMLPurifier_VarParser::ALIST: + $value = implode(PHP_EOL, $value); + break; + case HTMLPurifier_VarParser::HASH: + $nvalue = ''; + foreach ($value as $i => $v) { + $nvalue .= "$i:$v" . PHP_EOL; + } + $value = $nvalue; + break; + default: + $value = ''; + } + } + if ($type === HTMLPurifier_VarParser::MIXED) { + return 'Not supported'; + $value = serialize($value); + } + $attr = array( + 'name' => "$name"."[$ns.$directive]", + 'id' => "$name:$ns.$directive" + ); + if ($value === null) $attr['disabled'] = 'disabled'; + if (isset($def->allowed)) { + $ret .= $this->start('select', $attr); + foreach ($def->allowed as $val => $b) { + $attr = array(); + if ($value == $val) $attr['selected'] = 'selected'; + $ret .= $this->element('option', $val, $attr); + } + $ret .= $this->end('select'); + } elseif ( + $type === HTMLPurifier_VarParser::TEXT || + $type === HTMLPurifier_VarParser::ITEXT || + $type === HTMLPurifier_VarParser::ALIST || + $type === HTMLPurifier_VarParser::HASH || + $type === HTMLPurifier_VarParser::LOOKUP + ) { + $attr['cols'] = $this->cols; + $attr['rows'] = $this->rows; + $ret .= $this->start('textarea', $attr); + $ret .= $this->text($value); + $ret .= $this->end('textarea'); + } else { + $attr['value'] = $value; + $attr['type'] = 'text'; + $ret .= $this->elementEmpty('input', $attr); + } + return $ret; + } +} + +/** + * Bool form field printer + */ +class HTMLPurifier_Printer_ConfigForm_bool extends HTMLPurifier_Printer { + public function render($ns, $directive, $value, $name, $config) { + if (is_array($config) && isset($config[0])) { + $gen_config = $config[0]; + $config = $config[1]; + } else { + $gen_config = $config; + } + $this->prepareGenerator($gen_config); + $ret = ''; + $ret .= $this->start('div', array('id' => "$name:$ns.$directive")); + + $ret .= $this->start('label', array('for' => "$name:Yes_$ns.$directive")); + $ret .= $this->element('span', "$ns.$directive:", array('class' => 'verbose')); + $ret .= $this->text(' Yes'); + $ret .= $this->end('label'); + + $attr = array( + 'type' => 'radio', + 'name' => "$name"."[$ns.$directive]", + 'id' => "$name:Yes_$ns.$directive", + 'value' => '1' + ); + if ($value === true) $attr['checked'] = 'checked'; + if ($value === null) $attr['disabled'] = 'disabled'; + $ret .= $this->elementEmpty('input', $attr); + + $ret .= $this->start('label', array('for' => "$name:No_$ns.$directive")); + $ret .= $this->element('span', "$ns.$directive:", array('class' => 'verbose')); + $ret .= $this->text(' No'); + $ret .= $this->end('label'); + + $attr = array( + 'type' => 'radio', + 'name' => "$name"."[$ns.$directive]", + 'id' => "$name:No_$ns.$directive", + 'value' => '0' + ); + if ($value === false) $attr['checked'] = 'checked'; + if ($value === null) $attr['disabled'] = 'disabled'; + $ret .= $this->elementEmpty('input', $attr); + + $ret .= $this->end('div'); + + return $ret; + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Printer/HTMLDefinition.php b/lib/htmlpurifier/library/HTMLPurifier/Printer/HTMLDefinition.php new file mode 100644 index 00000000..8a8f126b --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Printer/HTMLDefinition.php @@ -0,0 +1,272 @@ +config =& $config; + + $this->def = $config->getHTMLDefinition(); + + $ret .= $this->start('div', array('class' => 'HTMLPurifier_Printer')); + + $ret .= $this->renderDoctype(); + $ret .= $this->renderEnvironment(); + $ret .= $this->renderContentSets(); + $ret .= $this->renderInfo(); + + $ret .= $this->end('div'); + + return $ret; + } + + /** + * Renders the Doctype table + */ + protected function renderDoctype() { + $doctype = $this->def->doctype; + $ret = ''; + $ret .= $this->start('table'); + $ret .= $this->element('caption', 'Doctype'); + $ret .= $this->row('Name', $doctype->name); + $ret .= $this->row('XML', $doctype->xml ? 'Yes' : 'No'); + $ret .= $this->row('Default Modules', implode($doctype->modules, ', ')); + $ret .= $this->row('Default Tidy Modules', implode($doctype->tidyModules, ', ')); + $ret .= $this->end('table'); + return $ret; + } + + + /** + * Renders environment table, which is miscellaneous info + */ + protected function renderEnvironment() { + $def = $this->def; + + $ret = ''; + + $ret .= $this->start('table'); + $ret .= $this->element('caption', 'Environment'); + + $ret .= $this->row('Parent of fragment', $def->info_parent); + $ret .= $this->renderChildren($def->info_parent_def->child); + $ret .= $this->row('Block wrap name', $def->info_block_wrapper); + + $ret .= $this->start('tr'); + $ret .= $this->element('th', 'Global attributes'); + $ret .= $this->element('td', $this->listifyAttr($def->info_global_attr),0,0); + $ret .= $this->end('tr'); + + $ret .= $this->start('tr'); + $ret .= $this->element('th', 'Tag transforms'); + $list = array(); + foreach ($def->info_tag_transform as $old => $new) { + $new = $this->getClass($new, 'TagTransform_'); + $list[] = "<$old> with $new"; + } + $ret .= $this->element('td', $this->listify($list)); + $ret .= $this->end('tr'); + + $ret .= $this->start('tr'); + $ret .= $this->element('th', 'Pre-AttrTransform'); + $ret .= $this->element('td', $this->listifyObjectList($def->info_attr_transform_pre)); + $ret .= $this->end('tr'); + + $ret .= $this->start('tr'); + $ret .= $this->element('th', 'Post-AttrTransform'); + $ret .= $this->element('td', $this->listifyObjectList($def->info_attr_transform_post)); + $ret .= $this->end('tr'); + + $ret .= $this->end('table'); + return $ret; + } + + /** + * Renders the Content Sets table + */ + protected function renderContentSets() { + $ret = ''; + $ret .= $this->start('table'); + $ret .= $this->element('caption', 'Content Sets'); + foreach ($this->def->info_content_sets as $name => $lookup) { + $ret .= $this->heavyHeader($name); + $ret .= $this->start('tr'); + $ret .= $this->element('td', $this->listifyTagLookup($lookup)); + $ret .= $this->end('tr'); + } + $ret .= $this->end('table'); + return $ret; + } + + /** + * Renders the Elements ($info) table + */ + protected function renderInfo() { + $ret = ''; + $ret .= $this->start('table'); + $ret .= $this->element('caption', 'Elements ($info)'); + ksort($this->def->info); + $ret .= $this->heavyHeader('Allowed tags', 2); + $ret .= $this->start('tr'); + $ret .= $this->element('td', $this->listifyTagLookup($this->def->info), array('colspan' => 2)); + $ret .= $this->end('tr'); + foreach ($this->def->info as $name => $def) { + $ret .= $this->start('tr'); + $ret .= $this->element('th', "<$name>", array('class'=>'heavy', 'colspan' => 2)); + $ret .= $this->end('tr'); + $ret .= $this->start('tr'); + $ret .= $this->element('th', 'Inline content'); + $ret .= $this->element('td', $def->descendants_are_inline ? 'Yes' : 'No'); + $ret .= $this->end('tr'); + if (!empty($def->excludes)) { + $ret .= $this->start('tr'); + $ret .= $this->element('th', 'Excludes'); + $ret .= $this->element('td', $this->listifyTagLookup($def->excludes)); + $ret .= $this->end('tr'); + } + if (!empty($def->attr_transform_pre)) { + $ret .= $this->start('tr'); + $ret .= $this->element('th', 'Pre-AttrTransform'); + $ret .= $this->element('td', $this->listifyObjectList($def->attr_transform_pre)); + $ret .= $this->end('tr'); + } + if (!empty($def->attr_transform_post)) { + $ret .= $this->start('tr'); + $ret .= $this->element('th', 'Post-AttrTransform'); + $ret .= $this->element('td', $this->listifyObjectList($def->attr_transform_post)); + $ret .= $this->end('tr'); + } + if (!empty($def->auto_close)) { + $ret .= $this->start('tr'); + $ret .= $this->element('th', 'Auto closed by'); + $ret .= $this->element('td', $this->listifyTagLookup($def->auto_close)); + $ret .= $this->end('tr'); + } + $ret .= $this->start('tr'); + $ret .= $this->element('th', 'Allowed attributes'); + $ret .= $this->element('td',$this->listifyAttr($def->attr), array(), 0); + $ret .= $this->end('tr'); + + if (!empty($def->required_attr)) { + $ret .= $this->row('Required attributes', $this->listify($def->required_attr)); + } + + $ret .= $this->renderChildren($def->child); + } + $ret .= $this->end('table'); + return $ret; + } + + /** + * Renders a row describing the allowed children of an element + * @param $def HTMLPurifier_ChildDef of pertinent element + */ + protected function renderChildren($def) { + $context = new HTMLPurifier_Context(); + $ret = ''; + $ret .= $this->start('tr'); + $elements = array(); + $attr = array(); + if (isset($def->elements)) { + if ($def->type == 'strictblockquote') { + $def->validateChildren(array(), $this->config, $context); + } + $elements = $def->elements; + } + if ($def->type == 'chameleon') { + $attr['rowspan'] = 2; + } elseif ($def->type == 'empty') { + $elements = array(); + } elseif ($def->type == 'table') { + $elements = array_flip(array('col', 'caption', 'colgroup', 'thead', + 'tfoot', 'tbody', 'tr')); + } + $ret .= $this->element('th', 'Allowed children', $attr); + + if ($def->type == 'chameleon') { + + $ret .= $this->element('td', + 'Block: ' . + $this->escape($this->listifyTagLookup($def->block->elements)),0,0); + $ret .= $this->end('tr'); + $ret .= $this->start('tr'); + $ret .= $this->element('td', + 'Inline: ' . + $this->escape($this->listifyTagLookup($def->inline->elements)),0,0); + + } elseif ($def->type == 'custom') { + + $ret .= $this->element('td', ''.ucfirst($def->type).': ' . + $def->dtd_regex); + + } else { + $ret .= $this->element('td', + ''.ucfirst($def->type).': ' . + $this->escape($this->listifyTagLookup($elements)),0,0); + } + $ret .= $this->end('tr'); + return $ret; + } + + /** + * Listifies a tag lookup table. + * @param $array Tag lookup array in form of array('tagname' => true) + */ + protected function listifyTagLookup($array) { + ksort($array); + $list = array(); + foreach ($array as $name => $discard) { + if ($name !== '#PCDATA' && !isset($this->def->info[$name])) continue; + $list[] = $name; + } + return $this->listify($list); + } + + /** + * Listifies a list of objects by retrieving class names and internal state + * @param $array List of objects + * @todo Also add information about internal state + */ + protected function listifyObjectList($array) { + ksort($array); + $list = array(); + foreach ($array as $discard => $obj) { + $list[] = $this->getClass($obj, 'AttrTransform_'); + } + return $this->listify($list); + } + + /** + * Listifies a hash of attributes to AttrDef classes + * @param $array Array hash in form of array('attrname' => HTMLPurifier_AttrDef) + */ + protected function listifyAttr($array) { + ksort($array); + $list = array(); + foreach ($array as $name => $obj) { + if ($obj === false) continue; + $list[] = "$name = " . $this->getClass($obj, 'AttrDef_') . ''; + } + return $this->listify($list); + } + + /** + * Creates a heavy header row + */ + protected function heavyHeader($text, $num = 1) { + $ret = ''; + $ret .= $this->start('tr'); + $ret .= $this->element('th', $text, array('colspan' => $num, 'class' => 'heavy')); + $ret .= $this->end('tr'); + return $ret; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/PropertyList.php b/lib/htmlpurifier/library/HTMLPurifier/PropertyList.php new file mode 100644 index 00000000..2b99fb7b --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/PropertyList.php @@ -0,0 +1,86 @@ +parent = $parent; + } + + /** + * Recursively retrieves the value for a key + */ + public function get($name) { + if ($this->has($name)) return $this->data[$name]; + // possible performance bottleneck, convert to iterative if necessary + if ($this->parent) return $this->parent->get($name); + throw new HTMLPurifier_Exception("Key '$name' not found"); + } + + /** + * Sets the value of a key, for this plist + */ + public function set($name, $value) { + $this->data[$name] = $value; + } + + /** + * Returns true if a given key exists + */ + public function has($name) { + return array_key_exists($name, $this->data); + } + + /** + * Resets a value to the value of it's parent, usually the default. If + * no value is specified, the entire plist is reset. + */ + public function reset($name = null) { + if ($name == null) $this->data = array(); + else unset($this->data[$name]); + } + + /** + * Squashes this property list and all of its property lists into a single + * array, and returns the array. This value is cached by default. + * @param $force If true, ignores the cache and regenerates the array. + */ + public function squash($force = false) { + if ($this->cache !== null && !$force) return $this->cache; + if ($this->parent) { + return $this->cache = array_merge($this->parent->squash($force), $this->data); + } else { + return $this->cache = $this->data; + } + } + + /** + * Returns the parent plist. + */ + public function getParent() { + return $this->parent; + } + + /** + * Sets the parent plist. + */ + public function setParent($plist) { + $this->parent = $plist; + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/PropertyListIterator.php b/lib/htmlpurifier/library/HTMLPurifier/PropertyListIterator.php new file mode 100644 index 00000000..8f250443 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/PropertyListIterator.php @@ -0,0 +1,32 @@ +l = strlen($filter); + $this->filter = $filter; + } + + public function accept() { + $key = $this->getInnerIterator()->key(); + if( strncmp($key, $this->filter, $this->l) !== 0 ) { + return false; + } + return true; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Strategy.php b/lib/htmlpurifier/library/HTMLPurifier/Strategy.php new file mode 100644 index 00000000..24628652 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Strategy.php @@ -0,0 +1,26 @@ +strategies as $strategy) { + $tokens = $strategy->execute($tokens, $config, $context); + } + return $tokens; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Strategy/Core.php b/lib/htmlpurifier/library/HTMLPurifier/Strategy/Core.php new file mode 100644 index 00000000..d90e1586 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Strategy/Core.php @@ -0,0 +1,18 @@ +strategies[] = new HTMLPurifier_Strategy_RemoveForeignElements(); + $this->strategies[] = new HTMLPurifier_Strategy_MakeWellFormed(); + $this->strategies[] = new HTMLPurifier_Strategy_FixNesting(); + $this->strategies[] = new HTMLPurifier_Strategy_ValidateAttributes(); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Strategy/FixNesting.php b/lib/htmlpurifier/library/HTMLPurifier/Strategy/FixNesting.php new file mode 100644 index 00000000..f8180239 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Strategy/FixNesting.php @@ -0,0 +1,328 @@ +getHTMLDefinition(); + + // insert implicit "parent" node, will be removed at end. + // DEFINITION CALL + $parent_name = $definition->info_parent; + array_unshift($tokens, new HTMLPurifier_Token_Start($parent_name)); + $tokens[] = new HTMLPurifier_Token_End($parent_name); + + // setup the context variable 'IsInline', for chameleon processing + // is 'false' when we are not inline, 'true' when it must always + // be inline, and an integer when it is inline for a certain + // branch of the document tree + $is_inline = $definition->info_parent_def->descendants_are_inline; + $context->register('IsInline', $is_inline); + + // setup error collector + $e =& $context->get('ErrorCollector', true); + + //####################################################################// + // Loop initialization + + // stack that contains the indexes of all parents, + // $stack[count($stack)-1] being the current parent + $stack = array(); + + // stack that contains all elements that are excluded + // it is organized by parent elements, similar to $stack, + // but it is only populated when an element with exclusions is + // processed, i.e. there won't be empty exclusions. + $exclude_stack = array(); + + // variable that contains the start token while we are processing + // nodes. This enables error reporting to do its job + $start_token = false; + $context->register('CurrentToken', $start_token); + + //####################################################################// + // Loop + + // iterate through all start nodes. Determining the start node + // is complicated so it has been omitted from the loop construct + for ($i = 0, $size = count($tokens) ; $i < $size; ) { + + //################################################################// + // Gather information on children + + // child token accumulator + $child_tokens = array(); + + // scroll to the end of this node, report number, and collect + // all children + for ($j = $i, $depth = 0; ; $j++) { + if ($tokens[$j] instanceof HTMLPurifier_Token_Start) { + $depth++; + // skip token assignment on first iteration, this is the + // token we currently are on + if ($depth == 1) continue; + } elseif ($tokens[$j] instanceof HTMLPurifier_Token_End) { + $depth--; + // skip token assignment on last iteration, this is the + // end token of the token we're currently on + if ($depth == 0) break; + } + $child_tokens[] = $tokens[$j]; + } + + // $i is index of start token + // $j is index of end token + + $start_token = $tokens[$i]; // to make token available via CurrentToken + + //################################################################// + // Gather information on parent + + // calculate parent information + if ($count = count($stack)) { + $parent_index = $stack[$count-1]; + $parent_name = $tokens[$parent_index]->name; + if ($parent_index == 0) { + $parent_def = $definition->info_parent_def; + } else { + $parent_def = $definition->info[$parent_name]; + } + } else { + // processing as if the parent were the "root" node + // unknown info, it won't be used anyway, in the future, + // we may want to enforce one element only (this is + // necessary for HTML Purifier to clean entire documents + $parent_index = $parent_name = $parent_def = null; + } + + // calculate context + if ($is_inline === false) { + // check if conditions make it inline + if (!empty($parent_def) && $parent_def->descendants_are_inline) { + $is_inline = $count - 1; + } + } else { + // check if we're out of inline + if ($count === $is_inline) { + $is_inline = false; + } + } + + //################################################################// + // Determine whether element is explicitly excluded SGML-style + + // determine whether or not element is excluded by checking all + // parent exclusions. The array should not be very large, two + // elements at most. + $excluded = false; + if (!empty($exclude_stack)) { + foreach ($exclude_stack as $lookup) { + if (isset($lookup[$tokens[$i]->name])) { + $excluded = true; + // no need to continue processing + break; + } + } + } + + //################################################################// + // Perform child validation + + if ($excluded) { + // there is an exclusion, remove the entire node + $result = false; + $excludes = array(); // not used, but good to initialize anyway + } else { + // DEFINITION CALL + if ($i === 0) { + // special processing for the first node + $def = $definition->info_parent_def; + } else { + $def = $definition->info[$tokens[$i]->name]; + + } + + if (!empty($def->child)) { + // have DTD child def validate children + $result = $def->child->validateChildren( + $child_tokens, $config, $context); + } else { + // weird, no child definition, get rid of everything + $result = false; + } + + // determine whether or not this element has any exclusions + $excludes = $def->excludes; + } + + // $result is now a bool or array + + //################################################################// + // Process result by interpreting $result + + if ($result === true || $child_tokens === $result) { + // leave the node as is + + // register start token as a parental node start + $stack[] = $i; + + // register exclusions if there are any + if (!empty($excludes)) $exclude_stack[] = $excludes; + + // move cursor to next possible start node + $i++; + + } elseif($result === false) { + // remove entire node + + if ($e) { + if ($excluded) { + $e->send(E_ERROR, 'Strategy_FixNesting: Node excluded'); + } else { + $e->send(E_ERROR, 'Strategy_FixNesting: Node removed'); + } + } + + // calculate length of inner tokens and current tokens + $length = $j - $i + 1; + + // perform removal + array_splice($tokens, $i, $length); + + // update size + $size -= $length; + + // there is no start token to register, + // current node is now the next possible start node + // unless it turns out that we need to do a double-check + + // this is a rought heuristic that covers 100% of HTML's + // cases and 99% of all other cases. A child definition + // that would be tricked by this would be something like: + // ( | a b c) where it's all or nothing. Fortunately, + // our current implementation claims that that case would + // not allow empty, even if it did + if (!$parent_def->child->allow_empty) { + // we need to do a double-check + $i = $parent_index; + array_pop($stack); + } + + // PROJECTED OPTIMIZATION: Process all children elements before + // reprocessing parent node. + + } else { + // replace node with $result + + // calculate length of inner tokens + $length = $j - $i - 1; + + if ($e) { + if (empty($result) && $length) { + $e->send(E_ERROR, 'Strategy_FixNesting: Node contents removed'); + } else { + $e->send(E_WARNING, 'Strategy_FixNesting: Node reorganized'); + } + } + + // perform replacement + array_splice($tokens, $i + 1, $length, $result); + + // update size + $size -= $length; + $size += count($result); + + // register start token as a parental node start + $stack[] = $i; + + // register exclusions if there are any + if (!empty($excludes)) $exclude_stack[] = $excludes; + + // move cursor to next possible start node + $i++; + + } + + //################################################################// + // Scroll to next start node + + // We assume, at this point, that $i is the index of the token + // that is the first possible new start point for a node. + + // Test if the token indeed is a start tag, if not, move forward + // and test again. + $size = count($tokens); + while ($i < $size and !$tokens[$i] instanceof HTMLPurifier_Token_Start) { + if ($tokens[$i] instanceof HTMLPurifier_Token_End) { + // pop a token index off the stack if we ended a node + array_pop($stack); + // pop an exclusion lookup off exclusion stack if + // we ended node and that node had exclusions + if ($i == 0 || $i == $size - 1) { + // use specialized var if it's the super-parent + $s_excludes = $definition->info_parent_def->excludes; + } else { + $s_excludes = $definition->info[$tokens[$i]->name]->excludes; + } + if ($s_excludes) { + array_pop($exclude_stack); + } + } + $i++; + } + + } + + //####################################################################// + // Post-processing + + // remove implicit parent tokens at the beginning and end + array_shift($tokens); + array_pop($tokens); + + // remove context variables + $context->destroy('IsInline'); + $context->destroy('CurrentToken'); + + //####################################################################// + // Return + + return $tokens; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Strategy/MakeWellFormed.php b/lib/htmlpurifier/library/HTMLPurifier/Strategy/MakeWellFormed.php new file mode 100644 index 00000000..c7aa1bb8 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Strategy/MakeWellFormed.php @@ -0,0 +1,532 @@ +getHTMLDefinition(); + + // local variables + $generator = new HTMLPurifier_Generator($config, $context); + $escape_invalid_tags = $config->get('Core.EscapeInvalidTags'); + // used for autoclose early abortion + $global_parent_allowed_elements = array(); + if (isset($definition->info[$definition->info_parent])) { + // may be unset under testing circumstances + $global_parent_allowed_elements = $definition->info[$definition->info_parent]->child->getAllowedElements($config); + } + $e = $context->get('ErrorCollector', true); + $t = false; // token index + $i = false; // injector index + $token = false; // the current token + $reprocess = false; // whether or not to reprocess the same token + $stack = array(); + + // member variables + $this->stack =& $stack; + $this->t =& $t; + $this->tokens =& $tokens; + $this->config = $config; + $this->context = $context; + + // context variables + $context->register('CurrentNesting', $stack); + $context->register('InputIndex', $t); + $context->register('InputTokens', $tokens); + $context->register('CurrentToken', $token); + + // -- begin INJECTOR -- + + $this->injectors = array(); + + $injectors = $config->getBatch('AutoFormat'); + $def_injectors = $definition->info_injector; + $custom_injectors = $injectors['Custom']; + unset($injectors['Custom']); // special case + foreach ($injectors as $injector => $b) { + // XXX: Fix with a legitimate lookup table of enabled filters + if (strpos($injector, '.') !== false) continue; + $injector = "HTMLPurifier_Injector_$injector"; + if (!$b) continue; + $this->injectors[] = new $injector; + } + foreach ($def_injectors as $injector) { + // assumed to be objects + $this->injectors[] = $injector; + } + foreach ($custom_injectors as $injector) { + if (!$injector) continue; + if (is_string($injector)) { + $injector = "HTMLPurifier_Injector_$injector"; + $injector = new $injector; + } + $this->injectors[] = $injector; + } + + // give the injectors references to the definition and context + // variables for performance reasons + foreach ($this->injectors as $ix => $injector) { + $error = $injector->prepare($config, $context); + if (!$error) continue; + array_splice($this->injectors, $ix, 1); // rm the injector + trigger_error("Cannot enable {$injector->name} injector because $error is not allowed", E_USER_WARNING); + } + + // -- end INJECTOR -- + + // a note on reprocessing: + // In order to reduce code duplication, whenever some code needs + // to make HTML changes in order to make things "correct", the + // new HTML gets sent through the purifier, regardless of its + // status. This means that if we add a start token, because it + // was totally necessary, we don't have to update nesting; we just + // punt ($reprocess = true; continue;) and it does that for us. + + // isset is in loop because $tokens size changes during loop exec + for ( + $t = 0; + $t == 0 || isset($tokens[$t - 1]); + // only increment if we don't need to reprocess + $reprocess ? $reprocess = false : $t++ + ) { + + // check for a rewind + if (is_int($i) && $i >= 0) { + // possibility: disable rewinding if the current token has a + // rewind set on it already. This would offer protection from + // infinite loop, but might hinder some advanced rewinding. + $rewind_to = $this->injectors[$i]->getRewind(); + if (is_int($rewind_to) && $rewind_to < $t) { + if ($rewind_to < 0) $rewind_to = 0; + while ($t > $rewind_to) { + $t--; + $prev = $tokens[$t]; + // indicate that other injectors should not process this token, + // but we need to reprocess it + unset($prev->skip[$i]); + $prev->rewind = $i; + if ($prev instanceof HTMLPurifier_Token_Start) array_pop($this->stack); + elseif ($prev instanceof HTMLPurifier_Token_End) $this->stack[] = $prev->start; + } + } + $i = false; + } + + // handle case of document end + if (!isset($tokens[$t])) { + // kill processing if stack is empty + if (empty($this->stack)) break; + + // peek + $top_nesting = array_pop($this->stack); + $this->stack[] = $top_nesting; + + // send error [TagClosedSuppress] + if ($e && !isset($top_nesting->armor['MakeWellFormed_TagClosedError'])) { + $e->send(E_NOTICE, 'Strategy_MakeWellFormed: Tag closed by document end', $top_nesting); + } + + // append, don't splice, since this is the end + $tokens[] = new HTMLPurifier_Token_End($top_nesting->name); + + // punt! + $reprocess = true; + continue; + } + + $token = $tokens[$t]; + + //echo '
'; printTokens($tokens, $t); printTokens($this->stack); + //flush(); + + // quick-check: if it's not a tag, no need to process + if (empty($token->is_tag)) { + if ($token instanceof HTMLPurifier_Token_Text) { + foreach ($this->injectors as $i => $injector) { + if (isset($token->skip[$i])) continue; + if ($token->rewind !== null && $token->rewind !== $i) continue; + $injector->handleText($token); + $this->processToken($token, $i); + $reprocess = true; + break; + } + } + // another possibility is a comment + continue; + } + + if (isset($definition->info[$token->name])) { + $type = $definition->info[$token->name]->child->type; + } else { + $type = false; // Type is unknown, treat accordingly + } + + // quick tag checks: anything that's *not* an end tag + $ok = false; + if ($type === 'empty' && $token instanceof HTMLPurifier_Token_Start) { + // claims to be a start tag but is empty + $token = new HTMLPurifier_Token_Empty($token->name, $token->attr, $token->line, $token->col, $token->armor); + $ok = true; + } elseif ($type && $type !== 'empty' && $token instanceof HTMLPurifier_Token_Empty) { + // claims to be empty but really is a start tag + $this->swap(new HTMLPurifier_Token_End($token->name)); + $this->insertBefore(new HTMLPurifier_Token_Start($token->name, $token->attr, $token->line, $token->col, $token->armor)); + // punt (since we had to modify the input stream in a non-trivial way) + $reprocess = true; + continue; + } elseif ($token instanceof HTMLPurifier_Token_Empty) { + // real empty token + $ok = true; + } elseif ($token instanceof HTMLPurifier_Token_Start) { + // start tag + + // ...unless they also have to close their parent + if (!empty($this->stack)) { + + // Performance note: you might think that it's rather + // inefficient, recalculating the autoclose information + // for every tag that a token closes (since when we + // do an autoclose, we push a new token into the + // stream and then /process/ that, before + // re-processing this token.) But this is + // necessary, because an injector can make an + // arbitrary transformations to the autoclosing + // tokens we introduce, so things may have changed + // in the meantime. Also, doing the inefficient thing is + // "easy" to reason about (for certain perverse definitions + // of "easy") + + $parent = array_pop($this->stack); + $this->stack[] = $parent; + + if (isset($definition->info[$parent->name])) { + $elements = $definition->info[$parent->name]->child->getAllowedElements($config); + $autoclose = !isset($elements[$token->name]); + } else { + $autoclose = false; + } + + if ($autoclose && $definition->info[$token->name]->wrap) { + // Check if an element can be wrapped by another + // element to make it valid in a context (for + // example,
needs a
in between) + $wrapname = $definition->info[$token->name]->wrap; + $wrapdef = $definition->info[$wrapname]; + $elements = $wrapdef->child->getAllowedElements($config); + $parent_elements = $definition->info[$parent->name]->child->getAllowedElements($config); + if (isset($elements[$token->name]) && isset($parent_elements[$wrapname])) { + $newtoken = new HTMLPurifier_Token_Start($wrapname); + $this->insertBefore($newtoken); + $reprocess = true; + continue; + } + } + + $carryover = false; + if ($autoclose && $definition->info[$parent->name]->formatting) { + $carryover = true; + } + + if ($autoclose) { + // check if this autoclose is doomed to fail + // (this rechecks $parent, which his harmless) + $autoclose_ok = isset($global_parent_allowed_elements[$token->name]); + if (!$autoclose_ok) { + foreach ($this->stack as $ancestor) { + $elements = $definition->info[$ancestor->name]->child->getAllowedElements($config); + if (isset($elements[$token->name])) { + $autoclose_ok = true; + break; + } + if ($definition->info[$token->name]->wrap) { + $wrapname = $definition->info[$token->name]->wrap; + $wrapdef = $definition->info[$wrapname]; + $wrap_elements = $wrapdef->child->getAllowedElements($config); + if (isset($wrap_elements[$token->name]) && isset($elements[$wrapname])) { + $autoclose_ok = true; + break; + } + } + } + } + if ($autoclose_ok) { + // errors need to be updated + $new_token = new HTMLPurifier_Token_End($parent->name); + $new_token->start = $parent; + if ($carryover) { + $element = clone $parent; + // [TagClosedAuto] + $element->armor['MakeWellFormed_TagClosedError'] = true; + $element->carryover = true; + $this->processToken(array($new_token, $token, $element)); + } else { + $this->insertBefore($new_token); + } + // [TagClosedSuppress] + if ($e && !isset($parent->armor['MakeWellFormed_TagClosedError'])) { + if (!$carryover) { + $e->send(E_NOTICE, 'Strategy_MakeWellFormed: Tag auto closed', $parent); + } else { + $e->send(E_NOTICE, 'Strategy_MakeWellFormed: Tag carryover', $parent); + } + } + } else { + $this->remove(); + } + $reprocess = true; + continue; + } + + } + $ok = true; + } + + if ($ok) { + foreach ($this->injectors as $i => $injector) { + if (isset($token->skip[$i])) continue; + if ($token->rewind !== null && $token->rewind !== $i) continue; + $injector->handleElement($token); + $this->processToken($token, $i); + $reprocess = true; + break; + } + if (!$reprocess) { + // ah, nothing interesting happened; do normal processing + $this->swap($token); + if ($token instanceof HTMLPurifier_Token_Start) { + $this->stack[] = $token; + } elseif ($token instanceof HTMLPurifier_Token_End) { + throw new HTMLPurifier_Exception('Improper handling of end tag in start code; possible error in MakeWellFormed'); + } + } + continue; + } + + // sanity check: we should be dealing with a closing tag + if (!$token instanceof HTMLPurifier_Token_End) { + throw new HTMLPurifier_Exception('Unaccounted for tag token in input stream, bug in HTML Purifier'); + } + + // make sure that we have something open + if (empty($this->stack)) { + if ($escape_invalid_tags) { + if ($e) $e->send(E_WARNING, 'Strategy_MakeWellFormed: Unnecessary end tag to text'); + $this->swap(new HTMLPurifier_Token_Text( + $generator->generateFromToken($token) + )); + } else { + $this->remove(); + if ($e) $e->send(E_WARNING, 'Strategy_MakeWellFormed: Unnecessary end tag removed'); + } + $reprocess = true; + continue; + } + + // first, check for the simplest case: everything closes neatly. + // Eventually, everything passes through here; if there are problems + // we modify the input stream accordingly and then punt, so that + // the tokens get processed again. + $current_parent = array_pop($this->stack); + if ($current_parent->name == $token->name) { + $token->start = $current_parent; + foreach ($this->injectors as $i => $injector) { + if (isset($token->skip[$i])) continue; + if ($token->rewind !== null && $token->rewind !== $i) continue; + $injector->handleEnd($token); + $this->processToken($token, $i); + $this->stack[] = $current_parent; + $reprocess = true; + break; + } + continue; + } + + // okay, so we're trying to close the wrong tag + + // undo the pop previous pop + $this->stack[] = $current_parent; + + // scroll back the entire nest, trying to find our tag. + // (feature could be to specify how far you'd like to go) + $size = count($this->stack); + // -2 because -1 is the last element, but we already checked that + $skipped_tags = false; + for ($j = $size - 2; $j >= 0; $j--) { + if ($this->stack[$j]->name == $token->name) { + $skipped_tags = array_slice($this->stack, $j); + break; + } + } + + // we didn't find the tag, so remove + if ($skipped_tags === false) { + if ($escape_invalid_tags) { + $this->swap(new HTMLPurifier_Token_Text( + $generator->generateFromToken($token) + )); + if ($e) $e->send(E_WARNING, 'Strategy_MakeWellFormed: Stray end tag to text'); + } else { + $this->remove(); + if ($e) $e->send(E_WARNING, 'Strategy_MakeWellFormed: Stray end tag removed'); + } + $reprocess = true; + continue; + } + + // do errors, in REVERSE $j order: a,b,c with + $c = count($skipped_tags); + if ($e) { + for ($j = $c - 1; $j > 0; $j--) { + // notice we exclude $j == 0, i.e. the current ending tag, from + // the errors... [TagClosedSuppress] + if (!isset($skipped_tags[$j]->armor['MakeWellFormed_TagClosedError'])) { + $e->send(E_NOTICE, 'Strategy_MakeWellFormed: Tag closed by element end', $skipped_tags[$j]); + } + } + } + + // insert tags, in FORWARD $j order: c,b,a with + $replace = array($token); + for ($j = 1; $j < $c; $j++) { + // ...as well as from the insertions + $new_token = new HTMLPurifier_Token_End($skipped_tags[$j]->name); + $new_token->start = $skipped_tags[$j]; + array_unshift($replace, $new_token); + if (isset($definition->info[$new_token->name]) && $definition->info[$new_token->name]->formatting) { + // [TagClosedAuto] + $element = clone $skipped_tags[$j]; + $element->carryover = true; + $element->armor['MakeWellFormed_TagClosedError'] = true; + $replace[] = $element; + } + } + $this->processToken($replace); + $reprocess = true; + continue; + } + + $context->destroy('CurrentNesting'); + $context->destroy('InputTokens'); + $context->destroy('InputIndex'); + $context->destroy('CurrentToken'); + + unset($this->injectors, $this->stack, $this->tokens, $this->t); + return $tokens; + } + + /** + * Processes arbitrary token values for complicated substitution patterns. + * In general: + * + * If $token is an array, it is a list of tokens to substitute for the + * current token. These tokens then get individually processed. If there + * is a leading integer in the list, that integer determines how many + * tokens from the stream should be removed. + * + * If $token is a regular token, it is swapped with the current token. + * + * If $token is false, the current token is deleted. + * + * If $token is an integer, that number of tokens (with the first token + * being the current one) will be deleted. + * + * @param $token Token substitution value + * @param $injector Injector that performed the substitution; default is if + * this is not an injector related operation. + */ + protected function processToken($token, $injector = -1) { + + // normalize forms of token + if (is_object($token)) $token = array(1, $token); + if (is_int($token)) $token = array($token); + if ($token === false) $token = array(1); + if (!is_array($token)) throw new HTMLPurifier_Exception('Invalid token type from injector'); + if (!is_int($token[0])) array_unshift($token, 1); + if ($token[0] === 0) throw new HTMLPurifier_Exception('Deleting zero tokens is not valid'); + + // $token is now an array with the following form: + // array(number nodes to delete, new node 1, new node 2, ...) + + $delete = array_shift($token); + $old = array_splice($this->tokens, $this->t, $delete, $token); + + if ($injector > -1) { + // determine appropriate skips + $oldskip = isset($old[0]) ? $old[0]->skip : array(); + foreach ($token as $object) { + $object->skip = $oldskip; + $object->skip[$injector] = true; + } + } + + } + + /** + * Inserts a token before the current token. Cursor now points to + * this token. You must reprocess after this. + */ + private function insertBefore($token) { + array_splice($this->tokens, $this->t, 0, array($token)); + } + + /** + * Removes current token. Cursor now points to new token occupying previously + * occupied space. You must reprocess after this. + */ + private function remove() { + array_splice($this->tokens, $this->t, 1); + } + + /** + * Swap current token with new token. Cursor points to new token (no + * change). You must reprocess after this. + */ + private function swap($token) { + $this->tokens[$this->t] = $token; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Strategy/RemoveForeignElements.php b/lib/htmlpurifier/library/HTMLPurifier/Strategy/RemoveForeignElements.php new file mode 100644 index 00000000..bccaf14d --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Strategy/RemoveForeignElements.php @@ -0,0 +1,188 @@ +getHTMLDefinition(); + $generator = new HTMLPurifier_Generator($config, $context); + $result = array(); + + $escape_invalid_tags = $config->get('Core.EscapeInvalidTags'); + $remove_invalid_img = $config->get('Core.RemoveInvalidImg'); + + // currently only used to determine if comments should be kept + $trusted = $config->get('HTML.Trusted'); + $comment_lookup = $config->get('HTML.AllowedComments'); + $comment_regexp = $config->get('HTML.AllowedCommentsRegexp'); + $check_comments = $comment_lookup !== array() || $comment_regexp !== null; + + $remove_script_contents = $config->get('Core.RemoveScriptContents'); + $hidden_elements = $config->get('Core.HiddenElements'); + + // remove script contents compatibility + if ($remove_script_contents === true) { + $hidden_elements['script'] = true; + } elseif ($remove_script_contents === false && isset($hidden_elements['script'])) { + unset($hidden_elements['script']); + } + + $attr_validator = new HTMLPurifier_AttrValidator(); + + // removes tokens until it reaches a closing tag with its value + $remove_until = false; + + // converts comments into text tokens when this is equal to a tag name + $textify_comments = false; + + $token = false; + $context->register('CurrentToken', $token); + + $e = false; + if ($config->get('Core.CollectErrors')) { + $e =& $context->get('ErrorCollector'); + } + + foreach($tokens as $token) { + if ($remove_until) { + if (empty($token->is_tag) || $token->name !== $remove_until) { + continue; + } + } + if (!empty( $token->is_tag )) { + // DEFINITION CALL + + // before any processing, try to transform the element + if ( + isset($definition->info_tag_transform[$token->name]) + ) { + $original_name = $token->name; + // there is a transformation for this tag + // DEFINITION CALL + $token = $definition-> + info_tag_transform[$token->name]-> + transform($token, $config, $context); + if ($e) $e->send(E_NOTICE, 'Strategy_RemoveForeignElements: Tag transform', $original_name); + } + + if (isset($definition->info[$token->name])) { + + // mostly everything's good, but + // we need to make sure required attributes are in order + if ( + ($token instanceof HTMLPurifier_Token_Start || $token instanceof HTMLPurifier_Token_Empty) && + $definition->info[$token->name]->required_attr && + ($token->name != 'img' || $remove_invalid_img) // ensure config option still works + ) { + $attr_validator->validateToken($token, $config, $context); + $ok = true; + foreach ($definition->info[$token->name]->required_attr as $name) { + if (!isset($token->attr[$name])) { + $ok = false; + break; + } + } + if (!$ok) { + if ($e) $e->send(E_ERROR, 'Strategy_RemoveForeignElements: Missing required attribute', $name); + continue; + } + $token->armor['ValidateAttributes'] = true; + } + + if (isset($hidden_elements[$token->name]) && $token instanceof HTMLPurifier_Token_Start) { + $textify_comments = $token->name; + } elseif ($token->name === $textify_comments && $token instanceof HTMLPurifier_Token_End) { + $textify_comments = false; + } + + } elseif ($escape_invalid_tags) { + // invalid tag, generate HTML representation and insert in + if ($e) $e->send(E_WARNING, 'Strategy_RemoveForeignElements: Foreign element to text'); + $token = new HTMLPurifier_Token_Text( + $generator->generateFromToken($token) + ); + } else { + // check if we need to destroy all of the tag's children + // CAN BE GENERICIZED + if (isset($hidden_elements[$token->name])) { + if ($token instanceof HTMLPurifier_Token_Start) { + $remove_until = $token->name; + } elseif ($token instanceof HTMLPurifier_Token_Empty) { + // do nothing: we're still looking + } else { + $remove_until = false; + } + if ($e) $e->send(E_ERROR, 'Strategy_RemoveForeignElements: Foreign meta element removed'); + } else { + if ($e) $e->send(E_ERROR, 'Strategy_RemoveForeignElements: Foreign element removed'); + } + continue; + } + } elseif ($token instanceof HTMLPurifier_Token_Comment) { + // textify comments in script tags when they are allowed + if ($textify_comments !== false) { + $data = $token->data; + $token = new HTMLPurifier_Token_Text($data); + } elseif ($trusted || $check_comments) { + // always cleanup comments + $trailing_hyphen = false; + if ($e) { + // perform check whether or not there's a trailing hyphen + if (substr($token->data, -1) == '-') { + $trailing_hyphen = true; + } + } + $token->data = rtrim($token->data, '-'); + $found_double_hyphen = false; + while (strpos($token->data, '--') !== false) { + $found_double_hyphen = true; + $token->data = str_replace('--', '-', $token->data); + } + if ($trusted || !empty($comment_lookup[trim($token->data)]) || ($comment_regexp !== NULL && preg_match($comment_regexp, trim($token->data)))) { + // OK good + if ($e) { + if ($trailing_hyphen) { + $e->send(E_NOTICE, 'Strategy_RemoveForeignElements: Trailing hyphen in comment removed'); + } + if ($found_double_hyphen) { + $e->send(E_NOTICE, 'Strategy_RemoveForeignElements: Hyphens in comment collapsed'); + } + } + } else { + if ($e) { + $e->send(E_NOTICE, 'Strategy_RemoveForeignElements: Comment removed'); + } + continue; + } + } else { + // strip comments + if ($e) $e->send(E_NOTICE, 'Strategy_RemoveForeignElements: Comment removed'); + continue; + } + } elseif ($token instanceof HTMLPurifier_Token_Text) { + } else { + continue; + } + $result[] = $token; + } + if ($remove_until && $e) { + // we removed tokens until the end, throw error + $e->send(E_ERROR, 'Strategy_RemoveForeignElements: Token removed to end', $remove_until); + } + + $context->destroy('CurrentToken'); + + return $result; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Strategy/ValidateAttributes.php b/lib/htmlpurifier/library/HTMLPurifier/Strategy/ValidateAttributes.php new file mode 100644 index 00000000..c3328a9d --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Strategy/ValidateAttributes.php @@ -0,0 +1,39 @@ +register('CurrentToken', $token); + + foreach ($tokens as $key => $token) { + + // only process tokens that have attributes, + // namely start and empty tags + if (!$token instanceof HTMLPurifier_Token_Start && !$token instanceof HTMLPurifier_Token_Empty) continue; + + // skip tokens that are armored + if (!empty($token->armor['ValidateAttributes'])) continue; + + // note that we have no facilities here for removing tokens + $validator->validateToken($token, $config, $context); + + $tokens[$key] = $token; // for PHP 4 + } + $context->destroy('CurrentToken'); + + return $tokens; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/StringHash.php b/lib/htmlpurifier/library/HTMLPurifier/StringHash.php new file mode 100644 index 00000000..62085c5c --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/StringHash.php @@ -0,0 +1,39 @@ +accessed[$index] = true; + return parent::offsetGet($index); + } + + /** + * Returns a lookup array of all array indexes that have been accessed. + * @return Array in form array($index => true). + */ + public function getAccessed() { + return $this->accessed; + } + + /** + * Resets the access array. + */ + public function resetAccessed() { + $this->accessed = array(); + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/StringHashParser.php b/lib/htmlpurifier/library/HTMLPurifier/StringHashParser.php new file mode 100644 index 00000000..f3e70c71 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/StringHashParser.php @@ -0,0 +1,110 @@ + 'DefaultKeyValue', + * 'KEY' => 'Value', + * 'KEY2' => 'Value2', + * 'MULTILINE-KEY' => "Multiline\nvalue.\n", + * ) + * + * We use this as an easy to use file-format for configuration schema + * files, but the class itself is usage agnostic. + * + * You can use ---- to forcibly terminate parsing of a single string-hash; + * this marker is used in multi string-hashes to delimit boundaries. + */ +class HTMLPurifier_StringHashParser +{ + + public $default = 'ID'; + + /** + * Parses a file that contains a single string-hash. + */ + public function parseFile($file) { + if (!file_exists($file)) return false; + $fh = fopen($file, 'r'); + if (!$fh) return false; + $ret = $this->parseHandle($fh); + fclose($fh); + return $ret; + } + + /** + * Parses a file that contains multiple string-hashes delimited by '----' + */ + public function parseMultiFile($file) { + if (!file_exists($file)) return false; + $ret = array(); + $fh = fopen($file, 'r'); + if (!$fh) return false; + while (!feof($fh)) { + $ret[] = $this->parseHandle($fh); + } + fclose($fh); + return $ret; + } + + /** + * Internal parser that acepts a file handle. + * @note While it's possible to simulate in-memory parsing by using + * custom stream wrappers, if such a use-case arises we should + * factor out the file handle into its own class. + * @param $fh File handle with pointer at start of valid string-hash + * block. + */ + protected function parseHandle($fh) { + $state = false; + $single = false; + $ret = array(); + do { + $line = fgets($fh); + if ($line === false) break; + $line = rtrim($line, "\n\r"); + if (!$state && $line === '') continue; + if ($line === '----') break; + if (strncmp('--#', $line, 3) === 0) { + // Comment + continue; + } elseif (strncmp('--', $line, 2) === 0) { + // Multiline declaration + $state = trim($line, '- '); + if (!isset($ret[$state])) $ret[$state] = ''; + continue; + } elseif (!$state) { + $single = true; + if (strpos($line, ':') !== false) { + // Single-line declaration + list($state, $line) = explode(':', $line, 2); + $line = trim($line); + } else { + // Use default declaration + $state = $this->default; + } + } + if ($single) { + $ret[$state] = $line; + $single = false; + $state = false; + } else { + $ret[$state] .= "$line\n"; + } + } while (!feof($fh)); + return $ret; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/TagTransform.php b/lib/htmlpurifier/library/HTMLPurifier/TagTransform.php new file mode 100644 index 00000000..210a4472 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/TagTransform.php @@ -0,0 +1,36 @@ + 'xx-small', + '1' => 'xx-small', + '2' => 'small', + '3' => 'medium', + '4' => 'large', + '5' => 'x-large', + '6' => 'xx-large', + '7' => '300%', + '-1' => 'smaller', + '-2' => '60%', + '+1' => 'larger', + '+2' => '150%', + '+3' => '200%', + '+4' => '300%' + ); + + public function transform($tag, $config, $context) { + + if ($tag instanceof HTMLPurifier_Token_End) { + $new_tag = clone $tag; + $new_tag->name = $this->transform_to; + return $new_tag; + } + + $attr = $tag->attr; + $prepend_style = ''; + + // handle color transform + if (isset($attr['color'])) { + $prepend_style .= 'color:' . $attr['color'] . ';'; + unset($attr['color']); + } + + // handle face transform + if (isset($attr['face'])) { + $prepend_style .= 'font-family:' . $attr['face'] . ';'; + unset($attr['face']); + } + + // handle size transform + if (isset($attr['size'])) { + // normalize large numbers + if ($attr['size'] !== '') { + if ($attr['size']{0} == '+' || $attr['size']{0} == '-') { + $size = (int) $attr['size']; + if ($size < -2) $attr['size'] = '-2'; + if ($size > 4) $attr['size'] = '+4'; + } else { + $size = (int) $attr['size']; + if ($size > 7) $attr['size'] = '7'; + } + } + if (isset($this->_size_lookup[$attr['size']])) { + $prepend_style .= 'font-size:' . + $this->_size_lookup[$attr['size']] . ';'; + } + unset($attr['size']); + } + + if ($prepend_style) { + $attr['style'] = isset($attr['style']) ? + $prepend_style . $attr['style'] : + $prepend_style; + } + + $new_tag = clone $tag; + $new_tag->name = $this->transform_to; + $new_tag->attr = $attr; + + return $new_tag; + + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/TagTransform/Simple.php b/lib/htmlpurifier/library/HTMLPurifier/TagTransform/Simple.php new file mode 100644 index 00000000..0e36130f --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/TagTransform/Simple.php @@ -0,0 +1,35 @@ +transform_to = $transform_to; + $this->style = $style; + } + + public function transform($tag, $config, $context) { + $new_tag = clone $tag; + $new_tag->name = $this->transform_to; + if (!is_null($this->style) && + ($new_tag instanceof HTMLPurifier_Token_Start || $new_tag instanceof HTMLPurifier_Token_Empty) + ) { + $this->prependCSS($new_tag->attr, $this->style); + } + return $new_tag; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Token.php b/lib/htmlpurifier/library/HTMLPurifier/Token.php new file mode 100644 index 00000000..7900e6cb --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Token.php @@ -0,0 +1,57 @@ +line = $l; + $this->col = $c; + } + + /** + * Convenience function for DirectLex settings line/col position. + */ + public function rawPosition($l, $c) { + if ($c === -1) $l++; + $this->line = $l; + $this->col = $c; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Token/Comment.php b/lib/htmlpurifier/library/HTMLPurifier/Token/Comment.php new file mode 100644 index 00000000..dc6bdcab --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Token/Comment.php @@ -0,0 +1,22 @@ +data = $data; + $this->line = $line; + $this->col = $col; + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Token/Empty.php b/lib/htmlpurifier/library/HTMLPurifier/Token/Empty.php new file mode 100644 index 00000000..2a82b47a --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Token/Empty.php @@ -0,0 +1,11 @@ +!empty($obj->is_tag) + * without having to use a function call is_a(). + */ + public $is_tag = true; + + /** + * The lower-case name of the tag, like 'a', 'b' or 'blockquote'. + * + * @note Strictly speaking, XML tags are case sensitive, so we shouldn't + * be lower-casing them, but these tokens cater to HTML tags, which are + * insensitive. + */ + public $name; + + /** + * Associative array of the tag's attributes. + */ + public $attr = array(); + + /** + * Non-overloaded constructor, which lower-cases passed tag name. + * + * @param $name String name. + * @param $attr Associative array of attributes. + */ + public function __construct($name, $attr = array(), $line = null, $col = null, $armor = array()) { + $this->name = ctype_lower($name) ? $name : strtolower($name); + foreach ($attr as $key => $value) { + // normalization only necessary when key is not lowercase + if (!ctype_lower($key)) { + $new_key = strtolower($key); + if (!isset($attr[$new_key])) { + $attr[$new_key] = $attr[$key]; + } + if ($new_key !== $key) { + unset($attr[$key]); + } + } + } + $this->attr = $attr; + $this->line = $line; + $this->col = $col; + $this->armor = $armor; + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Token/Text.php b/lib/htmlpurifier/library/HTMLPurifier/Token/Text.php new file mode 100644 index 00000000..82efd823 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Token/Text.php @@ -0,0 +1,33 @@ +data = $data; + $this->is_whitespace = ctype_space($data); + $this->line = $line; + $this->col = $col; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/TokenFactory.php b/lib/htmlpurifier/library/HTMLPurifier/TokenFactory.php new file mode 100644 index 00000000..7cf48fb4 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/TokenFactory.php @@ -0,0 +1,94 @@ +p_start = new HTMLPurifier_Token_Start('', array()); + $this->p_end = new HTMLPurifier_Token_End(''); + $this->p_empty = new HTMLPurifier_Token_Empty('', array()); + $this->p_text = new HTMLPurifier_Token_Text(''); + $this->p_comment= new HTMLPurifier_Token_Comment(''); + } + + /** + * Creates a HTMLPurifier_Token_Start. + * @param $name Tag name + * @param $attr Associative array of attributes + * @return Generated HTMLPurifier_Token_Start + */ + public function createStart($name, $attr = array()) { + $p = clone $this->p_start; + $p->__construct($name, $attr); + return $p; + } + + /** + * Creates a HTMLPurifier_Token_End. + * @param $name Tag name + * @return Generated HTMLPurifier_Token_End + */ + public function createEnd($name) { + $p = clone $this->p_end; + $p->__construct($name); + return $p; + } + + /** + * Creates a HTMLPurifier_Token_Empty. + * @param $name Tag name + * @param $attr Associative array of attributes + * @return Generated HTMLPurifier_Token_Empty + */ + public function createEmpty($name, $attr = array()) { + $p = clone $this->p_empty; + $p->__construct($name, $attr); + return $p; + } + + /** + * Creates a HTMLPurifier_Token_Text. + * @param $data Data of text token + * @return Generated HTMLPurifier_Token_Text + */ + public function createText($data) { + $p = clone $this->p_text; + $p->__construct($data); + return $p; + } + + /** + * Creates a HTMLPurifier_Token_Comment. + * @param $data Data of comment token + * @return Generated HTMLPurifier_Token_Comment + */ + public function createComment($data) { + $p = clone $this->p_comment; + $p->__construct($data); + return $p; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/URI.php b/lib/htmlpurifier/library/HTMLPurifier/URI.php new file mode 100644 index 00000000..f158ef5e --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URI.php @@ -0,0 +1,242 @@ +scheme = is_null($scheme) || ctype_lower($scheme) ? $scheme : strtolower($scheme); + $this->userinfo = $userinfo; + $this->host = $host; + $this->port = is_null($port) ? $port : (int) $port; + $this->path = $path; + $this->query = $query; + $this->fragment = $fragment; + } + + /** + * Retrieves a scheme object corresponding to the URI's scheme/default + * @param $config Instance of HTMLPurifier_Config + * @param $context Instance of HTMLPurifier_Context + * @return Scheme object appropriate for validating this URI + */ + public function getSchemeObj($config, $context) { + $registry = HTMLPurifier_URISchemeRegistry::instance(); + if ($this->scheme !== null) { + $scheme_obj = $registry->getScheme($this->scheme, $config, $context); + if (!$scheme_obj) return false; // invalid scheme, clean it out + } else { + // no scheme: retrieve the default one + $def = $config->getDefinition('URI'); + $scheme_obj = $def->getDefaultScheme($config, $context); + if (!$scheme_obj) { + // something funky happened to the default scheme object + trigger_error( + 'Default scheme object "' . $def->defaultScheme . '" was not readable', + E_USER_WARNING + ); + return false; + } + } + return $scheme_obj; + } + + /** + * Generic validation method applicable for all schemes. May modify + * this URI in order to get it into a compliant form. + * @param $config Instance of HTMLPurifier_Config + * @param $context Instance of HTMLPurifier_Context + * @return True if validation/filtering succeeds, false if failure + */ + public function validate($config, $context) { + + // ABNF definitions from RFC 3986 + $chars_sub_delims = '!$&\'()*+,;='; + $chars_gen_delims = ':/?#[]@'; + $chars_pchar = $chars_sub_delims . ':@'; + + // validate host + if (!is_null($this->host)) { + $host_def = new HTMLPurifier_AttrDef_URI_Host(); + $this->host = $host_def->validate($this->host, $config, $context); + if ($this->host === false) $this->host = null; + } + + // validate scheme + // NOTE: It's not appropriate to check whether or not this + // scheme is in our registry, since a URIFilter may convert a + // URI that we don't allow into one we do. So instead, we just + // check if the scheme can be dropped because there is no host + // and it is our default scheme. + if (!is_null($this->scheme) && is_null($this->host) || $this->host === '') { + // support for relative paths is pretty abysmal when the + // scheme is present, so axe it when possible + $def = $config->getDefinition('URI'); + if ($def->defaultScheme === $this->scheme) { + $this->scheme = null; + } + } + + // validate username + if (!is_null($this->userinfo)) { + $encoder = new HTMLPurifier_PercentEncoder($chars_sub_delims . ':'); + $this->userinfo = $encoder->encode($this->userinfo); + } + + // validate port + if (!is_null($this->port)) { + if ($this->port < 1 || $this->port > 65535) $this->port = null; + } + + // validate path + $path_parts = array(); + $segments_encoder = new HTMLPurifier_PercentEncoder($chars_pchar . '/'); + if (!is_null($this->host)) { // this catches $this->host === '' + // path-abempty (hier and relative) + // http://www.example.com/my/path + // //www.example.com/my/path (looks odd, but works, and + // recognized by most browsers) + // (this set is valid or invalid on a scheme by scheme + // basis, so we'll deal with it later) + // file:///my/path + // ///my/path + $this->path = $segments_encoder->encode($this->path); + } elseif ($this->path !== '') { + if ($this->path[0] === '/') { + // path-absolute (hier and relative) + // http:/my/path + // /my/path + if (strlen($this->path) >= 2 && $this->path[1] === '/') { + // This could happen if both the host gets stripped + // out + // http://my/path + // //my/path + $this->path = ''; + } else { + $this->path = $segments_encoder->encode($this->path); + } + } elseif (!is_null($this->scheme)) { + // path-rootless (hier) + // http:my/path + // Short circuit evaluation means we don't need to check nz + $this->path = $segments_encoder->encode($this->path); + } else { + // path-noscheme (relative) + // my/path + // (once again, not checking nz) + $segment_nc_encoder = new HTMLPurifier_PercentEncoder($chars_sub_delims . '@'); + $c = strpos($this->path, '/'); + if ($c !== false) { + $this->path = + $segment_nc_encoder->encode(substr($this->path, 0, $c)) . + $segments_encoder->encode(substr($this->path, $c)); + } else { + $this->path = $segment_nc_encoder->encode($this->path); + } + } + } else { + // path-empty (hier and relative) + $this->path = ''; // just to be safe + } + + // qf = query and fragment + $qf_encoder = new HTMLPurifier_PercentEncoder($chars_pchar . '/?'); + + if (!is_null($this->query)) { + $this->query = $qf_encoder->encode($this->query); + } + + if (!is_null($this->fragment)) { + $this->fragment = $qf_encoder->encode($this->fragment); + } + + return true; + + } + + /** + * Convert URI back to string + * @return String URI appropriate for output + */ + public function toString() { + // reconstruct authority + $authority = null; + // there is a rendering difference between a null authority + // (http:foo-bar) and an empty string authority + // (http:///foo-bar). + if (!is_null($this->host)) { + $authority = ''; + if(!is_null($this->userinfo)) $authority .= $this->userinfo . '@'; + $authority .= $this->host; + if(!is_null($this->port)) $authority .= ':' . $this->port; + } + + // Reconstruct the result + // One might wonder about parsing quirks from browsers after + // this reconstruction. Unfortunately, parsing behavior depends + // on what *scheme* was employed (file:///foo is handled *very* + // differently than http:///foo), so unfortunately we have to + // defer to the schemes to do the right thing. + $result = ''; + if (!is_null($this->scheme)) $result .= $this->scheme . ':'; + if (!is_null($authority)) $result .= '//' . $authority; + $result .= $this->path; + if (!is_null($this->query)) $result .= '?' . $this->query; + if (!is_null($this->fragment)) $result .= '#' . $this->fragment; + + return $result; + } + + /** + * Returns true if this URL might be considered a 'local' URL given + * the current context. This is true when the host is null, or + * when it matches the host supplied to the configuration. + * + * Note that this does not do any scheme checking, so it is mostly + * only appropriate for metadata that doesn't care about protocol + * security. isBenign is probably what you actually want. + */ + public function isLocal($config, $context) { + if ($this->host === null) return true; + $uri_def = $config->getDefinition('URI'); + if ($uri_def->host === $this->host) return true; + return false; + } + + /** + * Returns true if this URL should be considered a 'benign' URL, + * that is: + * + * - It is a local URL (isLocal), and + * - It has a equal or better level of security + */ + public function isBenign($config, $context) { + if (!$this->isLocal($config, $context)) return false; + + $scheme_obj = $this->getSchemeObj($config, $context); + if (!$scheme_obj) return false; // conservative approach + + $current_scheme_obj = $config->getDefinition('URI')->getDefaultScheme($config, $context); + if ($current_scheme_obj->secure) { + if (!$scheme_obj->secure) { + return false; + } + } + return true; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/URIDefinition.php b/lib/htmlpurifier/library/HTMLPurifier/URIDefinition.php new file mode 100644 index 00000000..40e57bb7 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URIDefinition.php @@ -0,0 +1,102 @@ +registerFilter(new HTMLPurifier_URIFilter_DisableExternal()); + $this->registerFilter(new HTMLPurifier_URIFilter_DisableExternalResources()); + $this->registerFilter(new HTMLPurifier_URIFilter_HostBlacklist()); + $this->registerFilter(new HTMLPurifier_URIFilter_SafeIframe()); + $this->registerFilter(new HTMLPurifier_URIFilter_MakeAbsolute()); + $this->registerFilter(new HTMLPurifier_URIFilter_Munge()); + } + + public function registerFilter($filter) { + $this->registeredFilters[$filter->name] = $filter; + } + + public function addFilter($filter, $config) { + $r = $filter->prepare($config); + if ($r === false) return; // null is ok, for backwards compat + if ($filter->post) { + $this->postFilters[$filter->name] = $filter; + } else { + $this->filters[$filter->name] = $filter; + } + } + + protected function doSetup($config) { + $this->setupMemberVariables($config); + $this->setupFilters($config); + } + + protected function setupFilters($config) { + foreach ($this->registeredFilters as $name => $filter) { + if ($filter->always_load) { + $this->addFilter($filter, $config); + } else { + $conf = $config->get('URI.' . $name); + if ($conf !== false && $conf !== null) { + $this->addFilter($filter, $config); + } + } + } + unset($this->registeredFilters); + } + + protected function setupMemberVariables($config) { + $this->host = $config->get('URI.Host'); + $base_uri = $config->get('URI.Base'); + if (!is_null($base_uri)) { + $parser = new HTMLPurifier_URIParser(); + $this->base = $parser->parse($base_uri); + $this->defaultScheme = $this->base->scheme; + if (is_null($this->host)) $this->host = $this->base->host; + } + if (is_null($this->defaultScheme)) $this->defaultScheme = $config->get('URI.DefaultScheme'); + } + + public function getDefaultScheme($config, $context) { + return HTMLPurifier_URISchemeRegistry::instance()->getScheme($this->defaultScheme, $config, $context); + } + + public function filter(&$uri, $config, $context) { + foreach ($this->filters as $name => $f) { + $result = $f->filter($uri, $config, $context); + if (!$result) return false; + } + return true; + } + + public function postFilter(&$uri, $config, $context) { + foreach ($this->postFilters as $name => $f) { + $result = $f->filter($uri, $config, $context); + if (!$result) return false; + } + return true; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/URIFilter.php b/lib/htmlpurifier/library/HTMLPurifier/URIFilter.php new file mode 100644 index 00000000..6a1b0b08 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URIFilter.php @@ -0,0 +1,67 @@ +getDefinition('URI')->host; + if ($our_host !== null) $this->ourHostParts = array_reverse(explode('.', $our_host)); + } + public function filter(&$uri, $config, $context) { + if (is_null($uri->host)) return true; + if ($this->ourHostParts === false) return false; + $host_parts = array_reverse(explode('.', $uri->host)); + foreach ($this->ourHostParts as $i => $x) { + if (!isset($host_parts[$i])) return false; + if ($host_parts[$i] != $this->ourHostParts[$i]) return false; + } + return true; + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/URIFilter/DisableExternalResources.php b/lib/htmlpurifier/library/HTMLPurifier/URIFilter/DisableExternalResources.php new file mode 100644 index 00000000..881abc43 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URIFilter/DisableExternalResources.php @@ -0,0 +1,12 @@ +get('EmbeddedURI', true)) return true; + return parent::filter($uri, $config, $context); + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/URIFilter/DisableResources.php b/lib/htmlpurifier/library/HTMLPurifier/URIFilter/DisableResources.php new file mode 100644 index 00000000..67538c7b --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URIFilter/DisableResources.php @@ -0,0 +1,11 @@ +get('EmbeddedURI', true); + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/URIFilter/HostBlacklist.php b/lib/htmlpurifier/library/HTMLPurifier/URIFilter/HostBlacklist.php new file mode 100644 index 00000000..55fde3bf --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URIFilter/HostBlacklist.php @@ -0,0 +1,25 @@ +blacklist = $config->get('URI.HostBlacklist'); + return true; + } + public function filter(&$uri, $config, $context) { + foreach($this->blacklist as $blacklisted_host_fragment) { + if (strpos($uri->host, $blacklisted_host_fragment) !== false) { + return false; + } + } + return true; + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/URIFilter/MakeAbsolute.php b/lib/htmlpurifier/library/HTMLPurifier/URIFilter/MakeAbsolute.php new file mode 100644 index 00000000..f46ab263 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URIFilter/MakeAbsolute.php @@ -0,0 +1,114 @@ +getDefinition('URI'); + $this->base = $def->base; + if (is_null($this->base)) { + trigger_error('URI.MakeAbsolute is being ignored due to lack of value for URI.Base configuration', E_USER_WARNING); + return false; + } + $this->base->fragment = null; // fragment is invalid for base URI + $stack = explode('/', $this->base->path); + array_pop($stack); // discard last segment + $stack = $this->_collapseStack($stack); // do pre-parsing + $this->basePathStack = $stack; + return true; + } + public function filter(&$uri, $config, $context) { + if (is_null($this->base)) return true; // abort early + if ( + $uri->path === '' && is_null($uri->scheme) && + is_null($uri->host) && is_null($uri->query) && is_null($uri->fragment) + ) { + // reference to current document + $uri = clone $this->base; + return true; + } + if (!is_null($uri->scheme)) { + // absolute URI already: don't change + if (!is_null($uri->host)) return true; + $scheme_obj = $uri->getSchemeObj($config, $context); + if (!$scheme_obj) { + // scheme not recognized + return false; + } + if (!$scheme_obj->hierarchical) { + // non-hierarchal URI with explicit scheme, don't change + return true; + } + // special case: had a scheme but always is hierarchical and had no authority + } + if (!is_null($uri->host)) { + // network path, don't bother + return true; + } + if ($uri->path === '') { + $uri->path = $this->base->path; + } elseif ($uri->path[0] !== '/') { + // relative path, needs more complicated processing + $stack = explode('/', $uri->path); + $new_stack = array_merge($this->basePathStack, $stack); + if ($new_stack[0] !== '' && !is_null($this->base->host)) { + array_unshift($new_stack, ''); + } + $new_stack = $this->_collapseStack($new_stack); + $uri->path = implode('/', $new_stack); + } else { + // absolute path, but still we should collapse + $uri->path = implode('/', $this->_collapseStack(explode('/', $uri->path))); + } + // re-combine + $uri->scheme = $this->base->scheme; + if (is_null($uri->userinfo)) $uri->userinfo = $this->base->userinfo; + if (is_null($uri->host)) $uri->host = $this->base->host; + if (is_null($uri->port)) $uri->port = $this->base->port; + return true; + } + + /** + * Resolve dots and double-dots in a path stack + */ + private function _collapseStack($stack) { + $result = array(); + $is_folder = false; + for ($i = 0; isset($stack[$i]); $i++) { + $is_folder = false; + // absorb an internally duplicated slash + if ($stack[$i] == '' && $i && isset($stack[$i+1])) continue; + if ($stack[$i] == '..') { + if (!empty($result)) { + $segment = array_pop($result); + if ($segment === '' && empty($result)) { + // error case: attempted to back out too far: + // restore the leading slash + $result[] = ''; + } elseif ($segment === '..') { + $result[] = '..'; // cannot remove .. with .. + } + } else { + // relative path, preserve the double-dots + $result[] = '..'; + } + $is_folder = true; + continue; + } + if ($stack[$i] == '.') { + // silently absorb + $is_folder = true; + continue; + } + $result[] = $stack[$i]; + } + if ($is_folder) $result[] = ''; + return $result; + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/URIFilter/Munge.php b/lib/htmlpurifier/library/HTMLPurifier/URIFilter/Munge.php new file mode 100644 index 00000000..de695df1 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URIFilter/Munge.php @@ -0,0 +1,53 @@ +target = $config->get('URI.' . $this->name); + $this->parser = new HTMLPurifier_URIParser(); + $this->doEmbed = $config->get('URI.MungeResources'); + $this->secretKey = $config->get('URI.MungeSecretKey'); + return true; + } + public function filter(&$uri, $config, $context) { + if ($context->get('EmbeddedURI', true) && !$this->doEmbed) return true; + + $scheme_obj = $uri->getSchemeObj($config, $context); + if (!$scheme_obj) return true; // ignore unknown schemes, maybe another postfilter did it + if (!$scheme_obj->browsable) return true; // ignore non-browseable schemes, since we can't munge those in a reasonable way + if ($uri->isBenign($config, $context)) return true; // don't redirect if a benign URL + + $this->makeReplace($uri, $config, $context); + $this->replace = array_map('rawurlencode', $this->replace); + + $new_uri = strtr($this->target, $this->replace); + $new_uri = $this->parser->parse($new_uri); + // don't redirect if the target host is the same as the + // starting host + if ($uri->host === $new_uri->host) return true; + $uri = $new_uri; // overwrite + return true; + } + + protected function makeReplace($uri, $config, $context) { + $string = $uri->toString(); + // always available + $this->replace['%s'] = $string; + $this->replace['%r'] = $context->get('EmbeddedURI', true); + $token = $context->get('CurrentToken', true); + $this->replace['%n'] = $token ? $token->name : null; + $this->replace['%m'] = $context->get('CurrentAttr', true); + $this->replace['%p'] = $context->get('CurrentCSSProperty', true); + // not always available + if ($this->secretKey) $this->replace['%t'] = sha1($this->secretKey . ':' . $string); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/URIFilter/SafeIframe.php b/lib/htmlpurifier/library/HTMLPurifier/URIFilter/SafeIframe.php new file mode 100644 index 00000000..284bb13d --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URIFilter/SafeIframe.php @@ -0,0 +1,35 @@ +regexp = $config->get('URI.SafeIframeRegexp'); + return true; + } + public function filter(&$uri, $config, $context) { + // check if filter not applicable + if (!$config->get('HTML.SafeIframe')) return true; + // check if the filter should actually trigger + if (!$context->get('EmbeddedURI', true)) return true; + $token = $context->get('CurrentToken', true); + if (!($token && $token->name == 'iframe')) return true; + // check if we actually have some whitelists enabled + if ($this->regexp === null) return false; + // actually check the whitelists + return preg_match($this->regexp, $uri->toString()); + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/URIParser.php b/lib/htmlpurifier/library/HTMLPurifier/URIParser.php new file mode 100644 index 00000000..7179e4ab --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URIParser.php @@ -0,0 +1,70 @@ +percentEncoder = new HTMLPurifier_PercentEncoder(); + } + + /** + * Parses a URI. + * @param $uri string URI to parse + * @return HTMLPurifier_URI representation of URI. This representation has + * not been validated yet and may not conform to RFC. + */ + public function parse($uri) { + + $uri = $this->percentEncoder->normalize($uri); + + // Regexp is as per Appendix B. + // Note that ["<>] are an addition to the RFC's recommended + // characters, because they represent external delimeters. + $r_URI = '!'. + '(([^:/?#"<>]+):)?'. // 2. Scheme + '(//([^/?#"<>]*))?'. // 4. Authority + '([^?#"<>]*)'. // 5. Path + '(\?([^#"<>]*))?'. // 7. Query + '(#([^"<>]*))?'. // 8. Fragment + '!'; + + $matches = array(); + $result = preg_match($r_URI, $uri, $matches); + + if (!$result) return false; // *really* invalid URI + + // seperate out parts + $scheme = !empty($matches[1]) ? $matches[2] : null; + $authority = !empty($matches[3]) ? $matches[4] : null; + $path = $matches[5]; // always present, can be empty + $query = !empty($matches[6]) ? $matches[7] : null; + $fragment = !empty($matches[8]) ? $matches[9] : null; + + // further parse authority + if ($authority !== null) { + $r_authority = "/^((.+?)@)?(\[[^\]]+\]|[^:]*)(:(\d*))?/"; + $matches = array(); + preg_match($r_authority, $authority, $matches); + $userinfo = !empty($matches[1]) ? $matches[2] : null; + $host = !empty($matches[3]) ? $matches[3] : ''; + $port = !empty($matches[4]) ? (int) $matches[5] : null; + } else { + $port = $host = $userinfo = null; + } + + return new HTMLPurifier_URI( + $scheme, $userinfo, $host, $port, $path, $query, $fragment); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/URIScheme.php b/lib/htmlpurifier/library/HTMLPurifier/URIScheme.php new file mode 100644 index 00000000..7be95814 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URIScheme.php @@ -0,0 +1,95 @@ +, resolves edge cases + * with making relative URIs absolute + */ + public $hierarchical = false; + + /** + * Whether or not the URI may omit a hostname when the scheme is + * explicitly specified, ala file:///path/to/file. As of writing, + * 'file' is the only scheme that browsers support his properly. + */ + public $may_omit_host = false; + + /** + * Validates the components of a URI for a specific scheme. + * @param $uri Reference to a HTMLPurifier_URI object + * @param $config HTMLPurifier_Config object + * @param $context HTMLPurifier_Context object + * @return Bool success or failure + */ + public abstract function doValidate(&$uri, $config, $context); + + /** + * Public interface for validating components of a URI. Performs a + * bunch of default actions. Don't overload this method. + * @param $uri Reference to a HTMLPurifier_URI object + * @param $config HTMLPurifier_Config object + * @param $context HTMLPurifier_Context object + * @return Bool success or failure + */ + public function validate(&$uri, $config, $context) { + if ($this->default_port == $uri->port) $uri->port = null; + // kludge: browsers do funny things when the scheme but not the + // authority is set + if (!$this->may_omit_host && + // if the scheme is present, a missing host is always in error + (!is_null($uri->scheme) && ($uri->host === '' || is_null($uri->host))) || + // if the scheme is not present, a *blank* host is in error, + // since this translates into '///path' which most browsers + // interpret as being 'http://path'. + (is_null($uri->scheme) && $uri->host === '') + ) { + do { + if (is_null($uri->scheme)) { + if (substr($uri->path, 0, 2) != '//') { + $uri->host = null; + break; + } + // URI is '////path', so we cannot nullify the + // host to preserve semantics. Try expanding the + // hostname instead (fall through) + } + // first see if we can manually insert a hostname + $host = $config->get('URI.Host'); + if (!is_null($host)) { + $uri->host = $host; + } else { + // we can't do anything sensible, reject the URL. + return false; + } + } while (false); + } + return $this->doValidate($uri, $config, $context); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/URIScheme/data.php b/lib/htmlpurifier/library/HTMLPurifier/URIScheme/data.php new file mode 100644 index 00000000..a5c43989 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URIScheme/data.php @@ -0,0 +1,96 @@ + true, + 'image/gif' => true, + 'image/png' => true, + ); + // this is actually irrelevant since we only write out the path + // component + public $may_omit_host = true; + + public function doValidate(&$uri, $config, $context) { + $result = explode(',', $uri->path, 2); + $is_base64 = false; + $charset = null; + $content_type = null; + if (count($result) == 2) { + list($metadata, $data) = $result; + // do some legwork on the metadata + $metas = explode(';', $metadata); + while(!empty($metas)) { + $cur = array_shift($metas); + if ($cur == 'base64') { + $is_base64 = true; + break; + } + if (substr($cur, 0, 8) == 'charset=') { + // doesn't match if there are arbitrary spaces, but + // whatever dude + if ($charset !== null) continue; // garbage + $charset = substr($cur, 8); // not used + } else { + if ($content_type !== null) continue; // garbage + $content_type = $cur; + } + } + } else { + $data = $result[0]; + } + if ($content_type !== null && empty($this->allowed_types[$content_type])) { + return false; + } + if ($charset !== null) { + // error; we don't allow plaintext stuff + $charset = null; + } + $data = rawurldecode($data); + if ($is_base64) { + $raw_data = base64_decode($data); + } else { + $raw_data = $data; + } + // XXX probably want to refactor this into a general mechanism + // for filtering arbitrary content types + $file = tempnam("/tmp", ""); + file_put_contents($file, $raw_data); + if (function_exists('exif_imagetype')) { + $image_code = exif_imagetype($file); + } elseif (function_exists('getimagesize')) { + set_error_handler(array($this, 'muteErrorHandler')); + $info = getimagesize($file); + restore_error_handler(); + if ($info == false) return false; + $image_code = $info[2]; + } else { + trigger_error("could not find exif_imagetype or getimagesize functions", E_USER_ERROR); + } + $real_content_type = image_type_to_mime_type($image_code); + if ($real_content_type != $content_type) { + // we're nice guys; if the content type is something else we + // support, change it over + if (empty($this->allowed_types[$real_content_type])) return false; + $content_type = $real_content_type; + } + // ok, it's kosher, rewrite what we need + $uri->userinfo = null; + $uri->host = null; + $uri->port = null; + $uri->fragment = null; + $uri->query = null; + $uri->path = "$content_type;base64," . base64_encode($raw_data); + return true; + } + + public function muteErrorHandler($errno, $errstr) {} + +} + diff --git a/lib/htmlpurifier/library/HTMLPurifier/URIScheme/file.php b/lib/htmlpurifier/library/HTMLPurifier/URIScheme/file.php new file mode 100644 index 00000000..d74a3f19 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URIScheme/file.php @@ -0,0 +1,32 @@ +userinfo = null; + // file:// makes no provisions for accessing the resource + $uri->port = null; + // While it seems to work on Firefox, the querystring has + // no possible effect and is thus stripped. + $uri->query = null; + return true; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/URIScheme/ftp.php b/lib/htmlpurifier/library/HTMLPurifier/URIScheme/ftp.php new file mode 100644 index 00000000..0fb2abf6 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URIScheme/ftp.php @@ -0,0 +1,42 @@ +query = null; + + // typecode check + $semicolon_pos = strrpos($uri->path, ';'); // reverse + if ($semicolon_pos !== false) { + $type = substr($uri->path, $semicolon_pos + 1); // no semicolon + $uri->path = substr($uri->path, 0, $semicolon_pos); + $type_ret = ''; + if (strpos($type, '=') !== false) { + // figure out whether or not the declaration is correct + list($key, $typecode) = explode('=', $type, 2); + if ($key !== 'type') { + // invalid key, tack it back on encoded + $uri->path .= '%3B' . $type; + } elseif ($typecode === 'a' || $typecode === 'i' || $typecode === 'd') { + $type_ret = ";type=$typecode"; + } + } else { + $uri->path .= '%3B' . $type; + } + $uri->path = str_replace(';', '%3B', $uri->path); + $uri->path .= $type_ret; + } + + return true; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/URIScheme/http.php b/lib/htmlpurifier/library/HTMLPurifier/URIScheme/http.php new file mode 100644 index 00000000..959b8daf --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URIScheme/http.php @@ -0,0 +1,19 @@ +userinfo = null; + return true; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/URIScheme/https.php b/lib/htmlpurifier/library/HTMLPurifier/URIScheme/https.php new file mode 100644 index 00000000..159c2874 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URIScheme/https.php @@ -0,0 +1,13 @@ +userinfo = null; + $uri->host = null; + $uri->port = null; + // we need to validate path against RFC 2368's addr-spec + return true; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/URIScheme/news.php b/lib/htmlpurifier/library/HTMLPurifier/URIScheme/news.php new file mode 100644 index 00000000..84a6748d --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URIScheme/news.php @@ -0,0 +1,22 @@ +userinfo = null; + $uri->host = null; + $uri->port = null; + $uri->query = null; + // typecode check needed on path + return true; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/URIScheme/nntp.php b/lib/htmlpurifier/library/HTMLPurifier/URIScheme/nntp.php new file mode 100644 index 00000000..4ccea0df --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URIScheme/nntp.php @@ -0,0 +1,19 @@ +userinfo = null; + $uri->query = null; + return true; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/URISchemeRegistry.php b/lib/htmlpurifier/library/HTMLPurifier/URISchemeRegistry.php new file mode 100644 index 00000000..576bf7b6 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URISchemeRegistry.php @@ -0,0 +1,68 @@ +get('URI.AllowedSchemes'); + if (!$config->get('URI.OverrideAllowedSchemes') && + !isset($allowed_schemes[$scheme]) + ) { + return; + } + + if (isset($this->schemes[$scheme])) return $this->schemes[$scheme]; + if (!isset($allowed_schemes[$scheme])) return; + + $class = 'HTMLPurifier_URIScheme_' . $scheme; + if (!class_exists($class)) return; + $this->schemes[$scheme] = new $class(); + return $this->schemes[$scheme]; + } + + /** + * Registers a custom scheme to the cache, bypassing reflection. + * @param $scheme Scheme name + * @param $scheme_obj HTMLPurifier_URIScheme object + */ + public function register($scheme, $scheme_obj) { + $this->schemes[$scheme] = $scheme_obj; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/UnitConverter.php b/lib/htmlpurifier/library/HTMLPurifier/UnitConverter.php new file mode 100644 index 00000000..545d4262 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/UnitConverter.php @@ -0,0 +1,254 @@ + array( + 'px' => 3, // This is as per CSS 2.1 and Firefox. Your mileage may vary + 'pt' => 4, + 'pc' => 48, + 'in' => 288, + self::METRIC => array('pt', '0.352777778', 'mm'), + ), + self::METRIC => array( + 'mm' => 1, + 'cm' => 10, + self::ENGLISH => array('mm', '2.83464567', 'pt'), + ), + ); + + /** + * Minimum bcmath precision for output. + */ + protected $outputPrecision; + + /** + * Bcmath precision for internal calculations. + */ + protected $internalPrecision; + + /** + * Whether or not BCMath is available + */ + private $bcmath; + + public function __construct($output_precision = 4, $internal_precision = 10, $force_no_bcmath = false) { + $this->outputPrecision = $output_precision; + $this->internalPrecision = $internal_precision; + $this->bcmath = !$force_no_bcmath && function_exists('bcmul'); + } + + /** + * Converts a length object of one unit into another unit. + * @param HTMLPurifier_Length $length + * Instance of HTMLPurifier_Length to convert. You must validate() + * it before passing it here! + * @param string $to_unit + * Unit to convert to. + * @note + * About precision: This conversion function pays very special + * attention to the incoming precision of values and attempts + * to maintain a number of significant figure. Results are + * fairly accurate up to nine digits. Some caveats: + * - If a number is zero-padded as a result of this significant + * figure tracking, the zeroes will be eliminated. + * - If a number contains less than four sigfigs ($outputPrecision) + * and this causes some decimals to be excluded, those + * decimals will be added on. + */ + public function convert($length, $to_unit) { + + if (!$length->isValid()) return false; + + $n = $length->getN(); + $unit = $length->getUnit(); + + if ($n === '0' || $unit === false) { + return new HTMLPurifier_Length('0', false); + } + + $state = $dest_state = false; + foreach (self::$units as $k => $x) { + if (isset($x[$unit])) $state = $k; + if (isset($x[$to_unit])) $dest_state = $k; + } + if (!$state || !$dest_state) return false; + + // Some calculations about the initial precision of the number; + // this will be useful when we need to do final rounding. + $sigfigs = $this->getSigFigs($n); + if ($sigfigs < $this->outputPrecision) $sigfigs = $this->outputPrecision; + + // BCMath's internal precision deals only with decimals. Use + // our default if the initial number has no decimals, or increase + // it by how ever many decimals, thus, the number of guard digits + // will always be greater than or equal to internalPrecision. + $log = (int) floor(log(abs($n), 10)); + $cp = ($log < 0) ? $this->internalPrecision - $log : $this->internalPrecision; // internal precision + + for ($i = 0; $i < 2; $i++) { + + // Determine what unit IN THIS SYSTEM we need to convert to + if ($dest_state === $state) { + // Simple conversion + $dest_unit = $to_unit; + } else { + // Convert to the smallest unit, pending a system shift + $dest_unit = self::$units[$state][$dest_state][0]; + } + + // Do the conversion if necessary + if ($dest_unit !== $unit) { + $factor = $this->div(self::$units[$state][$unit], self::$units[$state][$dest_unit], $cp); + $n = $this->mul($n, $factor, $cp); + $unit = $dest_unit; + } + + // Output was zero, so bail out early. Shouldn't ever happen. + if ($n === '') { + $n = '0'; + $unit = $to_unit; + break; + } + + // It was a simple conversion, so bail out + if ($dest_state === $state) { + break; + } + + if ($i !== 0) { + // Conversion failed! Apparently, the system we forwarded + // to didn't have this unit. This should never happen! + return false; + } + + // Pre-condition: $i == 0 + + // Perform conversion to next system of units + $n = $this->mul($n, self::$units[$state][$dest_state][1], $cp); + $unit = self::$units[$state][$dest_state][2]; + $state = $dest_state; + + // One more loop around to convert the unit in the new system. + + } + + // Post-condition: $unit == $to_unit + if ($unit !== $to_unit) return false; + + // Useful for debugging: + //echo "
n"; + //echo "$n\nsigfigs = $sigfigs\nnew_log = $new_log\nlog = $log\nrp = $rp\n
\n"; + + $n = $this->round($n, $sigfigs); + if (strpos($n, '.') !== false) $n = rtrim($n, '0'); + $n = rtrim($n, '.'); + + return new HTMLPurifier_Length($n, $unit); + } + + /** + * Returns the number of significant figures in a string number. + * @param string $n Decimal number + * @return int number of sigfigs + */ + public function getSigFigs($n) { + $n = ltrim($n, '0+-'); + $dp = strpos($n, '.'); // decimal position + if ($dp === false) { + $sigfigs = strlen(rtrim($n, '0')); + } else { + $sigfigs = strlen(ltrim($n, '0.')); // eliminate extra decimal character + if ($dp !== 0) $sigfigs--; + } + return $sigfigs; + } + + /** + * Adds two numbers, using arbitrary precision when available. + */ + private function add($s1, $s2, $scale) { + if ($this->bcmath) return bcadd($s1, $s2, $scale); + else return $this->scale($s1 + $s2, $scale); + } + + /** + * Multiples two numbers, using arbitrary precision when available. + */ + private function mul($s1, $s2, $scale) { + if ($this->bcmath) return bcmul($s1, $s2, $scale); + else return $this->scale($s1 * $s2, $scale); + } + + /** + * Divides two numbers, using arbitrary precision when available. + */ + private function div($s1, $s2, $scale) { + if ($this->bcmath) return bcdiv($s1, $s2, $scale); + else return $this->scale($s1 / $s2, $scale); + } + + /** + * Rounds a number according to the number of sigfigs it should have, + * using arbitrary precision when available. + */ + private function round($n, $sigfigs) { + $new_log = (int) floor(log(abs($n), 10)); // Number of digits left of decimal - 1 + $rp = $sigfigs - $new_log - 1; // Number of decimal places needed + $neg = $n < 0 ? '-' : ''; // Negative sign + if ($this->bcmath) { + if ($rp >= 0) { + $n = bcadd($n, $neg . '0.' . str_repeat('0', $rp) . '5', $rp + 1); + $n = bcdiv($n, '1', $rp); + } else { + // This algorithm partially depends on the standardized + // form of numbers that comes out of bcmath. + $n = bcadd($n, $neg . '5' . str_repeat('0', $new_log - $sigfigs), 0); + $n = substr($n, 0, $sigfigs + strlen($neg)) . str_repeat('0', $new_log - $sigfigs + 1); + } + return $n; + } else { + return $this->scale(round($n, $sigfigs - $new_log - 1), $rp + 1); + } + } + + /** + * Scales a float to $scale digits right of decimal point, like BCMath. + */ + private function scale($r, $scale) { + if ($scale < 0) { + // The f sprintf type doesn't support negative numbers, so we + // need to cludge things manually. First get the string. + $r = sprintf('%.0f', (float) $r); + // Due to floating point precision loss, $r will more than likely + // look something like 4652999999999.9234. We grab one more digit + // than we need to precise from $r and then use that to round + // appropriately. + $precise = (string) round(substr($r, 0, strlen($r) + $scale), -1); + // Now we return it, truncating the zero that was rounded off. + return substr($precise, 0, -1) . str_repeat('0', -$scale + 1); + } + return sprintf('%.' . $scale . 'f', (float) $r); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/VarParser.php b/lib/htmlpurifier/library/HTMLPurifier/VarParser.php new file mode 100644 index 00000000..68e72ae8 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/VarParser.php @@ -0,0 +1,154 @@ + self::STRING, + 'istring' => self::ISTRING, + 'text' => self::TEXT, + 'itext' => self::ITEXT, + 'int' => self::INT, + 'float' => self::FLOAT, + 'bool' => self::BOOL, + 'lookup' => self::LOOKUP, + 'list' => self::ALIST, + 'hash' => self::HASH, + 'mixed' => self::MIXED + ); + + /** + * Lookup table of types that are string, and can have aliases or + * allowed value lists. + */ + static public $stringTypes = array( + self::STRING => true, + self::ISTRING => true, + self::TEXT => true, + self::ITEXT => true, + ); + + /** + * Validate a variable according to type. Throws + * HTMLPurifier_VarParserException if invalid. + * It may return NULL as a valid type if $allow_null is true. + * + * @param $var Variable to validate + * @param $type Type of variable, see HTMLPurifier_VarParser->types + * @param $allow_null Whether or not to permit null as a value + * @return Validated and type-coerced variable + */ + final public function parse($var, $type, $allow_null = false) { + if (is_string($type)) { + if (!isset(HTMLPurifier_VarParser::$types[$type])) { + throw new HTMLPurifier_VarParserException("Invalid type '$type'"); + } else { + $type = HTMLPurifier_VarParser::$types[$type]; + } + } + $var = $this->parseImplementation($var, $type, $allow_null); + if ($allow_null && $var === null) return null; + // These are basic checks, to make sure nothing horribly wrong + // happened in our implementations. + switch ($type) { + case (self::STRING): + case (self::ISTRING): + case (self::TEXT): + case (self::ITEXT): + if (!is_string($var)) break; + if ($type == self::ISTRING || $type == self::ITEXT) $var = strtolower($var); + return $var; + case (self::INT): + if (!is_int($var)) break; + return $var; + case (self::FLOAT): + if (!is_float($var)) break; + return $var; + case (self::BOOL): + if (!is_bool($var)) break; + return $var; + case (self::LOOKUP): + case (self::ALIST): + case (self::HASH): + if (!is_array($var)) break; + if ($type === self::LOOKUP) { + foreach ($var as $k) if ($k !== true) $this->error('Lookup table contains value other than true'); + } elseif ($type === self::ALIST) { + $keys = array_keys($var); + if (array_keys($keys) !== $keys) $this->error('Indices for list are not uniform'); + } + return $var; + case (self::MIXED): + return $var; + default: + $this->errorInconsistent(get_class($this), $type); + } + $this->errorGeneric($var, $type); + } + + /** + * Actually implements the parsing. Base implementation is to not + * do anything to $var. Subclasses should overload this! + */ + protected function parseImplementation($var, $type, $allow_null) { + return $var; + } + + /** + * Throws an exception. + */ + protected function error($msg) { + throw new HTMLPurifier_VarParserException($msg); + } + + /** + * Throws an inconsistency exception. + * @note This should not ever be called. It would be called if we + * extend the allowed values of HTMLPurifier_VarParser without + * updating subclasses. + */ + protected function errorInconsistent($class, $type) { + throw new HTMLPurifier_Exception("Inconsistency in $class: ".HTMLPurifier_VarParser::getTypeName($type)." not implemented"); + } + + /** + * Generic error for if a type didn't work. + */ + protected function errorGeneric($var, $type) { + $vtype = gettype($var); + $this->error("Expected type ".HTMLPurifier_VarParser::getTypeName($type).", got $vtype"); + } + + static public function getTypeName($type) { + static $lookup; + if (!$lookup) { + // Lazy load the alternative lookup table + $lookup = array_flip(HTMLPurifier_VarParser::$types); + } + if (!isset($lookup[$type])) return 'unknown'; + return $lookup[$type]; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/VarParser/Flexible.php b/lib/htmlpurifier/library/HTMLPurifier/VarParser/Flexible.php new file mode 100644 index 00000000..21b87675 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/VarParser/Flexible.php @@ -0,0 +1,103 @@ + $j) $var[$i] = trim($j); + if ($type === self::HASH) { + // key:value,key2:value2 + $nvar = array(); + foreach ($var as $keypair) { + $c = explode(':', $keypair, 2); + if (!isset($c[1])) continue; + $nvar[trim($c[0])] = trim($c[1]); + } + $var = $nvar; + } + } + if (!is_array($var)) break; + $keys = array_keys($var); + if ($keys === array_keys($keys)) { + if ($type == self::ALIST) return $var; + elseif ($type == self::LOOKUP) { + $new = array(); + foreach ($var as $key) { + $new[$key] = true; + } + return $new; + } else break; + } + if ($type === self::ALIST) { + trigger_error("Array list did not have consecutive integer indexes", E_USER_WARNING); + return array_values($var); + } + if ($type === self::LOOKUP) { + foreach ($var as $key => $value) { + if ($value !== true) { + trigger_error("Lookup array has non-true value at key '$key'; maybe your input array was not indexed numerically", E_USER_WARNING); + } + $var[$key] = true; + } + } + return $var; + default: + $this->errorInconsistent(__CLASS__, $type); + } + $this->errorGeneric($var, $type); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/VarParser/Native.php b/lib/htmlpurifier/library/HTMLPurifier/VarParser/Native.php new file mode 100644 index 00000000..b02a6de5 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/VarParser/Native.php @@ -0,0 +1,26 @@ +evalExpression($var); + } + + protected function evalExpression($expr) { + $var = null; + $result = eval("\$var = $expr;"); + if ($result === false) { + throw new HTMLPurifier_VarParserException("Fatal error in evaluated code"); + } + return $var; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/VarParserException.php b/lib/htmlpurifier/library/HTMLPurifier/VarParserException.php new file mode 100644 index 00000000..5df34149 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/VarParserException.php @@ -0,0 +1,11 @@ +

"); + $res = $purifier->purify($res); if (get_pref($link, "STRIP_IMAGES", $owner)) { $res = preg_replace('/]+>/is', '', $res); diff --git a/include/sanity_check.php b/include/sanity_check.php index 2195945b..11da8592 100644 --- a/include/sanity_check.php +++ b/include/sanity_check.php @@ -23,6 +23,12 @@ $array_push($errors, "Configuration file (config.php) has incorrect version. Update it with new options from config.php-dist and set CONFIG_VERSION to the correct value."); } + $purifier_cache_dir = CACHE_DIR . "/htmlpurifier"; + + if (!is_writable($purifier_cache_dir)) { + array_push($errors, "HTMLPurifier cache directory should be writable by anyone (chmod -R 777 $purifier_cache_dir)"); + } + if (!is_writable(CACHE_DIR . "/images")) { array_push($errors, "Image cache is not writable (chmod -R 777 ".CACHE_DIR."/images)"); } diff --git a/lib/htmlpurifier/CREDITS b/lib/htmlpurifier/CREDITS new file mode 100644 index 00000000..7921b45a --- /dev/null +++ b/lib/htmlpurifier/CREDITS @@ -0,0 +1,9 @@ + +CREDITS + +Almost everything written by Edward Z. Yang (Ambush Commander). Lots of thanks +to the DevNetwork Community for their help (see docs/ref-devnetwork.html for +more details), Feyd especially (namely IPv6 and optimization). Thanks to RSnake +for letting me package his fantastic XSS cheatsheet for a smoketest. + + vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/LICENSE b/lib/htmlpurifier/LICENSE new file mode 100644 index 00000000..8c88a20d --- /dev/null +++ b/lib/htmlpurifier/LICENSE @@ -0,0 +1,504 @@ + GNU LESSER GENERAL PUBLIC LICENSE + Version 2.1, February 1999 + + Copyright (C) 1991, 1999 Free Software Foundation, Inc. + 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA + Everyone is permitted to copy and distribute verbatim copies + of this license document, but changing it is not allowed. + +[This is the first released version of the Lesser GPL. It also counts + as the successor of the GNU Library Public License, version 2, hence + the version number 2.1.] + + Preamble + + The licenses for most software are designed to take away your +freedom to share and change it. By contrast, the GNU General Public +Licenses are intended to guarantee your freedom to share and change +free software--to make sure the software is free for all its users. + + This license, the Lesser General Public License, applies to some +specially designated software packages--typically libraries--of the +Free Software Foundation and other authors who decide to use it. You +can use it too, but we suggest you first think carefully about whether +this license or the ordinary General Public License is the better +strategy to use in any particular case, based on the explanations below. + + When we speak of free software, we are referring to freedom of use, +not price. Our General Public Licenses are designed to make sure that +you have the freedom to distribute copies of free software (and charge +for this service if you wish); that you receive source code or can get +it if you want it; that you can change the software and use pieces of +it in new free programs; and that you are informed that you can do +these things. + + To protect your rights, we need to make restrictions that forbid +distributors to deny you these rights or to ask you to surrender these +rights. These restrictions translate to certain responsibilities for +you if you distribute copies of the library or if you modify it. + + For example, if you distribute copies of the library, whether gratis +or for a fee, you must give the recipients all the rights that we gave +you. You must make sure that they, too, receive or can get the source +code. If you link other code with the library, you must provide +complete object files to the recipients, so that they can relink them +with the library after making changes to the library and recompiling +it. And you must show them these terms so they know their rights. + + We protect your rights with a two-step method: (1) we copyright the +library, and (2) we offer you this license, which gives you legal +permission to copy, distribute and/or modify the library. + + To protect each distributor, we want to make it very clear that +there is no warranty for the free library. Also, if the library is +modified by someone else and passed on, the recipients should know +that what they have is not the original version, so that the original +author's reputation will not be affected by problems that might be +introduced by others. + + Finally, software patents pose a constant threat to the existence of +any free program. We wish to make sure that a company cannot +effectively restrict the users of a free program by obtaining a +restrictive license from a patent holder. Therefore, we insist that +any patent license obtained for a version of the library must be +consistent with the full freedom of use specified in this license. + + Most GNU software, including some libraries, is covered by the +ordinary GNU General Public License. This license, the GNU Lesser +General Public License, applies to certain designated libraries, and +is quite different from the ordinary General Public License. We use +this license for certain libraries in order to permit linking those +libraries into non-free programs. + + When a program is linked with a library, whether statically or using +a shared library, the combination of the two is legally speaking a +combined work, a derivative of the original library. The ordinary +General Public License therefore permits such linking only if the +entire combination fits its criteria of freedom. The Lesser General +Public License permits more lax criteria for linking other code with +the library. + + We call this license the "Lesser" General Public License because it +does Less to protect the user's freedom than the ordinary General +Public License. It also provides other free software developers Less +of an advantage over competing non-free programs. These disadvantages +are the reason we use the ordinary General Public License for many +libraries. However, the Lesser license provides advantages in certain +special circumstances. + + For example, on rare occasions, there may be a special need to +encourage the widest possible use of a certain library, so that it becomes +a de-facto standard. To achieve this, non-free programs must be +allowed to use the library. A more frequent case is that a free +library does the same job as widely used non-free libraries. In this +case, there is little to gain by limiting the free library to free +software only, so we use the Lesser General Public License. + + In other cases, permission to use a particular library in non-free +programs enables a greater number of people to use a large body of +free software. For example, permission to use the GNU C Library in +non-free programs enables many more people to use the whole GNU +operating system, as well as its variant, the GNU/Linux operating +system. + + Although the Lesser General Public License is Less protective of the +users' freedom, it does ensure that the user of a program that is +linked with the Library has the freedom and the wherewithal to run +that program using a modified version of the Library. + + The precise terms and conditions for copying, distribution and +modification follow. Pay close attention to the difference between a +"work based on the library" and a "work that uses the library". The +former contains code derived from the library, whereas the latter must +be combined with the library in order to run. + + GNU LESSER GENERAL PUBLIC LICENSE + TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION + + 0. This License Agreement applies to any software library or other +program which contains a notice placed by the copyright holder or +other authorized party saying it may be distributed under the terms of +this Lesser General Public License (also called "this License"). +Each licensee is addressed as "you". + + A "library" means a collection of software functions and/or data +prepared so as to be conveniently linked with application programs +(which use some of those functions and data) to form executables. + + The "Library", below, refers to any such software library or work +which has been distributed under these terms. A "work based on the +Library" means either the Library or any derivative work under +copyright law: that is to say, a work containing the Library or a +portion of it, either verbatim or with modifications and/or translated +straightforwardly into another language. (Hereinafter, translation is +included without limitation in the term "modification".) + + "Source code" for a work means the preferred form of the work for +making modifications to it. For a library, complete source code means +all the source code for all modules it contains, plus any associated +interface definition files, plus the scripts used to control compilation +and installation of the library. + + Activities other than copying, distribution and modification are not +covered by this License; they are outside its scope. The act of +running a program using the Library is not restricted, and output from +such a program is covered only if its contents constitute a work based +on the Library (independent of the use of the Library in a tool for +writing it). Whether that is true depends on what the Library does +and what the program that uses the Library does. + + 1. You may copy and distribute verbatim copies of the Library's +complete source code as you receive it, in any medium, provided that +you conspicuously and appropriately publish on each copy an +appropriate copyright notice and disclaimer of warranty; keep intact +all the notices that refer to this License and to the absence of any +warranty; and distribute a copy of this License along with the +Library. + + You may charge a fee for the physical act of transferring a copy, +and you may at your option offer warranty protection in exchange for a +fee. + + 2. You may modify your copy or copies of the Library or any portion +of it, thus forming a work based on the Library, and copy and +distribute such modifications or work under the terms of Section 1 +above, provided that you also meet all of these conditions: + + a) The modified work must itself be a software library. + + b) You must cause the files modified to carry prominent notices + stating that you changed the files and the date of any change. + + c) You must cause the whole of the work to be licensed at no + charge to all third parties under the terms of this License. + + d) If a facility in the modified Library refers to a function or a + table of data to be supplied by an application program that uses + the facility, other than as an argument passed when the facility + is invoked, then you must make a good faith effort to ensure that, + in the event an application does not supply such function or + table, the facility still operates, and performs whatever part of + its purpose remains meaningful. + + (For example, a function in a library to compute square roots has + a purpose that is entirely well-defined independent of the + application. Therefore, Subsection 2d requires that any + application-supplied function or table used by this function must + be optional: if the application does not supply it, the square + root function must still compute square roots.) + +These requirements apply to the modified work as a whole. If +identifiable sections of that work are not derived from the Library, +and can be reasonably considered independent and separate works in +themselves, then this License, and its terms, do not apply to those +sections when you distribute them as separate works. But when you +distribute the same sections as part of a whole which is a work based +on the Library, the distribution of the whole must be on the terms of +this License, whose permissions for other licensees extend to the +entire whole, and thus to each and every part regardless of who wrote +it. + +Thus, it is not the intent of this section to claim rights or contest +your rights to work written entirely by you; rather, the intent is to +exercise the right to control the distribution of derivative or +collective works based on the Library. + +In addition, mere aggregation of another work not based on the Library +with the Library (or with a work based on the Library) on a volume of +a storage or distribution medium does not bring the other work under +the scope of this License. + + 3. You may opt to apply the terms of the ordinary GNU General Public +License instead of this License to a given copy of the Library. To do +this, you must alter all the notices that refer to this License, so +that they refer to the ordinary GNU General Public License, version 2, +instead of to this License. (If a newer version than version 2 of the +ordinary GNU General Public License has appeared, then you can specify +that version instead if you wish.) Do not make any other change in +these notices. + + Once this change is made in a given copy, it is irreversible for +that copy, so the ordinary GNU General Public License applies to all +subsequent copies and derivative works made from that copy. + + This option is useful when you wish to copy part of the code of +the Library into a program that is not a library. + + 4. You may copy and distribute the Library (or a portion or +derivative of it, under Section 2) in object code or executable form +under the terms of Sections 1 and 2 above provided that you accompany +it with the complete corresponding machine-readable source code, which +must be distributed under the terms of Sections 1 and 2 above on a +medium customarily used for software interchange. + + If distribution of object code is made by offering access to copy +from a designated place, then offering equivalent access to copy the +source code from the same place satisfies the requirement to +distribute the source code, even though third parties are not +compelled to copy the source along with the object code. + + 5. A program that contains no derivative of any portion of the +Library, but is designed to work with the Library by being compiled or +linked with it, is called a "work that uses the Library". Such a +work, in isolation, is not a derivative work of the Library, and +therefore falls outside the scope of this License. + + However, linking a "work that uses the Library" with the Library +creates an executable that is a derivative of the Library (because it +contains portions of the Library), rather than a "work that uses the +library". The executable is therefore covered by this License. +Section 6 states terms for distribution of such executables. + + When a "work that uses the Library" uses material from a header file +that is part of the Library, the object code for the work may be a +derivative work of the Library even though the source code is not. +Whether this is true is especially significant if the work can be +linked without the Library, or if the work is itself a library. The +threshold for this to be true is not precisely defined by law. + + If such an object file uses only numerical parameters, data +structure layouts and accessors, and small macros and small inline +functions (ten lines or less in length), then the use of the object +file is unrestricted, regardless of whether it is legally a derivative +work. (Executables containing this object code plus portions of the +Library will still fall under Section 6.) + + Otherwise, if the work is a derivative of the Library, you may +distribute the object code for the work under the terms of Section 6. +Any executables containing that work also fall under Section 6, +whether or not they are linked directly with the Library itself. + + 6. As an exception to the Sections above, you may also combine or +link a "work that uses the Library" with the Library to produce a +work containing portions of the Library, and distribute that work +under terms of your choice, provided that the terms permit +modification of the work for the customer's own use and reverse +engineering for debugging such modifications. + + You must give prominent notice with each copy of the work that the +Library is used in it and that the Library and its use are covered by +this License. You must supply a copy of this License. If the work +during execution displays copyright notices, you must include the +copyright notice for the Library among them, as well as a reference +directing the user to the copy of this License. Also, you must do one +of these things: + + a) Accompany the work with the complete corresponding + machine-readable source code for the Library including whatever + changes were used in the work (which must be distributed under + Sections 1 and 2 above); and, if the work is an executable linked + with the Library, with the complete machine-readable "work that + uses the Library", as object code and/or source code, so that the + user can modify the Library and then relink to produce a modified + executable containing the modified Library. (It is understood + that the user who changes the contents of definitions files in the + Library will not necessarily be able to recompile the application + to use the modified definitions.) + + b) Use a suitable shared library mechanism for linking with the + Library. A suitable mechanism is one that (1) uses at run time a + copy of the library already present on the user's computer system, + rather than copying library functions into the executable, and (2) + will operate properly with a modified version of the library, if + the user installs one, as long as the modified version is + interface-compatible with the version that the work was made with. + + c) Accompany the work with a written offer, valid for at + least three years, to give the same user the materials + specified in Subsection 6a, above, for a charge no more + than the cost of performing this distribution. + + d) If distribution of the work is made by offering access to copy + from a designated place, offer equivalent access to copy the above + specified materials from the same place. + + e) Verify that the user has already received a copy of these + materials or that you have already sent this user a copy. + + For an executable, the required form of the "work that uses the +Library" must include any data and utility programs needed for +reproducing the executable from it. However, as a special exception, +the materials to be distributed need not include anything that is +normally distributed (in either source or binary form) with the major +components (compiler, kernel, and so on) of the operating system on +which the executable runs, unless that component itself accompanies +the executable. + + It may happen that this requirement contradicts the license +restrictions of other proprietary libraries that do not normally +accompany the operating system. Such a contradiction means you cannot +use both them and the Library together in an executable that you +distribute. + + 7. You may place library facilities that are a work based on the +Library side-by-side in a single library together with other library +facilities not covered by this License, and distribute such a combined +library, provided that the separate distribution of the work based on +the Library and of the other library facilities is otherwise +permitted, and provided that you do these two things: + + a) Accompany the combined library with a copy of the same work + based on the Library, uncombined with any other library + facilities. This must be distributed under the terms of the + Sections above. + + b) Give prominent notice with the combined library of the fact + that part of it is a work based on the Library, and explaining + where to find the accompanying uncombined form of the same work. + + 8. You may not copy, modify, sublicense, link with, or distribute +the Library except as expressly provided under this License. Any +attempt otherwise to copy, modify, sublicense, link with, or +distribute the Library is void, and will automatically terminate your +rights under this License. However, parties who have received copies, +or rights, from you under this License will not have their licenses +terminated so long as such parties remain in full compliance. + + 9. You are not required to accept this License, since you have not +signed it. However, nothing else grants you permission to modify or +distribute the Library or its derivative works. These actions are +prohibited by law if you do not accept this License. Therefore, by +modifying or distributing the Library (or any work based on the +Library), you indicate your acceptance of this License to do so, and +all its terms and conditions for copying, distributing or modifying +the Library or works based on it. + + 10. Each time you redistribute the Library (or any work based on the +Library), the recipient automatically receives a license from the +original licensor to copy, distribute, link with or modify the Library +subject to these terms and conditions. You may not impose any further +restrictions on the recipients' exercise of the rights granted herein. +You are not responsible for enforcing compliance by third parties with +this License. + + 11. If, as a consequence of a court judgment or allegation of patent +infringement or for any other reason (not limited to patent issues), +conditions are imposed on you (whether by court order, agreement or +otherwise) that contradict the conditions of this License, they do not +excuse you from the conditions of this License. If you cannot +distribute so as to satisfy simultaneously your obligations under this +License and any other pertinent obligations, then as a consequence you +may not distribute the Library at all. For example, if a patent +license would not permit royalty-free redistribution of the Library by +all those who receive copies directly or indirectly through you, then +the only way you could satisfy both it and this License would be to +refrain entirely from distribution of the Library. + +If any portion of this section is held invalid or unenforceable under any +particular circumstance, the balance of the section is intended to apply, +and the section as a whole is intended to apply in other circumstances. + +It is not the purpose of this section to induce you to infringe any +patents or other property right claims or to contest validity of any +such claims; this section has the sole purpose of protecting the +integrity of the free software distribution system which is +implemented by public license practices. Many people have made +generous contributions to the wide range of software distributed +through that system in reliance on consistent application of that +system; it is up to the author/donor to decide if he or she is willing +to distribute software through any other system and a licensee cannot +impose that choice. + +This section is intended to make thoroughly clear what is believed to +be a consequence of the rest of this License. + + 12. If the distribution and/or use of the Library is restricted in +certain countries either by patents or by copyrighted interfaces, the +original copyright holder who places the Library under this License may add +an explicit geographical distribution limitation excluding those countries, +so that distribution is permitted only in or among countries not thus +excluded. In such case, this License incorporates the limitation as if +written in the body of this License. + + 13. The Free Software Foundation may publish revised and/or new +versions of the Lesser General Public License from time to time. +Such new versions will be similar in spirit to the present version, +but may differ in detail to address new problems or concerns. + +Each version is given a distinguishing version number. If the Library +specifies a version number of this License which applies to it and +"any later version", you have the option of following the terms and +conditions either of that version or of any later version published by +the Free Software Foundation. If the Library does not specify a +license version number, you may choose any version ever published by +the Free Software Foundation. + + 14. If you wish to incorporate parts of the Library into other free +programs whose distribution conditions are incompatible with these, +write to the author to ask for permission. For software which is +copyrighted by the Free Software Foundation, write to the Free +Software Foundation; we sometimes make exceptions for this. Our +decision will be guided by the two goals of preserving the free status +of all derivatives of our free software and of promoting the sharing +and reuse of software generally. + + NO WARRANTY + + 15. BECAUSE THE LIBRARY IS LICENSED FREE OF CHARGE, THERE IS NO +WARRANTY FOR THE LIBRARY, TO THE EXTENT PERMITTED BY APPLICABLE LAW. +EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR +OTHER PARTIES PROVIDE THE LIBRARY "AS IS" WITHOUT WARRANTY OF ANY +KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR +PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE +LIBRARY IS WITH YOU. SHOULD THE LIBRARY PROVE DEFECTIVE, YOU ASSUME +THE COST OF ALL NECESSARY SERVICING, REPAIR OR CORRECTION. + + 16. IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN +WRITING WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY +AND/OR REDISTRIBUTE THE LIBRARY AS PERMITTED ABOVE, BE LIABLE TO YOU +FOR DAMAGES, INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR +CONSEQUENTIAL DAMAGES ARISING OUT OF THE USE OR INABILITY TO USE THE +LIBRARY (INCLUDING BUT NOT LIMITED TO LOSS OF DATA OR DATA BEING +RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD PARTIES OR A +FAILURE OF THE LIBRARY TO OPERATE WITH ANY OTHER SOFTWARE), EVEN IF +SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH +DAMAGES. + + END OF TERMS AND CONDITIONS + + How to Apply These Terms to Your New Libraries + + If you develop a new library, and you want it to be of the greatest +possible use to the public, we recommend making it free software that +everyone can redistribute and change. You can do so by permitting +redistribution under these terms (or, alternatively, under the terms of the +ordinary General Public License). + + To apply these terms, attach the following notices to the library. It is +safest to attach them to the start of each source file to most effectively +convey the exclusion of warranty; and each file should have at least the +"copyright" line and a pointer to where the full notice is found. + + + Copyright (C) + + This library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + This library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with this library; if not, write to the Free Software + Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA + +Also add information on how to contact you by electronic and paper mail. + +You should also get your employer (if you work as a programmer) or your +school, if any, to sign a "copyright disclaimer" for the library, if +necessary. Here is a sample; alter the names: + + Yoyodyne, Inc., hereby disclaims all copyright interest in the + library `Frob' (a library for tweaking knobs) written by James Random Hacker. + + , 1 April 1990 + Ty Coon, President of Vice + +That's all there is to it! + + vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier.auto.php b/lib/htmlpurifier/library/HTMLPurifier.auto.php new file mode 100644 index 00000000..1960c399 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier.auto.php @@ -0,0 +1,11 @@ +purify($html, $config); +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier.includes.php b/lib/htmlpurifier/library/HTMLPurifier.includes.php new file mode 100644 index 00000000..0ceff6a9 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier.includes.php @@ -0,0 +1,221 @@ + $attributes) { + $allowed_elements[$element] = true; + foreach ($attributes as $attribute => $x) { + $allowed_attributes["$element.$attribute"] = true; + } + } + $config->set('HTML.AllowedElements', $allowed_elements); + $config->set('HTML.AllowedAttributes', $allowed_attributes); + $allowed_schemes = array(); + if ($allowed_protocols !== null) { + $config->set('URI.AllowedSchemes', $allowed_protocols); + } + $purifier = new HTMLPurifier($config); + return $purifier->purify($string); +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier.path.php b/lib/htmlpurifier/library/HTMLPurifier.path.php new file mode 100644 index 00000000..39b1b653 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier.path.php @@ -0,0 +1,11 @@ +config = HTMLPurifier_Config::create($config); + + $this->strategy = new HTMLPurifier_Strategy_Core(); + + } + + /** + * Adds a filter to process the output. First come first serve + * @param $filter HTMLPurifier_Filter object + / + public function addFilter($filter) { + trigger_error('HTMLPurifier->addFilter() is deprecated, use configuration directives in the Filter namespace or Filter.Custom', E_USER_WARNING); + $this->filters[] = $filter; + } + + /* + * Filters an HTML snippet/document to be XSS-free and standards-compliant. + * + * @param $html String of HTML to purify + * @param $config HTMLPurifier_Config object for this operation, if omitted, + * defaults to the config object specified during this + * object's construction. The parameter can also be any type + * that HTMLPurifier_Config::create() supports. + * @return Purified HTML + / + public function purify($html, $config = null) { + + // :TODO: make the config merge in, instead of replace + $config = $config ? HTMLPurifier_Config::create($config) : $this->config; + + // implementation is partially environment dependant, partially + // configuration dependant + $lexer = HTMLPurifier_Lexer::create($config); + + $context = new HTMLPurifier_Context(); + + // setup HTML generator + $this->generator = new HTMLPurifier_Generator($config, $context); + $context->register('Generator', $this->generator); + + // set up global context variables + if ($config->get('Core.CollectErrors')) { + // may get moved out if other facilities use it + $language_factory = HTMLPurifier_LanguageFactory::instance(); + $language = $language_factory->create($config, $context); + $context->register('Locale', $language); + + $error_collector = new HTMLPurifier_ErrorCollector($context); + $context->register('ErrorCollector', $error_collector); + } + + // setup id_accumulator context, necessary due to the fact that + // AttrValidator can be called from many places + $id_accumulator = HTMLPurifier_IDAccumulator::build($config, $context); + $context->register('IDAccumulator', $id_accumulator); + + $html = HTMLPurifier_Encoder::convertToUTF8($html, $config, $context); + + // setup filters + $filter_flags = $config->getBatch('Filter'); + $custom_filters = $filter_flags['Custom']; + unset($filter_flags['Custom']); + $filters = array(); + foreach ($filter_flags as $filter => $flag) { + if (!$flag) continue; + if (strpos($filter, '.') !== false) continue; + $class = "HTMLPurifier_Filter_$filter"; + $filters[] = new $class; + } + foreach ($custom_filters as $filter) { + // maybe "HTMLPurifier_Filter_$filter", but be consistent with AutoFormat + $filters[] = $filter; + } + $filters = array_merge($filters, $this->filters); + // maybe prepare(), but later + + for ($i = 0, $filter_size = count($filters); $i < $filter_size; $i++) { + $html = $filters[$i]->preFilter($html, $config, $context); + } + + // purified HTML + $html = + $this->generator->generateFromTokens( + // list of tokens + $this->strategy->execute( + // list of un-purified tokens + $lexer->tokenizeHTML( + // un-purified HTML + $html, $config, $context + ), + $config, $context + ) + ); + + for ($i = $filter_size - 1; $i >= 0; $i--) { + $html = $filters[$i]->postFilter($html, $config, $context); + } + + $html = HTMLPurifier_Encoder::convertFromUTF8($html, $config, $context); + $this->context =& $context; + return $html; + } + + /* + * Filters an array of HTML snippets + * @param $config Optional HTMLPurifier_Config object for this operation. + * See HTMLPurifier::purify() for more details. + * @return Array of purified HTML + / + public function purifyArray($array_of_html, $config = null) { + $context_array = array(); + foreach ($array_of_html as $key => $html) { + $array_of_html[$key] = $this->purify($html, $config); + $context_array[$key] = $this->context; + } + $this->context = $context_array; + return $array_of_html; + } + + /* + * Singleton for enforcing just one HTML Purifier in your system + * @param $prototype Optional prototype HTMLPurifier instance to + * overload singleton with, or HTMLPurifier_Config + * instance to configure the generated version with. + / + public static function instance($prototype = null) { + if (!self::$instance \|\| $prototype) { + if ($prototype instanceof HTMLPurifier) { + self::$instance = $prototype; + } elseif ($prototype) { + self::$instance = new HTMLPurifier($prototype); + } else { + self::$instance = new HTMLPurifier(); + } + } + return self::$instance; + } + + /* + * @note Backwards compatibility, see instance() + / + public static function getInstance($prototype = null) { + return HTMLPurifier::instance($prototype); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier.safe-includes.php b/lib/htmlpurifier/library/HTMLPurifier.safe-includes.php new file mode 100644 index 00000000..d49b196c --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier.safe-includes.php @@ -0,0 +1,215 @@ +attr_collections as $coll_i => $coll) { + if (!isset($this->info[$coll_i])) { + $this->info[$coll_i] = array(); + } + foreach ($coll as $attr_i => $attr) { + if ($attr_i === 0 && isset($this->info[$coll_i][$attr_i])) { + // merge in includes + $this->info[$coll_i][$attr_i] = array_merge( + $this->info[$coll_i][$attr_i], $attr); + continue; + } + $this->info[$coll_i][$attr_i] = $attr; + } + } + } + // perform internal expansions and inclusions + foreach ($this->info as $name => $attr) { + // merge attribute collections that include others + $this->performInclusions($this->info[$name]); + // replace string identifiers with actual attribute objects + $this->expandIdentifiers($this->info[$name], $attr_types); + } + } + + /* + * Takes a reference to an attribute associative array and performs + * all inclusions specified by the zero index. + * @param &$attr Reference to attribute array + / + public function performInclusions(&$attr) { + if (!isset($attr[0])) return; + $merge = $attr[0]; + $seen = array(); // recursion guard + // loop through all the inclusions + for ($i = 0; isset($merge[$i]); $i++) { + if (isset($seen[$merge[$i]])) continue; + $seen[$merge[$i]] = true; + // foreach attribute of the inclusion, copy it over + if (!isset($this->info[$merge[$i]])) continue; + foreach ($this->info[$merge[$i]] as $key => $value) { + if (isset($attr[$key])) continue; // also catches more inclusions + $attr[$key] = $value; + } + if (isset($this->info[$merge[$i]][0])) { + // recursion + $merge = array_merge($merge, $this->info[$merge[$i]][0]); + } + } + unset($attr[0]); + } + + /* + * Expands all string identifiers in an attribute array by replacing + * them with the appropriate values inside HTMLPurifier_AttrTypes + * @param &$attr Reference to attribute array + * @param $attr_types HTMLPurifier_AttrTypes instance + / + public function expandIdentifiers(&$attr, $attr_types) { + + // because foreach will process new elements we add, make sure we + // skip duplicates + $processed = array(); + + foreach ($attr as $def_i => $def) { + // skip inclusions + if ($def_i === 0) continue; + + if (isset($processed[$def_i])) continue; + + // determine whether or not attribute is required + if ($required = (strpos($def_i, '') !== false)) { + // rename the definition + unset($attr[$def_i]); + $def_i = trim($def_i, ''); + $attr[$def_i] = $def; + } + + $processed[$def_i] = true; + + // if we've already got a literal object, move on + if (is_object($def)) { + // preserve previous required + $attr[$def_i]->required = ($required \|\| $attr[$def_i]->required); + continue; + } + + if ($def === false) { + unset($attr[$def_i]); + continue; + } + + if ($t = $attr_types->get($def)) { + $attr[$def_i] = $t; + $attr[$def_i]->required = $required; + } else { + unset($attr[$def_i]); + } + } + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef.php new file mode 100644 index 00000000..b2e4f36c --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef.php @@ -0,0 +1,123 @@ + by removing + leading and trailing whitespace, ignoring line feeds, and replacing + * carriage returns and tabs with spaces. While most useful for HTML + * attributes specified as CDATA, it can also be applied to most CSS + * values. + * + * @note This method is not entirely standards compliant, as trim() removes + * more types of whitespace than specified in the spec. In practice, + * this is rarely a problem, as those extra characters usually have + * already been removed by HTMLPurifier_Encoder. + * + * @warning This processing is inconsistent with XML's whitespace handling + * as specified by section 3.3.3 and referenced XHTML 1.0 section + * 4.7. However, note that we are NOT necessarily + * parsing XML, thus, this behavior may still be correct. We + * assume that newlines have been normalized. + / + public function parseCDATA($string) { + $string = trim($string); + $string = str_replace(array("\n", "\t", "\r"), ' ', $string); + return $string; + } + + /* + * Factory method for creating this class from a string. + * @param $string String construction info + * @return Created AttrDef object corresponding to $string + / + public function make($string) { + // default implementation, return a flyweight of this object. + // If $string has an effect on the returned object (i.e. you + // need to overload this method), it is best + // to clone or instantiate new copies. (Instantiation is safer.) + return $this; + } + + /* + * Removes spaces from rgb(0, 0, 0) so that shorthand CSS properties work + * properly. THIS IS A HACK! + / + protected function mungeRgb($string) { + return preg_replace('/rgb\((\d+)\s,\s(\d+)\s,\s(\d+)\)/', 'rgb(\1,\2,\3)', $string); + } + + /* + * Parses a possibly escaped CSS string and returns the "pure" + * version of it. + / + protected function expandCSSEscape($string) { + // flexibly parse it + $ret = ''; + for ($i = 0, $c = strlen($string); $i < $c; $i++) { + if ($string[$i] === '\\') { + $i++; + if ($i >= $c) { + $ret .= '\\'; + break; + } + if (ctype_xdigit($string[$i])) { + $code = $string[$i]; + for ($a = 1, $i++; $i < $c && $a < 6; $i++, $a++) { + if (!ctype_xdigit($string[$i])) break; + $code .= $string[$i]; + } + // We have to be extremely careful when adding + // new characters, to make sure we're not breaking + // the encoding. + $char = HTMLPurifier_Encoder::unichr(hexdec($code)); + if (HTMLPurifier_Encoder::cleanUTF8($char) === '') continue; + $ret .= $char; + if ($i < $c && trim($string[$i]) !== '') $i--; + continue; + } + if ($string[$i] === "\n") continue; + } + $ret .= $string[$i]; + } + return $ret; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS.php new file mode 100644 index 00000000..953e7067 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS.php @@ -0,0 +1,87 @@ +parseCDATA($css); + + $definition = $config->getCSSDefinition(); + + // we're going to break the spec and explode by semicolons. + // This is because semicolon rarely appears in escaped form + // Doing this is generally flaky but fast + // IT MIGHT APPEAR IN URIs, see HTMLPurifier_AttrDef_CSSURI + // for details + + $declarations = explode(';', $css); + $propvalues = array(); + + /* + * Name of the current CSS property being validated. + / + $property = false; + $context->register('CurrentCSSProperty', $property); + + foreach ($declarations as $declaration) { + if (!$declaration) continue; + if (!strpos($declaration, ':')) continue; + list($property, $value) = explode(':', $declaration, 2); + $property = trim($property); + $value = trim($value); + $ok = false; + do { + if (isset($definition->info[$property])) { + $ok = true; + break; + } + if (ctype_lower($property)) break; + $property = strtolower($property); + if (isset($definition->info[$property])) { + $ok = true; + break; + } + } while(0); + if (!$ok) continue; + // inefficient call, since the validator will do this again + if (strtolower(trim($value)) !== 'inherit') { + // inherit works for everything (but only on the base property) + $result = $definition->info[$property]->validate( + $value, $config, $context ); + } else { + $result = 'inherit'; + } + if ($result === false) continue; + $propvalues[$property] = $result; + } + + $context->destroy('CurrentCSSProperty'); + + // procedure does not write the new CSS simultaneously, so it's + // slightly inefficient, but it's the only way of getting rid of + // duplicates. Perhaps config to optimize it, but not now. + + $new_declarations = ''; + foreach ($propvalues as $prop => $value) { + $new_declarations .= "$prop:$value;"; + } + + return $new_declarations ? $new_declarations : false; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/AlphaValue.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/AlphaValue.php new file mode 100644 index 00000000..292c040d --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/AlphaValue.php @@ -0,0 +1,21 @@ + 1.0) $result = '1'; + return $result; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Background.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Background.php new file mode 100644 index 00000000..3a3d20cd --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Background.php @@ -0,0 +1,87 @@ +getCSSDefinition(); + $this->info['background-color'] = $def->info['background-color']; + $this->info['background-image'] = $def->info['background-image']; + $this->info['background-repeat'] = $def->info['background-repeat']; + $this->info['background-attachment'] = $def->info['background-attachment']; + $this->info['background-position'] = $def->info['background-position']; + } + + public function validate($string, $config, $context) { + + // regular pre-processing + $string = $this->parseCDATA($string); + if ($string === '') return false; + + // munge rgb() decl if necessary + $string = $this->mungeRgb($string); + + // assumes URI doesn't have spaces in it + $bits = explode(' ', strtolower($string)); // bits to process + + $caught = array(); + $caught['color'] = false; + $caught['image'] = false; + $caught['repeat'] = false; + $caught['attachment'] = false; + $caught['position'] = false; + + $i = 0; // number of catches + $none = false; + + foreach ($bits as $bit) { + if ($bit === '') continue; + foreach ($caught as $key => $status) { + if ($key != 'position') { + if ($status !== false) continue; + $r = $this->info['background-' . $key]->validate($bit, $config, $context); + } else { + $r = $bit; + } + if ($r === false) continue; + if ($key == 'position') { + if ($caught[$key] === false) $caught[$key] = ''; + $caught[$key] .= $r . ' '; + } else { + $caught[$key] = $r; + } + $i++; + break; + } + } + + if (!$i) return false; + if ($caught['position'] !== false) { + $caught['position'] = $this->info['background-position']-> + validate($caught['position'], $config, $context); + } + + $ret = array(); + foreach ($caught as $value) { + if ($value === false) continue; + $ret[] = $value; + } + + if (empty($ret)) return false; + return implode(' ', $ret); + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/BackgroundPosition.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/BackgroundPosition.php new file mode 100644 index 00000000..fae82eae --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/BackgroundPosition.php @@ -0,0 +1,133 @@ + \| \| left \| center \| right + ] + [ + \| \| top \| center \| bottom + ]? + ] \| + [ // this signifies that the vertical and horizontal adjectives + // can be arbitrarily ordered, however, there can only be two, + // one of each, or none at all + [ + left \| center \| right + ] \|\| + [ + top \| center \| bottom + ] + ] + top, left = 0% + center, (none) = 50% + bottom, right = 100% +/ + +/* QuirksMode says: + keyword + length/percentage must be ordered correctly, as per W3C + + Internet Explorer and Opera, however, support arbitrary ordering. We + should fix it up. + + Minor issue though, not strictly necessary. +/ + +// control freaks may appreciate the ability to convert these to +// percentages or something, but it's not necessary + +/* + * Validates the value of background-position. + / +class HTMLPurifier_AttrDef_CSS_BackgroundPosition extends HTMLPurifier_AttrDef +{ + + protected $length; + protected $percentage; + + public function __construct() { + $this->length = new HTMLPurifier_AttrDef_CSS_Length(); + $this->percentage = new HTMLPurifier_AttrDef_CSS_Percentage(); + } + + public function validate($string, $config, $context) { + $string = $this->parseCDATA($string); + $bits = explode(' ', $string); + + $keywords = array(); + $keywords['h'] = false; // left, right + $keywords['v'] = false; // top, bottom + $keywords['ch'] = false; // center (first word) + $keywords['cv'] = false; // center (second word) + $measures = array(); + + $i = 0; + + $lookup = array( + 'top' => 'v', + 'bottom' => 'v', + 'left' => 'h', + 'right' => 'h', + 'center' => 'c' + ); + + foreach ($bits as $bit) { + if ($bit === '') continue; + + // test for keyword + $lbit = ctype_lower($bit) ? $bit : strtolower($bit); + if (isset($lookup[$lbit])) { + $status = $lookup[$lbit]; + if ($status == 'c') { + if ($i == 0) { + $status = 'ch'; + } else { + $status = 'cv'; + } + } + $keywords[$status] = $lbit; + $i++; + } + + // test for length + $r = $this->length->validate($bit, $config, $context); + if ($r !== false) { + $measures[] = $r; + $i++; + } + + // test for percentage + $r = $this->percentage->validate($bit, $config, $context); + if ($r !== false) { + $measures[] = $r; + $i++; + } + + } + + if (!$i) return false; // no valid values were caught + + $ret = array(); + + // first keyword + if ($keywords['h']) $ret[] = $keywords['h']; + elseif ($keywords['ch']) { + $ret[] = $keywords['ch']; + $keywords['cv'] = false; // prevent re-use: center = center center + } + elseif (count($measures)) $ret[] = array_shift($measures); + + if ($keywords['v']) $ret[] = $keywords['v']; + elseif ($keywords['cv']) $ret[] = $keywords['cv']; + elseif (count($measures)) $ret[] = array_shift($measures); + + if (empty($ret)) return false; + return implode(' ', $ret); + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Border.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Border.php new file mode 100644 index 00000000..42a1d1b4 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Border.php @@ -0,0 +1,43 @@ +getCSSDefinition(); + $this->info['border-width'] = $def->info['border-width']; + $this->info['border-style'] = $def->info['border-style']; + $this->info['border-top-color'] = $def->info['border-top-color']; + } + + public function validate($string, $config, $context) { + $string = $this->parseCDATA($string); + $string = $this->mungeRgb($string); + $bits = explode(' ', $string); + $done = array(); // segments we've finished + $ret = ''; // return value + foreach ($bits as $bit) { + foreach ($this->info as $propname => $validator) { + if (isset($done[$propname])) continue; + $r = $validator->validate($bit, $config, $context); + if ($r !== false) { + $ret .= $r . ' '; + $done[$propname] = true; + break; + } + } + } + return rtrim($ret); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Color.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Color.php new file mode 100644 index 00000000..07f95a67 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Color.php @@ -0,0 +1,78 @@ +get('Core.ColorKeywords'); + + $color = trim($color); + if ($color === '') return false; + + $lower = strtolower($color); + if (isset($colors[$lower])) return $colors[$lower]; + + if (strpos($color, 'rgb(') !== false) { + // rgb literal handling + $length = strlen($color); + if (strpos($color, ')') !== $length - 1) return false; + $triad = substr($color, 4, $length - 4 - 1); + $parts = explode(',', $triad); + if (count($parts) !== 3) return false; + $type = false; // to ensure that they're all the same type + $new_parts = array(); + foreach ($parts as $part) { + $part = trim($part); + if ($part === '') return false; + $length = strlen($part); + if ($part[$length - 1] === '%') { + // handle percents + if (!$type) { + $type = 'percentage'; + } elseif ($type !== 'percentage') { + return false; + } + $num = (float) substr($part, 0, $length - 1); + if ($num < 0) $num = 0; + if ($num > 100) $num = 100; + $new_parts[] = "$num%"; + } else { + // handle integers + if (!$type) { + $type = 'integer'; + } elseif ($type !== 'integer') { + return false; + } + $num = (int) $part; + if ($num < 0) $num = 0; + if ($num > 255) $num = 255; + $new_parts[] = (string) $num; + } + } + $new_triad = implode(',', $new_parts); + $color = "rgb($new_triad)"; + } else { + // hexadecimal handling + if ($color[0] === '#') { + $hex = substr($color, 1); + } else { + $hex = $color; + $color = '#' . $color; + } + $length = strlen($hex); + if ($length !== 3 && $length !== 6) return false; + if (!ctype_xdigit($hex)) return false; + } + + return $color; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Composite.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Composite.php new file mode 100644 index 00000000..de1289cb --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Composite.php @@ -0,0 +1,38 @@ +defs = $defs; + } + + public function validate($string, $config, $context) { + foreach ($this->defs as $i => $def) { + $result = $this->defs[$i]->validate($string, $config, $context); + if ($result !== false) return $result; + } + return false; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/DenyElementDecorator.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/DenyElementDecorator.php new file mode 100644 index 00000000..6599c5b2 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/DenyElementDecorator.php @@ -0,0 +1,28 @@ +def = $def; + $this->element = $element; + } + /* + * Checks if CurrentToken is set and equal to $this->element + / + public function validate($string, $config, $context) { + $token = $context->get('CurrentToken', true); + if ($token && $token->name == $this->element) return false; + return $this->def->validate($string, $config, $context); + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Filter.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Filter.php new file mode 100644 index 00000000..147894b8 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Filter.php @@ -0,0 +1,54 @@ +intValidator = new HTMLPurifier_AttrDef_Integer(); + } + + public function validate($value, $config, $context) { + $value = $this->parseCDATA($value); + if ($value === 'none') return $value; + // if we looped this we could support multiple filters + $function_length = strcspn($value, '('); + $function = trim(substr($value, 0, $function_length)); + if ($function !== 'alpha' && + $function !== 'Alpha' && + $function !== 'progid:DXImageTransform.Microsoft.Alpha' + ) return false; + $cursor = $function_length + 1; + $parameters_length = strcspn($value, ')', $cursor); + $parameters = substr($value, $cursor, $parameters_length); + $params = explode(',', $parameters); + $ret_params = array(); + $lookup = array(); + foreach ($params as $param) { + list($key, $value) = explode('=', $param); + $key = trim($key); + $value = trim($value); + if (isset($lookup[$key])) continue; + if ($key !== 'opacity') continue; + $value = $this->intValidator->validate($value, $config, $context); + if ($value === false) continue; + $int = (int) $value; + if ($int > 100) $value = '100'; + if ($int < 0) $value = '0'; + $ret_params[] = "$key=$value"; + $lookup[$key] = true; + } + $ret_parameters = implode(',', $ret_params); + $ret_function = "$function($ret_parameters)"; + return $ret_function; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Font.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Font.php new file mode 100644 index 00000000..699ee0b7 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Font.php @@ -0,0 +1,149 @@ +getCSSDefinition(); + $this->info['font-style'] = $def->info['font-style']; + $this->info['font-variant'] = $def->info['font-variant']; + $this->info['font-weight'] = $def->info['font-weight']; + $this->info['font-size'] = $def->info['font-size']; + $this->info['line-height'] = $def->info['line-height']; + $this->info['font-family'] = $def->info['font-family']; + } + + public function validate($string, $config, $context) { + + static $system_fonts = array( + 'caption' => true, + 'icon' => true, + 'menu' => true, + 'message-box' => true, + 'small-caption' => true, + 'status-bar' => true + ); + + // regular pre-processing + $string = $this->parseCDATA($string); + if ($string === '') return false; + + // check if it's one of the keywords + $lowercase_string = strtolower($string); + if (isset($system_fonts[$lowercase_string])) { + return $lowercase_string; + } + + $bits = explode(' ', $string); // bits to process + $stage = 0; // this indicates what we're looking for + $caught = array(); // which stage 0 properties have we caught? + $stage_1 = array('font-style', 'font-variant', 'font-weight'); + $final = ''; // output + + for ($i = 0, $size = count($bits); $i < $size; $i++) { + if ($bits[$i] === '') continue; + switch ($stage) { + + // attempting to catch font-style, font-variant or font-weight + case 0: + foreach ($stage_1 as $validator_name) { + if (isset($caught[$validator_name])) continue; + $r = $this->info[$validator_name]->validate( + $bits[$i], $config, $context); + if ($r !== false) { + $final .= $r . ' '; + $caught[$validator_name] = true; + break; + } + } + // all three caught, continue on + if (count($caught) >= 3) $stage = 1; + if ($r !== false) break; + + // attempting to catch font-size and perhaps line-height + case 1: + $found_slash = false; + if (strpos($bits[$i], '/') !== false) { + list($font_size, $line_height) = + explode('/', $bits[$i]); + if ($line_height === '') { + // ooh, there's a space after the slash! + $line_height = false; + $found_slash = true; + } + } else { + $font_size = $bits[$i]; + $line_height = false; + } + $r = $this->info['font-size']->validate( + $font_size, $config, $context); + if ($r !== false) { + $final .= $r; + // attempt to catch line-height + if ($line_height === false) { + // we need to scroll forward + for ($j = $i + 1; $j < $size; $j++) { + if ($bits[$j] === '') continue; + if ($bits[$j] === '/') { + if ($found_slash) { + return false; + } else { + $found_slash = true; + continue; + } + } + $line_height = $bits[$j]; + break; + } + } else { + // slash already found + $found_slash = true; + $j = $i; + } + if ($found_slash) { + $i = $j; + $r = $this->info['line-height']->validate( + $line_height, $config, $context); + if ($r !== false) { + $final .= '/' . $r; + } + } + $final .= ' '; + $stage = 2; + break; + } + return false; + + // attempting to catch font-family + case 2: + $font_family = + implode(' ', array_slice($bits, $i, $size - $i)); + $r = $this->info['font-family']->validate( + $font_family, $config, $context); + if ($r !== false) { + $final .= $r . ' '; + // processing completed successfully + return rtrim($final); + } + return false; + } + } + return false; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/FontFamily.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/FontFamily.php new file mode 100644 index 00000000..0d9a4e12 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/FontFamily.php @@ -0,0 +1,197 @@ +mask = '- '; + for ($c = 'a'; $c <= 'z'; $c++) $this->mask .= $c; + for ($c = 'A'; $c <= 'Z'; $c++) $this->mask .= $c; + for ($c = '0'; $c <= '9'; $c++) $this->mask .= $c; // cast-y, but should be fine + // special bytes used by UTF-8 + for ($i = 0x80; $i <= 0xFF; $i++) { + // We don't bother excluding invalid bytes in this range, + // because the our restriction of well-formed UTF-8 will + // prevent these from ever occurring. + $this->mask .= chr($i); + } + + / + PHP's internal strcspn implementation is + O(length of string * length of mask), making it inefficient + for large masks. However, it's still faster than + preg_match 8) + for (p = s1;;) { + spanp = s2; + do { + if (spanp == c \|\| p == s1_end) { + return p - s1; + } + } while (spanp++ < (s2_end - 1)); + c = ++p; + } + / + // possible optimization: invert the mask. + } + + public function validate($string, $config, $context) { + static $generic_names = array( + 'serif' => true, + 'sans-serif' => true, + 'monospace' => true, + 'fantasy' => true, + 'cursive' => true + ); + $allowed_fonts = $config->get('CSS.AllowedFonts'); + + // assume that no font names contain commas in them + $fonts = explode(',', $string); + $final = ''; + foreach($fonts as $font) { + $font = trim($font); + if ($font === '') continue; + // match a generic name + if (isset($generic_names[$font])) { + if ($allowed_fonts === null \|\| isset($allowed_fonts[$font])) { + $final .= $font . ', '; + } + continue; + } + // match a quoted name + if ($font[0] === '"' \|\| $font[0] === "'") { + $length = strlen($font); + if ($length <= 2) continue; + $quote = $font[0]; + if ($font[$length - 1] !== $quote) continue; + $font = substr($font, 1, $length - 2); + } + + $font = $this->expandCSSEscape($font); + + // $font is a pure representation of the font name + + if ($allowed_fonts !== null && !isset($allowed_fonts[$font])) { + continue; + } + + if (ctype_alnum($font) && $font !== '') { + // very simple font, allow it in unharmed + $final .= $font . ', '; + continue; + } + + // bugger out on whitespace. form feed (0C) really + // shouldn't show up regardless + $font = str_replace(array("\n", "\t", "\r", "\x0C"), ' ', $font); + + // Here, there are various classes of characters which need + // to be treated differently: + // - Alphanumeric characters are essentially safe. We + // handled these above. + // - Spaces require quoting, though most parsers will do + // the right thing if there aren't any characters that + // can be misinterpreted + // - Dashes rarely occur, but they fairly unproblematic + // for parsing/rendering purposes. + // The above characters cover the majority of Western font + // names. + // - Arbitrary Unicode characters not in ASCII. Because + // most parsers give little thought to Unicode, treatment + // of these codepoints is basically uniform, even for + // punctuation-like codepoints. These characters can + // show up in non-Western pages and are supported by most + // major browsers, for example: "ï¼ï¼³ ææ" is a + // legitimate font-name + // . See + // the CSS3 spec for more examples: + // + // You can see live samples of these on the Internet: + // + // However, most of these fonts have ASCII equivalents: + // for example, 'MS Mincho', and it's considered + // professional to use ASCII font names instead of + // Unicode font names. Thanks Takeshi Terada for + // providing this information. + // The following characters, to my knowledge, have not been + // used to name font names. + // - Single quote. While theoretically you might find a + // font name that has a single quote in its name (serving + // as an apostrophe, e.g. Dave's Scribble), I haven't + // been able to find any actual examples of this. + // Internet Explorer's cssText translation (which I + // believe is invoked by innerHTML) normalizes any + // quoting to single quotes, and fails to escape single + // quotes. (Note that this is not IE's behavior for all + // CSS properties, just some sort of special casing for + // font-family). So a single quote cannot* be used + // safely in the font-family context if there will be an + // innerHTML/cssText translation. Note that Firefox 3.x + // does this too. + // - Double quote. In IE, these get normalized to + // single-quotes, no matter what the encoding. (Fun + // fact, in IE8, the 'content' CSS property gained + // support, where they special cased to preserve encoded + // double quotes, but still translate unadorned double + // quotes into single quotes.) So, because their + // fixpoint behavior is identical to single quotes, they + // cannot be allowed either. Firefox 3.x displays + // single-quote style behavior. + // - Backslashes are reduced by one (so \\ -> \) every + // iteration, so they cannot be used safely. This shows + // up in IE7, IE8 and FF3 + // - Semicolons, commas and backticks are handled properly. + // - The rest of the ASCII punctuation is handled properly. + // We haven't checked what browsers do to unadorned + // versions, but this is not important as long as the + // browser doesn't /remove/ surrounding quotes (as IE does + // for HTML). + // + // With these results in hand, we conclude that there are + // various levels of safety: + // - Paranoid: alphanumeric, spaces and dashes(?) + // - International: Paranoid + non-ASCII Unicode + // - Edgy: Everything except quotes, backslashes + // - NoJS: Standards compliance, e.g. sod IE. Note that + // with some judicious character escaping (since certain + // types of escaping doesn't work) this is theoretically + // OK as long as innerHTML/cssText is not called. + // We believe that international is a reasonable default + // (that we will implement now), and once we do more + // extensive research, we may feel comfortable with dropping + // it down to edgy. + + // Edgy: alphanumeric, spaces, dashes and Unicode. Use of + // str(c)spn assumes that the string was already well formed + // Unicode (which of course it is). + if (strspn($font, $this->mask) !== strlen($font)) { + continue; + } + + // Historical: + // In the absence of innerHTML/cssText, these ugly + // transforms don't pose a security risk (as \\ and \" + // might--these escapes are not supported by most browsers). + // We could try to be clever and use single-quote wrapping + // when there is a double quote present, but I have choosen + // not to implement that. (NOTE: you can reduce the amount + // of escapes by one depending on what quoting style you use) + // $font = str_replace('\\', '\\5C ', $font); + // $font = str_replace('"', '\\22 ', $font); + // $font = str_replace("'", '\\27 ', $font); + + // font possibly with spaces, requires quoting + $final .= "'$font', "; + } + $final = rtrim($final, ', '); + if ($final === '') return false; + return $final; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Ident.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Ident.php new file mode 100644 index 00000000..779794a0 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Ident.php @@ -0,0 +1,24 @@ +def = $def; + $this->allow = $allow; + } + /** + * Intercepts and removes !important if necessary + / + public function validate($string, $config, $context) { + // test for ! and important tokens + $string = trim($string); + $is_important = false; + // :TODO: optimization: test directly for !important and ! important + if (strlen($string) >= 9 && substr($string, -9) === 'important') { + $temp = rtrim(substr($string, 0, -9)); + // use a temp, because we might want to restore important + if (strlen($temp) >= 1 && substr($temp, -1) === '!') { + $string = rtrim(substr($temp, 0, -1)); + $is_important = true; + } + } + $string = $this->def->validate($string, $config, $context); + if ($this->allow && $is_important) $string .= ' !important'; + return $string; + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Length.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Length.php new file mode 100644 index 00000000..a07ec581 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Length.php @@ -0,0 +1,47 @@ +min = $min !== null ? HTMLPurifier_Length::make($min) : null; + $this->max = $max !== null ? HTMLPurifier_Length::make($max) : null; + } + + public function validate($string, $config, $context) { + $string = $this->parseCDATA($string); + + // Optimizations + if ($string === '') return false; + if ($string === '0') return '0'; + if (strlen($string) === 1) return false; + + $length = HTMLPurifier_Length::make($string); + if (!$length->isValid()) return false; + + if ($this->min) { + $c = $length->compareTo($this->min); + if ($c === false) return false; + if ($c < 0) return false; + } + if ($this->max) { + $c = $length->compareTo($this->max); + if ($c === false) return false; + if ($c > 0) return false; + } + + return $length->toString(); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/ListStyle.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/ListStyle.php new file mode 100644 index 00000000..4406868c --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/ListStyle.php @@ -0,0 +1,78 @@ +getCSSDefinition(); + $this->info['list-style-type'] = $def->info['list-style-type']; + $this->info['list-style-position'] = $def->info['list-style-position']; + $this->info['list-style-image'] = $def->info['list-style-image']; + } + + public function validate($string, $config, $context) { + + // regular pre-processing + $string = $this->parseCDATA($string); + if ($string === '') return false; + + // assumes URI doesn't have spaces in it + $bits = explode(' ', strtolower($string)); // bits to process + + $caught = array(); + $caught['type'] = false; + $caught['position'] = false; + $caught['image'] = false; + + $i = 0; // number of catches + $none = false; + + foreach ($bits as $bit) { + if ($i >= 3) return; // optimization bit + if ($bit === '') continue; + foreach ($caught as $key => $status) { + if ($status !== false) continue; + $r = $this->info['list-style-' . $key]->validate($bit, $config, $context); + if ($r === false) continue; + if ($r === 'none') { + if ($none) continue; + else $none = true; + if ($key == 'image') continue; + } + $caught[$key] = $r; + $i++; + break; + } + } + + if (!$i) return false; + + $ret = array(); + + // construct type + if ($caught['type']) $ret[] = $caught['type']; + + // construct image + if ($caught['image']) $ret[] = $caught['image']; + + // construct position + if ($caught['position']) $ret[] = $caught['position']; + + if (empty($ret)) return false; + return implode(' ', $ret); + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Multiple.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Multiple.php new file mode 100644 index 00000000..4d62a40d --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Multiple.php @@ -0,0 +1,58 @@ +single = $single; + $this->max = $max; + } + + public function validate($string, $config, $context) { + $string = $this->parseCDATA($string); + if ($string === '') return false; + $parts = explode(' ', $string); // parseCDATA replaced \r, \t and \n + $length = count($parts); + $final = ''; + for ($i = 0, $num = 0; $i < $length && $num < $this->max; $i++) { + if (ctype_space($parts[$i])) continue; + $result = $this->single->validate($parts[$i], $config, $context); + if ($result !== false) { + $final .= $result . ' '; + $num++; + } + } + if ($final === '') return false; + return rtrim($final); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Number.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Number.php new file mode 100644 index 00000000..3f99e12e --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Number.php @@ -0,0 +1,69 @@ +non_negative = $non_negative; + } + + /* + * @warning Some contexts do not pass $config, $context. These + * variables should not be used without checking HTMLPurifier_Length + / + public function validate($number, $config, $context) { + + $number = $this->parseCDATA($number); + + if ($number === '') return false; + if ($number === '0') return '0'; + + $sign = ''; + switch ($number[0]) { + case '-': + if ($this->non_negative) return false; + $sign = '-'; + case '+': + $number = substr($number, 1); + } + + if (ctype_digit($number)) { + $number = ltrim($number, '0'); + return $number ? $sign . $number : '0'; + } + + // Period is the only non-numeric character allowed + if (strpos($number, '.') === false) return false; + + list($left, $right) = explode('.', $number, 2); + + if ($left === '' && $right === '') return false; + if ($left !== '' && !ctype_digit($left)) return false; + + $left = ltrim($left, '0'); + $right = rtrim($right, '0'); + + if ($right === '') { + return $left ? $sign . $left : '0'; + } elseif (!ctype_digit($right)) { + return false; + } + + return $sign . $left . '.' . $right; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Percentage.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Percentage.php new file mode 100644 index 00000000..c34b8fc3 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Percentage.php @@ -0,0 +1,40 @@ +number_def = new HTMLPurifier_AttrDef_CSS_Number($non_negative); + } + + public function validate($string, $config, $context) { + + $string = $this->parseCDATA($string); + + if ($string === '') return false; + $length = strlen($string); + if ($length === 1) return false; + if ($string[$length - 1] !== '%') return false; + + $number = substr($string, 0, $length - 1); + $number = $this->number_def->validate($number, $config, $context); + + if ($number === false) return false; + return "$number%"; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/TextDecoration.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/TextDecoration.php new file mode 100644 index 00000000..772c922d --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/TextDecoration.php @@ -0,0 +1,38 @@ + true, + 'overline' => true, + 'underline' => true, + ); + + $string = strtolower($this->parseCDATA($string)); + + if ($string === 'none') return $string; + + $parts = explode(' ', $string); + $final = ''; + foreach ($parts as $part) { + if (isset($allowed_values[$part])) { + $final .= $part . ' '; + } + } + $final = rtrim($final); + if ($final === '') return false; + return $final; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/URI.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/URI.php new file mode 100644 index 00000000..c2f767e5 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/URI.php @@ -0,0 +1,61 @@ +parseCDATA($uri_string); + if (strpos($uri_string, 'url(') !== 0) return false; + $uri_string = substr($uri_string, 4); + $new_length = strlen($uri_string) - 1; + if ($uri_string[$new_length] != ')') return false; + $uri = trim(substr($uri_string, 0, $new_length)); + + if (!empty($uri) && ($uri[0] == "'" \|\| $uri[0] == '"')) { + $quote = $uri[0]; + $new_length = strlen($uri) - 1; + if ($uri[$new_length] !== $quote) return false; + $uri = substr($uri, 1, $new_length - 1); + } + + $uri = $this->expandCSSEscape($uri); + + $result = parent::validate($uri, $config, $context); + + if ($result === false) return false; + + // extra sanity check; should have been done by URI + $result = str_replace(array('"', "\\", "\n", "\x0c", "\r"), "", $result); + + // suspicious characters are ()'; we're going to percent encode + // them for safety. + $result = str_replace(array('(', ')', "'"), array('%28', '%29', '%27'), $result); + + // there's an extra bug where ampersands lose their escaping on + // an innerHTML cycle, so a very unlucky query parameter could + // then change the meaning of the URL. Unfortunately, there's + // not much we can do about that... + + return "url(\"$result\")"; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/Clone.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/Clone.php new file mode 100644 index 00000000..ce68dbd5 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/Clone.php @@ -0,0 +1,28 @@ +clone = $clone; + } + + public function validate($v, $config, $context) { + return $this->clone->validate($v, $config, $context); + } + + public function make($string) { + return clone $this->clone; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/Enum.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/Enum.php new file mode 100644 index 00000000..5d603ebc --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/Enum.php @@ -0,0 +1,65 @@ +valid_values = array_flip($valid_values); + $this->case_sensitive = $case_sensitive; + } + + public function validate($string, $config, $context) { + $string = trim($string); + if (!$this->case_sensitive) { + // we may want to do full case-insensitive libraries + $string = ctype_lower($string) ? $string : strtolower($string); + } + $result = isset($this->valid_values[$string]); + + return $result ? $string : false; + } + + /* + * @param $string In form of comma-delimited list of case-insensitive + * valid values. Example: "foo,bar,baz". Prepend "s:" to make + * case sensitive + / + public function make($string) { + if (strlen($string) > 2 && $string[0] == 's' && $string[1] == ':') { + $string = substr($string, 2); + $sensitive = true; + } else { + $sensitive = false; + } + $values = explode(',', $string); + return new HTMLPurifier_AttrDef_Enum($values, $sensitive); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/Bool.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/Bool.php new file mode 100644 index 00000000..e06987eb --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/Bool.php @@ -0,0 +1,28 @@ +name = $name;} + + public function validate($string, $config, $context) { + if (empty($string)) return false; + return $this->name; + } + + /* + * @param $string Name of attribute + / + public function make($string) { + return new HTMLPurifier_AttrDef_HTML_Bool($string); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/Class.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/Class.php new file mode 100644 index 00000000..370068d9 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/Class.php @@ -0,0 +1,34 @@ +getDefinition('HTML')->doctype->name; + if ($name == "XHTML 1.1" \|\| $name == "XHTML 2.0") { + return parent::split($string, $config, $context); + } else { + return preg_split('/\s+/', $string); + } + } + protected function filter($tokens, $config, $context) { + $allowed = $config->get('Attr.AllowedClasses'); + $forbidden = $config->get('Attr.ForbiddenClasses'); + $ret = array(); + foreach ($tokens as $token) { + if ( + ($allowed === null \|\| isset($allowed[$token])) && + !isset($forbidden[$token]) && + // We need this O(n) check because of PHP's array + // implementation that casts -0 to 0. + !in_array($token, $ret, true) + ) { + $ret[] = $token; + } + } + return $ret; + } +} diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/Color.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/Color.php new file mode 100644 index 00000000..00d86572 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/Color.php @@ -0,0 +1,32 @@ +get('Core.ColorKeywords'); + + $string = trim($string); + + if (empty($string)) return false; + if (isset($colors[strtolower($string)])) return $colors[$string]; + if ($string[0] === '#') $hex = substr($string, 1); + else $hex = $string; + + $length = strlen($hex); + if ($length !== 3 && $length !== 6) return false; + if (!ctype_xdigit($hex)) return false; + if ($length === 3) $hex = $hex[0].$hex[0].$hex[1].$hex[1].$hex[2].$hex[2]; + + return "#$hex"; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/FrameTarget.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/FrameTarget.php new file mode 100644 index 00000000..ae6ea7c0 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/FrameTarget.php @@ -0,0 +1,21 @@ +valid_values === false) $this->valid_values = $config->get('Attr.AllowedFrameTargets'); + return parent::validate($string, $config, $context); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/ID.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/ID.php new file mode 100644 index 00000000..0015fa1e --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/ID.php @@ -0,0 +1,80 @@ +selector = $selector; + } + + public function validate($id, $config, $context) { + + if (!$this->selector && !$config->get('Attr.EnableID')) return false; + + $id = trim($id); // trim it first + + if ($id === '') return false; + + $prefix = $config->get('Attr.IDPrefix'); + if ($prefix !== '') { + $prefix .= $config->get('Attr.IDPrefixLocal'); + // prevent re-appending the prefix + if (strpos($id, $prefix) !== 0) $id = $prefix . $id; + } elseif ($config->get('Attr.IDPrefixLocal') !== '') { + trigger_error('%Attr.IDPrefixLocal cannot be used unless '. + '%Attr.IDPrefix is set', E_USER_WARNING); + } + + if (!$this->selector) { + $id_accumulator =& $context->get('IDAccumulator'); + if (isset($id_accumulator->ids[$id])) return false; + } + + // we purposely avoid using regex, hopefully this is faster + + if (ctype_alpha($id)) { + $result = true; + } else { + if (!ctype_alpha(@$id[0])) return false; + $trim = trim( // primitive style of regexps, I suppose + $id, + 'A..Za..z0..9:-._' + ); + $result = ($trim === ''); + } + + $regexp = $config->get('Attr.IDBlacklistRegexp'); + if ($regexp && preg_match($regexp, $id)) { + return false; + } + + if (!$this->selector && $result) $id_accumulator->add($id); + + // if no change was made to the ID, return the result + // else, return the new id if stripping whitespace made it + // valid, or return false. + return $result ? $id : false; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/Length.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/Length.php new file mode 100644 index 00000000..a242f9c2 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/Length.php @@ -0,0 +1,41 @@ + 100) return '100%'; + + return ((string) $points) . '%'; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/LinkTypes.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/LinkTypes.php new file mode 100644 index 00000000..76d25ed0 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/LinkTypes.php @@ -0,0 +1,53 @@ + 'AllowedRel', + 'rev' => 'AllowedRev' + ); + if (!isset($configLookup[$name])) { + trigger_error('Unrecognized attribute name for link '. + 'relationship.', E_USER_ERROR); + return; + } + $this->name = $configLookup[$name]; + } + + public function validate($string, $config, $context) { + + $allowed = $config->get('Attr.' . $this->name); + if (empty($allowed)) return false; + + $string = $this->parseCDATA($string); + $parts = explode(' ', $string); + + // lookup to prevent duplicates + $ret_lookup = array(); + foreach ($parts as $part) { + $part = strtolower(trim($part)); + if (!isset($allowed[$part])) continue; + $ret_lookup[$part] = true; + } + + if (empty($ret_lookup)) return false; + $string = implode(' ', array_keys($ret_lookup)); + + return $string; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/MultiLength.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/MultiLength.php new file mode 100644 index 00000000..c72fc76e --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/MultiLength.php @@ -0,0 +1,41 @@ +split($string, $config, $context); + $tokens = $this->filter($tokens, $config, $context); + if (empty($tokens)) return false; + return implode(' ', $tokens); + + } + + /* + * Splits a space separated list of tokens into its constituent parts. + / + protected function split($string, $config, $context) { + // OPTIMIZABLE! + // do the preg_match, capture all subpatterns for reformulation + + // we don't support U+00A1 and up codepoints or + // escaping because I don't know how to do that with regexps + // and plus it would complicate optimization efforts (you never + // see that anyway). + $pattern = '/(?:(?<=\s)\|\A)'. // look behind for space or string start + '((?:--\|-?[A-Za-z_])[A-Za-z_\-0-9])'. + '(?:(?=\s)\|\z)/'; // look ahead for space or string end + preg_match_all($pattern, $string, $matches); + return $matches[1]; + } + + /** + * Template method for removing certain tokens based on arbitrary criteria. + * @note If we wanted to be really functional, we'd do an array_filter + * with a callback. But... we're not. + / + protected function filter($tokens, $config, $context) { + return $tokens; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/Pixels.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/Pixels.php new file mode 100644 index 00000000..4cb2c1b8 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/Pixels.php @@ -0,0 +1,48 @@ +max = $max; + } + + public function validate($string, $config, $context) { + + $string = trim($string); + if ($string === '0') return $string; + if ($string === '') return false; + $length = strlen($string); + if (substr($string, $length - 2) == 'px') { + $string = substr($string, 0, $length - 2); + } + if (!is_numeric($string)) return false; + $int = (int) $string; + + if ($int < 0) return '0'; + + // upper-bound value, extremely high values can + // crash operating systems, see + // WARNING, above link WILL crash you if you're using Windows + + if ($this->max !== null && $int > $this->max) return (string) $this->max; + + return (string) $int; + + } + + public function make($string) { + if ($string === '') $max = null; + else $max = (int) $string; + $class = get_class($this); + return new $class($max); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/Integer.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/Integer.php new file mode 100644 index 00000000..d59738d2 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/Integer.php @@ -0,0 +1,73 @@ +negative = $negative; + $this->zero = $zero; + $this->positive = $positive; + } + + public function validate($integer, $config, $context) { + + $integer = $this->parseCDATA($integer); + if ($integer === '') return false; + + // we could possibly simply typecast it to integer, but there are + // certain fringe cases that must not return an integer. + + // clip leading sign + if ( $this->negative && $integer[0] === '-' ) { + $digits = substr($integer, 1); + if ($digits === '0') $integer = '0'; // rm minus sign for zero + } elseif( $this->positive && $integer[0] === '+' ) { + $digits = $integer = substr($integer, 1); // rm unnecessary plus + } else { + $digits = $integer; + } + + // test if it's numeric + if (!ctype_digit($digits)) return false; + + // perform scope tests + if (!$this->zero && $integer == 0) return false; + if (!$this->positive && $integer > 0) return false; + if (!$this->negative && $integer < 0) return false; + + return $integer; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/Lang.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/Lang.php new file mode 100644 index 00000000..10e6da56 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/Lang.php @@ -0,0 +1,73 @@ + 8 \|\| !ctype_alnum($subtags[1])) { + return $new_string; + } + if (!ctype_lower($subtags[1])) $subtags[1] = strtolower($subtags[1]); + + $new_string .= '-' . $subtags[1]; + if ($num_subtags == 2) return $new_string; + + // process all other subtags, index 2 and up + for ($i = 2; $i < $num_subtags; $i++) { + $length = strlen($subtags[$i]); + if ($length == 0 \|\| $length > 8 \|\| !ctype_alnum($subtags[$i])) { + return $new_string; + } + if (!ctype_lower($subtags[$i])) { + $subtags[$i] = strtolower($subtags[$i]); + } + $new_string .= '-' . $subtags[$i]; + } + + return $new_string; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/Switch.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/Switch.php new file mode 100644 index 00000000..c9e3ed19 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/Switch.php @@ -0,0 +1,34 @@ +tag = $tag; + $this->withTag = $with_tag; + $this->withoutTag = $without_tag; + } + + public function validate($string, $config, $context) { + $token = $context->get('CurrentToken', true); + if (!$token \|\| $token->name !== $this->tag) { + return $this->withoutTag->validate($string, $config, $context); + } else { + return $this->withTag->validate($string, $config, $context); + } + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/Text.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/Text.php new file mode 100644 index 00000000..c6216cc5 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/Text.php @@ -0,0 +1,15 @@ +parseCDATA($string); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/URI.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/URI.php new file mode 100644 index 00000000..c2b68467 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/URI.php @@ -0,0 +1,77 @@ +parser = new HTMLPurifier_URIParser(); + $this->embedsResource = (bool) $embeds_resource; + } + + public function make($string) { + $embeds = ($string === 'embedded'); + return new HTMLPurifier_AttrDef_URI($embeds); + } + + public function validate($uri, $config, $context) { + + if ($config->get('URI.Disable')) return false; + + $uri = $this->parseCDATA($uri); + + // parse the URI + $uri = $this->parser->parse($uri); + if ($uri === false) return false; + + // add embedded flag to context for validators + $context->register('EmbeddedURI', $this->embedsResource); + + $ok = false; + do { + + // generic validation + $result = $uri->validate($config, $context); + if (!$result) break; + + // chained filtering + $uri_def = $config->getDefinition('URI'); + $result = $uri_def->filter($uri, $config, $context); + if (!$result) break; + + // scheme-specific validation + $scheme_obj = $uri->getSchemeObj($config, $context); + if (!$scheme_obj) break; + if ($this->embedsResource && !$scheme_obj->browsable) break; + $result = $scheme_obj->validate($uri, $config, $context); + if (!$result) break; + + // Post chained filtering + $result = $uri_def->postFilter($uri, $config, $context); + if (!$result) break; + + // survived gauntlet + $ok = true; + + } while (false); + + $context->destroy('EmbeddedURI'); + if (!$ok) return false; + + // back to string + return $uri->toString(); + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/URI/Email.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/URI/Email.php new file mode 100644 index 00000000..bfee9d16 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/URI/Email.php @@ -0,0 +1,17 @@ +" + // that needs more percent encoding to be done + if ($string == '') return false; + $string = trim($string); + $result = preg_match('/^[A-Z0-9._%-]+@[A-Z0-9.-]+\.[A-Z]{2,4}$/i', $string); + return $result ? $string : false; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/URI/Host.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/URI/Host.php new file mode 100644 index 00000000..125decb2 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/URI/Host.php @@ -0,0 +1,101 @@ +ipv4 = new HTMLPurifier_AttrDef_URI_IPv4(); + $this->ipv6 = new HTMLPurifier_AttrDef_URI_IPv6(); + } + + public function validate($string, $config, $context) { + $length = strlen($string); + // empty hostname is OK; it's usually semantically equivalent: + // the default host as defined by a URI scheme is used: + // + // If the URI scheme defines a default for host, then that + // default applies when the host subcomponent is undefined + // or when the registered name is empty (zero length). + if ($string === '') return ''; + if ($length > 1 && $string[0] === '[' && $string[$length-1] === ']') { + //IPv6 + $ip = substr($string, 1, $length - 2); + $valid = $this->ipv6->validate($ip, $config, $context); + if ($valid === false) return false; + return '['. $valid . ']'; + } + + // need to do checks on unusual encodings too + $ipv4 = $this->ipv4->validate($string, $config, $context); + if ($ipv4 !== false) return $ipv4; + + // A regular domain name. + + // This doesn't match I18N domain names, but we don't have proper IRI support, + // so force users to insert Punycode. + + // The productions describing this are: + $a = '[a-z]'; // alpha + $an = '[a-z0-9]'; // alphanum + $and = '[a-z0-9-]'; // alphanum \| "-" + // domainlabel = alphanum \| alphanum ( alphanum \| "-" ) alphanum + $domainlabel = "$an($and$an)?"; + // toplabel = alpha \| alpha ( alphanum \| "-" ) alphanum + $toplabel = "$a($and$an)?"; + // hostname = ( domainlabel "." ) toplabel [ "." ] + if (preg_match("/^($domainlabel\.)$toplabel\.?$/i", $string)) { + return $string; + } + + // If we have Net_IDNA2 support, we can support IRIs by + // punycoding them. (This is the most portable thing to do, + // since otherwise we have to assume browsers support + + if ($config->get('Core.EnableIDNA')) { + $idna = new Net_IDNA2(array('encoding' => 'utf8', 'overlong' => false, 'strict' => true)); + // we need to encode each period separately + $parts = explode('.', $string); + try { + $new_parts = array(); + foreach ($parts as $part) { + $encodable = false; + for ($i = 0, $c = strlen($part); $i < $c; $i++) { + if (ord($part[$i]) > 0x7a) { + $encodable = true; + break; + } + } + if (!$encodable) { + $new_parts[] = $part; + } else { + $new_parts[] = $idna->encode($part); + } + } + $string = implode('.', $new_parts); + if (preg_match("/^($domainlabel\.)$toplabel\.?$/i", $string)) { + return $string; + } + } catch (Exception $e) { + // XXX error reporting + } + } + + return false; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/URI/IPv4.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/URI/IPv4.php new file mode 100644 index 00000000..ec4cf591 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/URI/IPv4.php @@ -0,0 +1,39 @@ +ip4) $this->_loadRegex(); + + if (preg_match('#^' . $this->ip4 . '$#s', $aIP)) + { + return $aIP; + } + + return false; + + } + + /** + * Lazy load function to prevent regex from being stuffed in + * cache. + / + protected function _loadRegex() { + $oct = '(?:25[0-5]\|2[0-4][0-9]\|1[0-9]{2}\|[1-9][0-9]\|[0-9])'; // 0-255 + $this->ip4 = "(?:{$oct}\\.{$oct}\\.{$oct}\\.{$oct})"; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/URI/IPv6.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/URI/IPv6.php new file mode 100644 index 00000000..9454e9be --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/URI/IPv6.php @@ -0,0 +1,99 @@ +ip4) $this->_loadRegex(); + + $original = $aIP; + + $hex = '[0-9a-fA-F]'; + $blk = '(?:' . $hex . '{1,4})'; + $pre = '(?:/(?:12[0-8]\|1[0-1][0-9]\|[1-9][0-9]\|[0-9]))'; // /0 - /128 + + // prefix check + if (strpos($aIP, '/') !== false) + { + if (preg_match('#' . $pre . '$#s', $aIP, $find)) + { + $aIP = substr($aIP, 0, 0-strlen($find[0])); + unset($find); + } + else + { + return false; + } + } + + // IPv4-compatiblity check + if (preg_match('#(?<=:'.')' . $this->ip4 . '$#s', $aIP, $find)) + { + $aIP = substr($aIP, 0, 0-strlen($find[0])); + $ip = explode('.', $find[0]); + $ip = array_map('dechex', $ip); + $aIP .= $ip[0] . $ip[1] . ':' . $ip[2] . $ip[3]; + unset($find, $ip); + } + + // compression check + $aIP = explode('::', $aIP); + $c = count($aIP); + if ($c > 2) + { + return false; + } + elseif ($c == 2) + { + list($first, $second) = $aIP; + $first = explode(':', $first); + $second = explode(':', $second); + + if (count($first) + count($second) > 8) + { + return false; + } + + while(count($first) < 8) + { + array_push($first, '0'); + } + + array_splice($first, 8 - count($second), 8, $second); + $aIP = $first; + unset($first,$second); + } + else + { + $aIP = explode(':', $aIP[0]); + } + $c = count($aIP); + + if ($c != 8) + { + return false; + } + + // All the pieces should be 16-bit hex strings. Are they? + foreach ($aIP as $piece) + { + if (!preg_match('#^[0-9a-fA-F]{4}$#s', sprintf('%04s', $piece))) + { + return false; + } + } + + return $original; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTransform.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform.php new file mode 100644 index 00000000..e61d3e01 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform.php @@ -0,0 +1,56 @@ +confiscateAttr($attr, 'background'); + // some validation should happen here + + $this->prependCSS($attr, "background-image:url($background);"); + + return $attr; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/BdoDir.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/BdoDir.php new file mode 100644 index 00000000..4d1a0566 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/BdoDir.php @@ -0,0 +1,19 @@ +get('Attr.DefaultTextDir'); + return $attr; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/BgColor.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/BgColor.php new file mode 100644 index 00000000..ad3916bb --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/BgColor.php @@ -0,0 +1,23 @@ +confiscateAttr($attr, 'bgcolor'); + // some validation should happen here + + $this->prependCSS($attr, "background-color:$bgcolor;"); + + return $attr; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/BoolToCSS.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/BoolToCSS.php new file mode 100644 index 00000000..51159b67 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/BoolToCSS.php @@ -0,0 +1,36 @@ +attr = $attr; + $this->css = $css; + } + + public function transform($attr, $config, $context) { + if (!isset($attr[$this->attr])) return $attr; + unset($attr[$this->attr]); + $this->prependCSS($attr, $this->css); + return $attr; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Border.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Border.php new file mode 100644 index 00000000..476b0b07 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Border.php @@ -0,0 +1,18 @@ +confiscateAttr($attr, 'border'); + // some validation should happen here + $this->prependCSS($attr, "border:{$border_width}px solid;"); + return $attr; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/EnumToCSS.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/EnumToCSS.php new file mode 100644 index 00000000..2a5b4514 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/EnumToCSS.php @@ -0,0 +1,58 @@ +attr = $attr; + $this->enumToCSS = $enum_to_css; + $this->caseSensitive = (bool) $case_sensitive; + } + + public function transform($attr, $config, $context) { + + if (!isset($attr[$this->attr])) return $attr; + + $value = trim($attr[$this->attr]); + unset($attr[$this->attr]); + + if (!$this->caseSensitive) $value = strtolower($value); + + if (!isset($this->enumToCSS[$value])) { + return $attr; + } + + $this->prependCSS($attr, $this->enumToCSS[$value]); + + return $attr; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/ImgRequired.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/ImgRequired.php new file mode 100644 index 00000000..7f0e4b7a --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/ImgRequired.php @@ -0,0 +1,43 @@ +get('Core.RemoveInvalidImg')) return $attr; + $attr['src'] = $config->get('Attr.DefaultInvalidImage'); + $src = false; + } + + if (!isset($attr['alt'])) { + if ($src) { + $alt = $config->get('Attr.DefaultImageAlt'); + if ($alt === null) { + // truncate if the alt is too long + $attr['alt'] = substr(basename($attr['src']),0,40); + } else { + $attr['alt'] = $alt; + } + } else { + $attr['alt'] = $config->get('Attr.DefaultInvalidImageAlt'); + } + } + + return $attr; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/ImgSpace.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/ImgSpace.php new file mode 100644 index 00000000..fd84c10c --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/ImgSpace.php @@ -0,0 +1,44 @@ + array('left', 'right'), + 'vspace' => array('top', 'bottom') + ); + + public function __construct($attr) { + $this->attr = $attr; + if (!isset($this->css[$attr])) { + trigger_error(htmlspecialchars($attr) . ' is not valid space attribute'); + } + } + + public function transform($attr, $config, $context) { + + if (!isset($attr[$this->attr])) return $attr; + + $width = $this->confiscateAttr($attr, $this->attr); + // some validation could happen here + + if (!isset($this->css[$this->attr])) return $attr; + + $style = ''; + foreach ($this->css[$this->attr] as $suffix) { + $property = "margin-$suffix"; + $style .= "$property:{$width}px;"; + } + + $this->prependCSS($attr, $style); + + return $attr; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Input.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Input.php new file mode 100644 index 00000000..16829552 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Input.php @@ -0,0 +1,40 @@ +pixels = new HTMLPurifier_AttrDef_HTML_Pixels(); + } + + public function transform($attr, $config, $context) { + if (!isset($attr['type'])) $t = 'text'; + else $t = strtolower($attr['type']); + if (isset($attr['checked']) && $t !== 'radio' && $t !== 'checkbox') { + unset($attr['checked']); + } + if (isset($attr['maxlength']) && $t !== 'text' && $t !== 'password') { + unset($attr['maxlength']); + } + if (isset($attr['size']) && $t !== 'text' && $t !== 'password') { + $result = $this->pixels->validate($attr['size'], $config, $context); + if ($result === false) unset($attr['size']); + else $attr['size'] = $result; + } + if (isset($attr['src']) && $t !== 'image') { + unset($attr['src']); + } + if (!isset($attr['value']) && ($t === 'radio' \|\| $t === 'checkbox')) { + $attr['value'] = ''; + } + return $attr; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Lang.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Lang.php new file mode 100644 index 00000000..5869e7f8 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Lang.php @@ -0,0 +1,28 @@ +name = $name; + $this->cssName = $css_name ? $css_name : $name; + } + + public function transform($attr, $config, $context) { + if (!isset($attr[$this->name])) return $attr; + $length = $this->confiscateAttr($attr, $this->name); + if(ctype_digit($length)) $length .= 'px'; + $this->prependCSS($attr, $this->cssName . ":$length;"); + return $attr; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Name.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Name.php new file mode 100644 index 00000000..15315bc7 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Name.php @@ -0,0 +1,21 @@ +get('HTML.Attr.Name.UseCDATA')) return $attr; + if (!isset($attr['name'])) return $attr; + $id = $this->confiscateAttr($attr, 'name'); + if ( isset($attr['id'])) return $attr; + $attr['id'] = $id; + return $attr; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/NameSync.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/NameSync.php new file mode 100644 index 00000000..a95638c1 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/NameSync.php @@ -0,0 +1,27 @@ +idDef = new HTMLPurifier_AttrDef_HTML_ID(); + } + + public function transform($attr, $config, $context) { + if (!isset($attr['name'])) return $attr; + $name = $attr['name']; + if (isset($attr['id']) && $attr['id'] === $name) return $attr; + $result = $this->idDef->validate($name, $config, $context); + if ($result === false) unset($attr['name']); + else $attr['name'] = $result; + return $attr; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Nofollow.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Nofollow.php new file mode 100644 index 00000000..f7fb1209 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Nofollow.php @@ -0,0 +1,45 @@ +parser = new HTMLPurifier_URIParser(); + } + + public function transform($attr, $config, $context) { + + if (!isset($attr['href'])) { + return $attr; + } + + // XXX Kind of inefficient + $url = $this->parser->parse($attr['href']); + $scheme = $url->getSchemeObj($config, $context); + + if ($scheme->browsable && !$url->isLocal($config, $context)) { + if (isset($attr['rel'])) { + $rels = explode(' ', $attr); + if (!in_array('nofollow', $rels)) { + $rels[] = 'nofollow'; + } + $attr['rel'] = implode(' ', $rels); + } else { + $attr['rel'] = 'nofollow'; + } + } + + return $attr; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/SafeEmbed.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/SafeEmbed.php new file mode 100644 index 00000000..4da44998 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/SafeEmbed.php @@ -0,0 +1,15 @@ +uri = new HTMLPurifier_AttrDef_URI(true); // embedded + $this->wmode = new HTMLPurifier_AttrDef_Enum(array('window', 'opaque', 'transparent')); + } + + public function transform($attr, $config, $context) { + // If we add support for other objects, we'll need to alter the + // transforms. + switch ($attr['name']) { + // application/x-shockwave-flash + // Keep this synchronized with Injector/SafeObject.php + case 'allowScriptAccess': + $attr['value'] = 'never'; + break; + case 'allowNetworking': + $attr['value'] = 'internal'; + break; + case 'allowFullScreen': + if ($config->get('HTML.FlashAllowFullScreen')) { + $attr['value'] = ($attr['value'] == 'true') ? 'true' : 'false'; + } else { + $attr['value'] = 'false'; + } + break; + case 'wmode': + $attr['value'] = $this->wmode->validate($attr['value'], $config, $context); + break; + case 'movie': + case 'src': + $attr['name'] = "movie"; + $attr['value'] = $this->uri->validate($attr['value'], $config, $context); + break; + case 'flashvars': + // we're going to allow arbitrary inputs to the SWF, on + // the reasoning that it could only hack the SWF, not us. + break; + // add other cases to support other param name/value pairs + default: + $attr['name'] = $attr['value'] = null; + } + return $attr; + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/ScriptRequired.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/ScriptRequired.php new file mode 100644 index 00000000..4499050a --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/ScriptRequired.php @@ -0,0 +1,16 @@ + + / +class HTMLPurifier_AttrTransform_ScriptRequired extends HTMLPurifier_AttrTransform +{ + public function transform($attr, $config, $context) { + if (!isset($attr['type'])) { + $attr['type'] = 'text/javascript'; + } + return $attr; + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/TargetBlank.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/TargetBlank.php new file mode 100644 index 00000000..a6502c74 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/TargetBlank.php @@ -0,0 +1,38 @@ +parser = new HTMLPurifier_URIParser(); + } + + public function transform($attr, $config, $context) { + + if (!isset($attr['href'])) { + return $attr; + } + + // XXX Kind of inefficient + $url = $this->parser->parse($attr['href']); + $scheme = $url->getSchemeObj($config, $context); + + if ($scheme->browsable && !$url->isBenign($config, $context)) { + $attr['target'] = 'blank'; + } + + return $attr; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Textarea.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Textarea.php new file mode 100644 index 00000000..81ac3488 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Textarea.php @@ -0,0 +1,18 @@ + + / +class HTMLPurifier_AttrTransform_Textarea extends HTMLPurifier_AttrTransform +{ + + public function transform($attr, $config, $context) { + // Calculated from Firefox + if (!isset($attr['cols'])) $attr['cols'] = '22'; + if (!isset($attr['rows'])) $attr['rows'] = '3'; + return $attr; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTypes.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTypes.php new file mode 100644 index 00000000..6f985ff9 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTypes.php @@ -0,0 +1,91 @@ +info['Enum'] = new HTMLPurifier_AttrDef_Enum(); + $this->info['Bool'] = new HTMLPurifier_AttrDef_HTML_Bool(); + + $this->info['CDATA'] = new HTMLPurifier_AttrDef_Text(); + $this->info['ID'] = new HTMLPurifier_AttrDef_HTML_ID(); + $this->info['Length'] = new HTMLPurifier_AttrDef_HTML_Length(); + $this->info['MultiLength'] = new HTMLPurifier_AttrDef_HTML_MultiLength(); + $this->info['NMTOKENS'] = new HTMLPurifier_AttrDef_HTML_Nmtokens(); + $this->info['Pixels'] = new HTMLPurifier_AttrDef_HTML_Pixels(); + $this->info['Text'] = new HTMLPurifier_AttrDef_Text(); + $this->info['URI'] = new HTMLPurifier_AttrDef_URI(); + $this->info['LanguageCode'] = new HTMLPurifier_AttrDef_Lang(); + $this->info['Color'] = new HTMLPurifier_AttrDef_HTML_Color(); + $this->info['IAlign'] = self::makeEnum('top,middle,bottom,left,right'); + $this->info['LAlign'] = self::makeEnum('top,bottom,left,right'); + $this->info['FrameTarget'] = new HTMLPurifier_AttrDef_HTML_FrameTarget(); + + // unimplemented aliases + $this->info['ContentType'] = new HTMLPurifier_AttrDef_Text(); + $this->info['ContentTypes'] = new HTMLPurifier_AttrDef_Text(); + $this->info['Charsets'] = new HTMLPurifier_AttrDef_Text(); + $this->info['Character'] = new HTMLPurifier_AttrDef_Text(); + + // "proprietary" types + $this->info['Class'] = new HTMLPurifier_AttrDef_HTML_Class(); + + // number is really a positive integer (one or more digits) + // FIXME: ^^ not always, see start and value of list items + $this->info['Number'] = new HTMLPurifier_AttrDef_Integer(false, false, true); + } + + private static function makeEnum($in) { + return new HTMLPurifier_AttrDef_Clone(new HTMLPurifier_AttrDef_Enum(explode(',', $in))); + } + + /* + * Retrieves a type + * @param $type String type name + * @return Object AttrDef for type + / + public function get($type) { + + // determine if there is any extra info tacked on + if (strpos($type, '#') !== false) list($type, $string) = explode('#', $type, 2); + else $string = ''; + + if (!isset($this->info[$type])) { + trigger_error('Cannot retrieve undefined attribute type ' . $type, E_USER_ERROR); + return; + } + + return $this->info[$type]->make($string); + + } + + /* + * Sets a new implementation for a type + * @param $type String type name + * @param $impl Object AttrDef for type + / + public function set($type, $impl) { + $this->info[$type] = $impl; + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrValidator.php b/lib/htmlpurifier/library/HTMLPurifier/AttrValidator.php new file mode 100644 index 00000000..829a0f8f --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrValidator.php @@ -0,0 +1,162 @@ +getHTMLDefinition(); + $e =& $context->get('ErrorCollector', true); + + // initialize IDAccumulator if necessary + $ok =& $context->get('IDAccumulator', true); + if (!$ok) { + $id_accumulator = HTMLPurifier_IDAccumulator::build($config, $context); + $context->register('IDAccumulator', $id_accumulator); + } + + // initialize CurrentToken if necessary + $current_token =& $context->get('CurrentToken', true); + if (!$current_token) $context->register('CurrentToken', $token); + + if ( + !$token instanceof HTMLPurifier_Token_Start && + !$token instanceof HTMLPurifier_Token_Empty + ) return $token; + + // create alias to global definition array, see also $defs + // DEFINITION CALL + $d_defs = $definition->info_global_attr; + + // don't update token until the very end, to ensure an atomic update + $attr = $token->attr; + + // do global transformations (pre) + // nothing currently utilizes this + foreach ($definition->info_attr_transform_pre as $transform) { + $attr = $transform->transform($o = $attr, $config, $context); + if ($e) { + if ($attr != $o) $e->send(E_NOTICE, 'AttrValidator: Attributes transformed', $o, $attr); + } + } + + // do local transformations only applicable to this element (pre) + // ex. to + foreach ($definition->info[$token->name]->attr_transform_pre as $transform) { + $attr = $transform->transform($o = $attr, $config, $context); + if ($e) { + if ($attr != $o) $e->send(E_NOTICE, 'AttrValidator: Attributes transformed', $o, $attr); + } + } + + // create alias to this element's attribute definition array, see + // also $d_defs (global attribute definition array) + // DEFINITION CALL + $defs = $definition->info[$token->name]->attr; + + $attr_key = false; + $context->register('CurrentAttr', $attr_key); + + // iterate through all the attribute keypairs + // Watch out for name collisions: $key has previously been used + foreach ($attr as $attr_key => $value) { + + // call the definition + if ( isset($defs[$attr_key]) ) { + // there is a local definition defined + if ($defs[$attr_key] === false) { + // We've explicitly been told not to allow this element. + // This is usually when there's a global definition + // that must be overridden. + // Theoretically speaking, we could have a + // AttrDef_DenyAll, but this is faster! + $result = false; + } else { + // validate according to the element's definition + $result = $defs[$attr_key]->validate( + $value, $config, $context + ); + } + } elseif ( isset($d_defs[$attr_key]) ) { + // there is a global definition defined, validate according + // to the global definition + $result = $d_defs[$attr_key]->validate( + $value, $config, $context + ); + } else { + // system never heard of the attribute? DELETE! + $result = false; + } + + // put the results into effect + if ($result === false \|\| $result === null) { + // this is a generic error message that should replaced + // with more specific ones when possible + if ($e) $e->send(E_ERROR, 'AttrValidator: Attribute removed'); + + // remove the attribute + unset($attr[$attr_key]); + } elseif (is_string($result)) { + // generally, if a substitution is happening, there + // was some sort of implicit correction going on. We'll + // delegate it to the attribute classes to say exactly what. + + // simple substitution + $attr[$attr_key] = $result; + } else { + // nothing happens + } + + // we'd also want slightly more complicated substitution + // involving an array as the return value, + // although we're not sure how colliding attributes would + // resolve (certain ones would be completely overriden, + // others would prepend themselves). + } + + $context->destroy('CurrentAttr'); + + // post transforms + + // global (error reporting untested) + foreach ($definition->info_attr_transform_post as $transform) { + $attr = $transform->transform($o = $attr, $config, $context); + if ($e) { + if ($attr != $o) $e->send(E_NOTICE, 'AttrValidator: Attributes transformed', $o, $attr); + } + } + + // local (error reporting untested) + foreach ($definition->info[$token->name]->attr_transform_post as $transform) { + $attr = $transform->transform($o = $attr, $config, $context); + if ($e) { + if ($attr != $o) $e->send(E_NOTICE, 'AttrValidator: Attributes transformed', $o, $attr); + } + } + + $token->attr = $attr; + + // destroy CurrentToken if we made it ourselves + if (!$current_token) $context->destroy('CurrentToken'); + + } + + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Bootstrap.php b/lib/htmlpurifier/library/HTMLPurifier/Bootstrap.php new file mode 100644 index 00000000..607c5b18 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Bootstrap.php @@ -0,0 +1,104 @@ + +if (!defined('PHP_EOL')) { + switch (strtoupper(substr(PHP_OS, 0, 3))) { + case 'WIN': + define('PHP_EOL', "\r\n"); + break; + case 'DAR': + define('PHP_EOL', "\r"); + break; + default: + define('PHP_EOL', "\n"); + } +} + +/* + * Bootstrap class that contains meta-functionality for HTML Purifier such as + * the autoload function. + * + * @note + * This class may be used without any other files from HTML Purifier. + / +class HTMLPurifier_Bootstrap +{ + + /* + * Autoload function for HTML Purifier + * @param $class Class to load + / + public static function autoload($class) { + $file = HTMLPurifier_Bootstrap::getPath($class); + if (!$file) return false; + // Technically speaking, it should be ok and more efficient to + // just do 'require', but Antonio Parraga reports that with + // Zend extensions such as Zend debugger and APC, this invariant + // may be broken. Since we have efficient alternatives, pay + // the cost here and avoid the bug. + require_once HTMLPURIFIER_PREFIX . '/' . $file; + return true; + } + + /* + * Returns the path for a specific class. + / + public static function getPath($class) { + if (strncmp('HTMLPurifier', $class, 12) !== 0) return false; + // Custom implementations + if (strncmp('HTMLPurifier_Language_', $class, 22) === 0) { + $code = str_replace('_', '-', substr($class, 22)); + $file = 'HTMLPurifier/Language/classes/' . $code . '.php'; + } else { + $file = str_replace('_', '/', $class) . '.php'; + } + if (!file_exists(HTMLPURIFIER_PREFIX . '/' . $file)) return false; + return $file; + } + + /* + * "Pre-registers" our autoloader on the SPL stack. + / + public static function registerAutoload() { + $autoload = array('HTMLPurifier_Bootstrap', 'autoload'); + if ( ($funcs = spl_autoload_functions()) === false ) { + spl_autoload_register($autoload); + } elseif (function_exists('spl_autoload_unregister')) { + $buggy = version_compare(PHP_VERSION, '5.2.11', '<'); + $compat = version_compare(PHP_VERSION, '5.1.2', '<=') && + version_compare(PHP_VERSION, '5.1.0', '>='); + foreach ($funcs as $func) { + if ($buggy && is_array($func)) { + // :TRICKY: There are some compatibility issues and some + // places where we need to error out + $reflector = new ReflectionMethod($func[0], $func[1]); + if (!$reflector->isStatic()) { + throw new Exception(' + HTML Purifier autoloader registrar is not compatible + with non-static object methods due to PHP Bug #44144; + Please do not use HTMLPurifier.autoload.php (or any + file that includes this file); instead, place the code: + spl_autoload_register(array(\'HTMLPurifier_Bootstrap\', \'autoload\')) + after your own autoloaders. + '); + } + // Suprisingly, spl_autoload_register supports the + // Class::staticMethod callback format, although call_user_func doesn't + if ($compat) $func = implode('::', $func); + } + spl_autoload_unregister($func); + } + spl_autoload_register($autoload); + foreach ($funcs as $func) spl_autoload_register($func); + } + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/CSSDefinition.php b/lib/htmlpurifier/library/HTMLPurifier/CSSDefinition.php new file mode 100644 index 00000000..91619f5d --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/CSSDefinition.php @@ -0,0 +1,322 @@ +info['text-align'] = new HTMLPurifier_AttrDef_Enum( + array('left', 'right', 'center', 'justify'), false); + + $border_style = + $this->info['border-bottom-style'] = + $this->info['border-right-style'] = + $this->info['border-left-style'] = + $this->info['border-top-style'] = new HTMLPurifier_AttrDef_Enum( + array('none', 'hidden', 'dotted', 'dashed', 'solid', 'double', + 'groove', 'ridge', 'inset', 'outset'), false); + + $this->info['border-style'] = new HTMLPurifier_AttrDef_CSS_Multiple($border_style); + + $this->info['clear'] = new HTMLPurifier_AttrDef_Enum( + array('none', 'left', 'right', 'both'), false); + $this->info['float'] = new HTMLPurifier_AttrDef_Enum( + array('none', 'left', 'right'), false); + $this->info['font-style'] = new HTMLPurifier_AttrDef_Enum( + array('normal', 'italic', 'oblique'), false); + $this->info['font-variant'] = new HTMLPurifier_AttrDef_Enum( + array('normal', 'small-caps'), false); + + $uri_or_none = new HTMLPurifier_AttrDef_CSS_Composite( + array( + new HTMLPurifier_AttrDef_Enum(array('none')), + new HTMLPurifier_AttrDef_CSS_URI() + ) + ); + + $this->info['list-style-position'] = new HTMLPurifier_AttrDef_Enum( + array('inside', 'outside'), false); + $this->info['list-style-type'] = new HTMLPurifier_AttrDef_Enum( + array('disc', 'circle', 'square', 'decimal', 'lower-roman', + 'upper-roman', 'lower-alpha', 'upper-alpha', 'none'), false); + $this->info['list-style-image'] = $uri_or_none; + + $this->info['list-style'] = new HTMLPurifier_AttrDef_CSS_ListStyle($config); + + $this->info['text-transform'] = new HTMLPurifier_AttrDef_Enum( + array('capitalize', 'uppercase', 'lowercase', 'none'), false); + $this->info['color'] = new HTMLPurifier_AttrDef_CSS_Color(); + + $this->info['background-image'] = $uri_or_none; + $this->info['background-repeat'] = new HTMLPurifier_AttrDef_Enum( + array('repeat', 'repeat-x', 'repeat-y', 'no-repeat') + ); + $this->info['background-attachment'] = new HTMLPurifier_AttrDef_Enum( + array('scroll', 'fixed') + ); + $this->info['background-position'] = new HTMLPurifier_AttrDef_CSS_BackgroundPosition(); + + $border_color = + $this->info['border-top-color'] = + $this->info['border-bottom-color'] = + $this->info['border-left-color'] = + $this->info['border-right-color'] = + $this->info['background-color'] = new HTMLPurifier_AttrDef_CSS_Composite(array( + new HTMLPurifier_AttrDef_Enum(array('transparent')), + new HTMLPurifier_AttrDef_CSS_Color() + )); + + $this->info['background'] = new HTMLPurifier_AttrDef_CSS_Background($config); + + $this->info['border-color'] = new HTMLPurifier_AttrDef_CSS_Multiple($border_color); + + $border_width = + $this->info['border-top-width'] = + $this->info['border-bottom-width'] = + $this->info['border-left-width'] = + $this->info['border-right-width'] = new HTMLPurifier_AttrDef_CSS_Composite(array( + new HTMLPurifier_AttrDef_Enum(array('thin', 'medium', 'thick')), + new HTMLPurifier_AttrDef_CSS_Length('0') //disallow negative + )); + + $this->info['border-width'] = new HTMLPurifier_AttrDef_CSS_Multiple($border_width); + + $this->info['letter-spacing'] = new HTMLPurifier_AttrDef_CSS_Composite(array( + new HTMLPurifier_AttrDef_Enum(array('normal')), + new HTMLPurifier_AttrDef_CSS_Length() + )); + + $this->info['word-spacing'] = new HTMLPurifier_AttrDef_CSS_Composite(array( + new HTMLPurifier_AttrDef_Enum(array('normal')), + new HTMLPurifier_AttrDef_CSS_Length() + )); + + $this->info['font-size'] = new HTMLPurifier_AttrDef_CSS_Composite(array( + new HTMLPurifier_AttrDef_Enum(array('xx-small', 'x-small', + 'small', 'medium', 'large', 'x-large', 'xx-large', + 'larger', 'smaller')), + new HTMLPurifier_AttrDef_CSS_Percentage(), + new HTMLPurifier_AttrDef_CSS_Length() + )); + + $this->info['line-height'] = new HTMLPurifier_AttrDef_CSS_Composite(array( + new HTMLPurifier_AttrDef_Enum(array('normal')), + new HTMLPurifier_AttrDef_CSS_Number(true), // no negatives + new HTMLPurifier_AttrDef_CSS_Length('0'), + new HTMLPurifier_AttrDef_CSS_Percentage(true) + )); + + $margin = + $this->info['margin-top'] = + $this->info['margin-bottom'] = + $this->info['margin-left'] = + $this->info['margin-right'] = new HTMLPurifier_AttrDef_CSS_Composite(array( + new HTMLPurifier_AttrDef_CSS_Length(), + new HTMLPurifier_AttrDef_CSS_Percentage(), + new HTMLPurifier_AttrDef_Enum(array('auto')) + )); + + $this->info['margin'] = new HTMLPurifier_AttrDef_CSS_Multiple($margin); + + // non-negative + $padding = + $this->info['padding-top'] = + $this->info['padding-bottom'] = + $this->info['padding-left'] = + $this->info['padding-right'] = new HTMLPurifier_AttrDef_CSS_Composite(array( + new HTMLPurifier_AttrDef_CSS_Length('0'), + new HTMLPurifier_AttrDef_CSS_Percentage(true) + )); + + $this->info['padding'] = new HTMLPurifier_AttrDef_CSS_Multiple($padding); + + $this->info['text-indent'] = new HTMLPurifier_AttrDef_CSS_Composite(array( + new HTMLPurifier_AttrDef_CSS_Length(), + new HTMLPurifier_AttrDef_CSS_Percentage() + )); + + $trusted_wh = new HTMLPurifier_AttrDef_CSS_Composite(array( + new HTMLPurifier_AttrDef_CSS_Length('0'), + new HTMLPurifier_AttrDef_CSS_Percentage(true), + new HTMLPurifier_AttrDef_Enum(array('auto')) + )); + $max = $config->get('CSS.MaxImgLength'); + + $this->info['width'] = + $this->info['height'] = + $max === null ? + $trusted_wh : + new HTMLPurifier_AttrDef_Switch('img', + // For img tags: + new HTMLPurifier_AttrDef_CSS_Composite(array( + new HTMLPurifier_AttrDef_CSS_Length('0', $max), + new HTMLPurifier_AttrDef_Enum(array('auto')) + )), + // For everyone else: + $trusted_wh + ); + + $this->info['text-decoration'] = new HTMLPurifier_AttrDef_CSS_TextDecoration(); + + $this->info['font-family'] = new HTMLPurifier_AttrDef_CSS_FontFamily(); + + // this could use specialized code + $this->info['font-weight'] = new HTMLPurifier_AttrDef_Enum( + array('normal', 'bold', 'bolder', 'lighter', '100', '200', '300', + '400', '500', '600', '700', '800', '900'), false); + + // MUST be called after other font properties, as it references + // a CSSDefinition object + $this->info['font'] = new HTMLPurifier_AttrDef_CSS_Font($config); + + // same here + $this->info['border'] = + $this->info['border-bottom'] = + $this->info['border-top'] = + $this->info['border-left'] = + $this->info['border-right'] = new HTMLPurifier_AttrDef_CSS_Border($config); + + $this->info['border-collapse'] = new HTMLPurifier_AttrDef_Enum(array( + 'collapse', 'separate')); + + $this->info['caption-side'] = new HTMLPurifier_AttrDef_Enum(array( + 'top', 'bottom')); + + $this->info['table-layout'] = new HTMLPurifier_AttrDef_Enum(array( + 'auto', 'fixed')); + + $this->info['vertical-align'] = new HTMLPurifier_AttrDef_CSS_Composite(array( + new HTMLPurifier_AttrDef_Enum(array('baseline', 'sub', 'super', + 'top', 'text-top', 'middle', 'bottom', 'text-bottom')), + new HTMLPurifier_AttrDef_CSS_Length(), + new HTMLPurifier_AttrDef_CSS_Percentage() + )); + + $this->info['border-spacing'] = new HTMLPurifier_AttrDef_CSS_Multiple(new HTMLPurifier_AttrDef_CSS_Length(), 2); + + // partial support + $this->info['white-space'] = new HTMLPurifier_AttrDef_Enum(array('nowrap')); + + if ($config->get('CSS.Proprietary')) { + $this->doSetupProprietary($config); + } + + if ($config->get('CSS.AllowTricky')) { + $this->doSetupTricky($config); + } + + if ($config->get('CSS.Trusted')) { + $this->doSetupTrusted($config); + } + + $allow_important = $config->get('CSS.AllowImportant'); + // wrap all attr-defs with decorator that handles !important + foreach ($this->info as $k => $v) { + $this->info[$k] = new HTMLPurifier_AttrDef_CSS_ImportantDecorator($v, $allow_important); + } + + $this->setupConfigStuff($config); + } + + protected function doSetupProprietary($config) { + // Internet Explorer only scrollbar colors + $this->info['scrollbar-arrow-color'] = new HTMLPurifier_AttrDef_CSS_Color(); + $this->info['scrollbar-base-color'] = new HTMLPurifier_AttrDef_CSS_Color(); + $this->info['scrollbar-darkshadow-color'] = new HTMLPurifier_AttrDef_CSS_Color(); + $this->info['scrollbar-face-color'] = new HTMLPurifier_AttrDef_CSS_Color(); + $this->info['scrollbar-highlight-color'] = new HTMLPurifier_AttrDef_CSS_Color(); + $this->info['scrollbar-shadow-color'] = new HTMLPurifier_AttrDef_CSS_Color(); + + // technically not proprietary, but CSS3, and no one supports it + $this->info['opacity'] = new HTMLPurifier_AttrDef_CSS_AlphaValue(); + $this->info['-moz-opacity'] = new HTMLPurifier_AttrDef_CSS_AlphaValue(); + $this->info['-khtml-opacity'] = new HTMLPurifier_AttrDef_CSS_AlphaValue(); + + // only opacity, for now + $this->info['filter'] = new HTMLPurifier_AttrDef_CSS_Filter(); + + } + + protected function doSetupTricky($config) { + $this->info['display'] = new HTMLPurifier_AttrDef_Enum(array( + 'inline', 'block', 'list-item', 'run-in', 'compact', + 'marker', 'table', 'inline-table', 'table-row-group', + 'table-header-group', 'table-footer-group', 'table-row', + 'table-column-group', 'table-column', 'table-cell', 'table-caption', 'none' + )); + $this->info['visibility'] = new HTMLPurifier_AttrDef_Enum(array( + 'visible', 'hidden', 'collapse' + )); + $this->info['overflow'] = new HTMLPurifier_AttrDef_Enum(array('visible', 'hidden', 'auto', 'scroll')); + } + + protected function doSetupTrusted($config) { + $this->info['position'] = new HTMLPurifier_AttrDef_Enum(array( + 'static', 'relative', 'absolute', 'fixed' + )); + $this->info['top'] = + $this->info['left'] = + $this->info['right'] = + $this->info['bottom'] = new HTMLPurifier_AttrDef_CSS_Composite(array( + new HTMLPurifier_AttrDef_CSS_Length(), + new HTMLPurifier_AttrDef_CSS_Percentage(), + new HTMLPurifier_AttrDef_Enum(array('auto')), + )); + $this->info['z-index'] = new HTMLPurifier_AttrDef_CSS_Composite(array( + new HTMLPurifier_AttrDef_Integer(), + new HTMLPurifier_AttrDef_Enum(array('auto')), + )); + } + + /* + * Performs extra config-based processing. Based off of + * HTMLPurifier_HTMLDefinition. + * @todo Refactor duplicate elements into common class (probably using + * composition, not inheritance). + / + protected function setupConfigStuff($config) { + + // setup allowed elements + $support = "(for information on implementing this, see the ". + "support forums) "; + $allowed_properties = $config->get('CSS.AllowedProperties'); + if ($allowed_properties !== null) { + foreach ($this->info as $name => $d) { + if(!isset($allowed_properties[$name])) unset($this->info[$name]); + unset($allowed_properties[$name]); + } + // emit errors + foreach ($allowed_properties as $name => $d) { + // :TODO: Is this htmlspecialchars() call really necessary? + $name = htmlspecialchars($name); + trigger_error("Style attribute '$name' is not supported $support", E_USER_WARNING); + } + } + + $forbidden_properties = $config->get('CSS.ForbiddenProperties'); + if ($forbidden_properties !== null) { + foreach ($this->info as $name => $d) { + if (isset($forbidden_properties[$name])) { + unset($this->info[$name]); + } + } + } + + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ChildDef.php b/lib/htmlpurifier/library/HTMLPurifier/ChildDef.php new file mode 100644 index 00000000..c5d5216d --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ChildDef.php @@ -0,0 +1,48 @@ +elements; + } + + /* + * Validates nodes according to definition and returns modification. + * + * @param $tokens_of_children Array of HTMLPurifier_Token + * @param $config HTMLPurifier_Config object + * @param $context HTMLPurifier_Context object + * @return bool true to leave nodes as is + * @return bool false to remove parent node + * @return array of replacement child tokens + / + abstract public function validateChildren($tokens_of_children, $config, $context); +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ChildDef/Chameleon.php b/lib/htmlpurifier/library/HTMLPurifier/ChildDef/Chameleon.php new file mode 100644 index 00000000..15c364ee --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ChildDef/Chameleon.php @@ -0,0 +1,48 @@ +inline = new HTMLPurifier_ChildDef_Optional($inline); + $this->block = new HTMLPurifier_ChildDef_Optional($block); + $this->elements = $this->block->elements; + } + + public function validateChildren($tokens_of_children, $config, $context) { + if ($context->get('IsInline') === false) { + return $this->block->validateChildren( + $tokens_of_children, $config, $context); + } else { + return $this->inline->validateChildren( + $tokens_of_children, $config, $context); + } + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ChildDef/Custom.php b/lib/htmlpurifier/library/HTMLPurifier/ChildDef/Custom.php new file mode 100644 index 00000000..b68047b4 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ChildDef/Custom.php @@ -0,0 +1,90 @@ +dtd_regex = $dtd_regex; + $this->_compileRegex(); + } + /* + * Compiles the PCRE regex from a DTD regex ($dtd_regex to $_pcre_regex) + */ + protected function _compileRegex() { + $raw = str_replace(' ', '', $this->dtd_regex); + if ($raw{0} != '(') { + $raw = "($raw)"; + } + $el = '[#a-zA-Z0-9_.-]+'; + $reg = $raw; + + // COMPLICATED! AND MIGHT BE BUGGY! I HAVE NO CLUE WHAT I'M + // DOING! Seriously: if there's problems, please report them. + + // collect all elements into the $elements array + preg_match_all("/$el/", $reg, $matches); + foreach ($matches[0] as $match) { + $this->elements[$match] = true; + } + + // setup all elements as parentheticals with leading commas + $reg = preg_replace("/$el/", '(,\\0)', $reg); + + // remove commas when they were not solicited + $reg = preg_replace("/([^,(\|]\(+),/", '\\1', $reg); + + // remove all non-paranthetical commas: they are handled by first regex + $reg = preg_replace("/,\(/", '(', $reg); + + $this->_pcre_regex = $reg; + } + public function validateChildren($tokens_of_children, $config, $context) { + $list_of_children = ''; + $nesting = 0; // depth into the nest + foreach ($tokens_of_children as $token) { + if (!empty($token->is_whitespace)) continue; + + $is_child = ($nesting == 0); // direct + + if ($token instanceof HTMLPurifier_Token_Start) { + $nesting++; + } elseif ($token instanceof HTMLPurifier_Token_End) { + $nesting--; + } + + if ($is_child) { + $list_of_children .= $token->name . ','; + } + } + // add leading comma to deal with stray comma declarations + $list_of_children = ',' . rtrim($list_of_children, ','); + $okay = + preg_match( + '/^,?'.$this->_pcre_regex.'$/', + $list_of_children + ); + + return (bool) $okay; + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ChildDef/Empty.php b/lib/htmlpurifier/library/HTMLPurifier/ChildDef/Empty.php new file mode 100644 index 00000000..13171f66 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ChildDef/Empty.php @@ -0,0 +1,20 @@ + true, 'ul' => true, 'ol' => true); + public function validateChildren($tokens_of_children, $config, $context) { + // Flag for subclasses + $this->whitespace = false; + + // if there are no tokens, delete parent node + if (empty($tokens_of_children)) return false; + + // the new set of children + $result = array(); + + // current depth into the nest + $nesting = 0; + + // a little sanity check to make sure it's not ALL whitespace + $all_whitespace = true; + + $seen_li = false; + $need_close_li = false; + + foreach ($tokens_of_children as $token) { + if (!empty($token->is_whitespace)) { + $result[] = $token; + continue; + } + $all_whitespace = false; // phew, we're not talking about whitespace + + if ($nesting == 1 && $need_close_li) { + $result[] = new HTMLPurifier_Token_End('li'); + $nesting--; + $need_close_li = false; + } + + $is_child = ($nesting == 0); + + if ($token instanceof HTMLPurifier_Token_Start) { + $nesting++; + } elseif ($token instanceof HTMLPurifier_Token_End) { + $nesting--; + } + + if ($is_child) { + if ($token->name === 'li') { + // good + $seen_li = true; + } elseif ($token->name === 'ul' \|\| $token->name === 'ol') { + // we want to tuck this into the previous li + $need_close_li = true; + $nesting++; + if (!$seen_li) { + // create a new li element + $result[] = new HTMLPurifier_Token_Start('li'); + } else { + // backtrack until found + while(true) { + $t = array_pop($result); + if ($t instanceof HTMLPurifier_Token_End) { + // XXX actually, these invariants could very plausibly be violated + // if we are doing silly things with modifying the set of allowed elements. + // FORTUNATELY, it doesn't make a difference, since the allowed + // elements are hard-coded here! + if ($t->name !== 'li') { + trigger_error("Only li present invariant violated in List ChildDef", E_USER_ERROR); + return false; + } + break; + } elseif ($t instanceof HTMLPurifier_Token_Empty) { // bleagh + if ($t->name !== 'li') { + trigger_error("Only li present invariant violated in List ChildDef", E_USER_ERROR); + return false; + } + // XXX this should have a helper for it... + $result[] = new HTMLPurifier_Token_Start('li', $t->attr, $t->line, $t->col, $t->armor); + break; + } else { + if (!$t->is_whitespace) { + trigger_error("Only whitespace present invariant violated in List ChildDef", E_USER_ERROR); + return false; + } + } + } + } + } else { + // start wrapping (this doesn't precisely mimic + // browser behavior, but what browsers do is kind of + // hard to mimic in a standards compliant way + // XXX Actually, this has no impact in practice, + // because this gets handled earlier. Arguably, + // we should rip out all of that processing + $result[] = new HTMLPurifier_Token_Start('li'); + $nesting++; + $seen_li = true; + $need_close_li = true; + } + } + $result[] = $token; + } + if ($need_close_li) { + $result[] = new HTMLPurifier_Token_End('li'); + } + if (empty($result)) return false; + if ($all_whitespace) { + return false; + } + if ($tokens_of_children == $result) return true; + return $result; + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ChildDef/Optional.php b/lib/htmlpurifier/library/HTMLPurifier/ChildDef/Optional.php new file mode 100644 index 00000000..32bcb989 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ChildDef/Optional.php @@ -0,0 +1,26 @@ +whitespace) return $tokens_of_children; + else return array(); + } + return $result; + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ChildDef/Required.php b/lib/htmlpurifier/library/HTMLPurifier/ChildDef/Required.php new file mode 100644 index 00000000..4889f249 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ChildDef/Required.php @@ -0,0 +1,117 @@ + $x) { + $elements[$i] = true; + if (empty($i)) unset($elements[$i]); // remove blank + } + } + $this->elements = $elements; + } + public $allow_empty = false; + public $type = 'required'; + public function validateChildren($tokens_of_children, $config, $context) { + // Flag for subclasses + $this->whitespace = false; + + // if there are no tokens, delete parent node + if (empty($tokens_of_children)) return false; + + // the new set of children + $result = array(); + + // current depth into the nest + $nesting = 0; + + // whether or not we're deleting a node + $is_deleting = false; + + // whether or not parsed character data is allowed + // this controls whether or not we silently drop a tag + // or generate escaped HTML from it + $pcdata_allowed = isset($this->elements['#PCDATA']); + + // a little sanity check to make sure it's not ALL whitespace + $all_whitespace = true; + + // some configuration + $escape_invalid_children = $config->get('Core.EscapeInvalidChildren'); + + // generator + $gen = new HTMLPurifier_Generator($config, $context); + + foreach ($tokens_of_children as $token) { + if (!empty($token->is_whitespace)) { + $result[] = $token; + continue; + } + $all_whitespace = false; // phew, we're not talking about whitespace + + $is_child = ($nesting == 0); + + if ($token instanceof HTMLPurifier_Token_Start) { + $nesting++; + } elseif ($token instanceof HTMLPurifier_Token_End) { + $nesting--; + } + + if ($is_child) { + $is_deleting = false; + if (!isset($this->elements[$token->name])) { + $is_deleting = true; + if ($pcdata_allowed && $token instanceof HTMLPurifier_Token_Text) { + $result[] = $token; + } elseif ($pcdata_allowed && $escape_invalid_children) { + $result[] = new HTMLPurifier_Token_Text( + $gen->generateFromToken($token) + ); + } + continue; + } + } + if (!$is_deleting \|\| ($pcdata_allowed && $token instanceof HTMLPurifier_Token_Text)) { + $result[] = $token; + } elseif ($pcdata_allowed && $escape_invalid_children) { + $result[] = + new HTMLPurifier_Token_Text( + $gen->generateFromToken($token) + ); + } else { + // drop silently + } + } + if (empty($result)) return false; + if ($all_whitespace) { + $this->whitespace = true; + return false; + } + if ($tokens_of_children == $result) return true; + return $result; + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ChildDef/StrictBlockquote.php b/lib/htmlpurifier/library/HTMLPurifier/ChildDef/StrictBlockquote.php new file mode 100644 index 00000000..dfae8a6e --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ChildDef/StrictBlockquote.php @@ -0,0 +1,88 @@ +init($config); + return $this->fake_elements; + } + + public function validateChildren($tokens_of_children, $config, $context) { + + $this->init($config); + + // trick the parent class into thinking it allows more + $this->elements = $this->fake_elements; + $result = parent::validateChildren($tokens_of_children, $config, $context); + $this->elements = $this->real_elements; + + if ($result === false) return array(); + if ($result === true) $result = $tokens_of_children; + + $def = $config->getHTMLDefinition(); + $block_wrap_start = new HTMLPurifier_Token_Start($def->info_block_wrapper); + $block_wrap_end = new HTMLPurifier_Token_End( $def->info_block_wrapper); + $is_inline = false; + $depth = 0; + $ret = array(); + + // assuming that there are no comment tokens + foreach ($result as $i => $token) { + $token = $result[$i]; + // ifs are nested for readability + if (!$is_inline) { + if (!$depth) { + if ( + ($token instanceof HTMLPurifier_Token_Text && !$token->is_whitespace) \|\| + (!$token instanceof HTMLPurifier_Token_Text && !isset($this->elements[$token->name])) + ) { + $is_inline = true; + $ret[] = $block_wrap_start; + } + } + } else { + if (!$depth) { + // starting tokens have been inline text / empty + if ($token instanceof HTMLPurifier_Token_Start \|\| $token instanceof HTMLPurifier_Token_Empty) { + if (isset($this->elements[$token->name])) { + // ended + $ret[] = $block_wrap_end; + $is_inline = false; + } + } + } + } + $ret[] = $token; + if ($token instanceof HTMLPurifier_Token_Start) $depth++; + if ($token instanceof HTMLPurifier_Token_End) $depth--; + } + if ($is_inline) $ret[] = $block_wrap_end; + return $ret; + } + + private function init($config) { + if (!$this->init) { + $def = $config->getHTMLDefinition(); + // allow all inline elements + $this->real_elements = $this->elements; + $this->fake_elements = $def->info_content_sets['Flow']; + $this->fake_elements['#PCDATA'] = true; + $this->init = true; + } + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ChildDef/Table.php b/lib/htmlpurifier/library/HTMLPurifier/ChildDef/Table.php new file mode 100644 index 00000000..9a93421a --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ChildDef/Table.php @@ -0,0 +1,227 @@ + true, 'tbody' => true, 'thead' => true, + 'tfoot' => true, 'caption' => true, 'colgroup' => true, 'col' => true); + public function __construct() {} + public function validateChildren($tokens_of_children, $config, $context) { + if (empty($tokens_of_children)) return false; + + // this ensures that the loop gets run one last time before closing + // up. It's a little bit of a hack, but it works! Just make sure you + // get rid of the token later. + $tokens_of_children[] = false; + + // only one of these elements is allowed in a table + $caption = false; + $thead = false; + $tfoot = false; + + // as many of these as you want + $cols = array(); + $content = array(); + + $nesting = 0; // current depth so we can determine nodes + $is_collecting = false; // are we globbing together tokens to package + // into one of the collectors? + $collection = array(); // collected nodes + $tag_index = 0; // the first node might be whitespace, + // so this tells us where the start tag is + $tbody_mode = false; // if true, then we need to wrap any stray + //

Key	Description	Example `<a href="">`
%r	1 - The URI embeds a resource (blank) - The URI is merely a link
%n	The name of the tag this URI came from	a
%m	The name of the attribute this URI came from	href
%p	The name of the CSS property this URI came from, or blank if irrelevant