/*
- * $XFree86: xc/lib/fontconfig/src/fclang.c,v 1.4 2002/07/12 21:06:03 keithp Exp $
+ * $RCSId: xc/lib/fontconfig/src/fclang.c,v 1.7 2002/08/26 23:34:31 keithp Exp $
*
- * Copyright © 2002 Keith Packard, member of The XFree86 Project, Inc.
+ * Copyright © 2002 Keith Packard
*
* Permission to use, copy, modify, distribute, and sell this software and its
* documentation for any purpose is hereby granted without fee, provided that
#include "fcint.h"
typedef struct {
- FcChar8 *lang;
- FcCharSet charset;
+ const FcChar8 *lang;
+ const FcCharSet charset;
} FcLangCharSet;
-#include "../fc-lang/fclang.h"
+typedef struct {
+ int begin;
+ int end;
+} FcLangCharSetRange;
-#define NUM_LANG_CHAR_SET (sizeof (fcLangCharSets) / sizeof (fcLangCharSets[0]))
-#define NUM_LANG_SET_MAP ((NUM_LANG_CHAR_SET + 31) / 32)
+#include "../fc-lang/fclang.h"
struct _FcLangSet {
FcChar32 map[NUM_LANG_SET_MAP];
FcFreeTypeLangSet (const FcCharSet *charset,
const FcChar8 *exclusiveLang)
{
- int i;
+ int i, j;
FcChar32 missing;
const FcCharSet *exclusiveCharset = 0;
FcLangSet *ls;
* not support other Han languages
*/
if (exclusiveCharset &&
- FcFreeTypeIsExclusiveLang (fcLangCharSets[i].lang) &&
- fcLangCharSets[i].charset.leaves != exclusiveCharset->leaves)
+ FcFreeTypeIsExclusiveLang (fcLangCharSets[i].lang))
{
- continue;
+ if (fcLangCharSets[i].charset.num != exclusiveCharset->num)
+ continue;
+
+ for (j = 0; j < fcLangCharSets[i].charset.num; j++)
+ if (FcCharSetGetLeaf(&fcLangCharSets[i].charset, j) !=
+ FcCharSetGetLeaf(exclusiveCharset, j))
+ continue;
}
missing = FcCharSetSubtractCount (&fcLangCharSets[i].charset, charset);
if (FcDebug() & FC_DBG_SCANV)
}
}
+/*
+ * Return FcTrue when super contains sub.
+ *
+ * super contains sub if super and sub have the same
+ * language and either the same country or one
+ * is missing the country
+ */
+
+static FcBool
+FcLangContains (const FcChar8 *super, const FcChar8 *sub)
+{
+ FcChar8 c1, c2;
+
+ for (;;)
+ {
+ c1 = *super++;
+ c2 = *sub++;
+
+ c1 = FcToLower (c1);
+ c2 = FcToLower (c2);
+ if (c1 != c2)
+ {
+ /* see if super has a country while sub is mising one */
+ if (c1 == '-' && c2 == '\0')
+ return FcTrue;
+ /* see if sub has a country while super is mising one */
+ if (c1 == '\0' && c2 == '-')
+ return FcTrue;
+ return FcFalse;
+ }
+ else if (!c1)
+ return FcTrue;
+ }
+}
+
const FcCharSet *
FcCharSetForLang (const FcChar8 *lang)
{
static int
FcLangSetIndex (const FcChar8 *lang)
{
- int low, high, mid;
- int cmp;
+ int low, high, mid = 0;
+ int cmp = 0;
+ FcChar8 firstChar = FcToLower(lang[0]);
+ FcChar8 secondChar = firstChar ? FcToLower(lang[1]) : '\0';
+
+ if (firstChar < 'a')
+ {
+ low = 0;
+ high = fcLangCharSetRanges[0].begin;
+ }
+ else if(firstChar > 'z')
+ {
+ low = fcLangCharSetRanges[25].begin;
+ high = NUM_LANG_CHAR_SET - 1;
+ }
+ else
+ {
+ low = fcLangCharSetRanges[firstChar - 'a'].begin;
+ high = fcLangCharSetRanges[firstChar - 'a'].end;
+ /* no matches */
+ if (low > high)
+ return -low; /* next entry after where it would be */
+ }
- low = 0;
- high = NUM_LANG_CHAR_SET - 1;
while (low <= high)
{
mid = (high + low) >> 1;
- cmp = FcStrCmpIgnoreCase (fcLangCharSets[mid].lang, lang);
- if (cmp == 0)
+ if(fcLangCharSets[mid].lang[0] != firstChar)
+ cmp = FcStrCmpIgnoreCase(fcLangCharSets[mid].lang, lang);
+ else
+ { /* fast path for resolving 2-letter languages (by far the most common) after
+ * finding the first char (probably already true because of the hash table) */
+ cmp = fcLangCharSets[mid].lang[1] - secondChar;
+ if (cmp == 0 &&
+ (fcLangCharSets[mid].lang[2] != '\0' ||
+ lang[2] != '\0'))
+ {
+ cmp = FcStrCmpIgnoreCase(fcLangCharSets[mid].lang+2,
+ lang+2);
+ }
+ }
+ if (cmp == 0)
return mid;
if (cmp < 0)
low = mid + 1;
int i;
id = FcLangSetIndex (lang);
- if (id >= 0)
+ if (id < 0)
+ id = -id - 1;
+ else if (FcLangSetBitGet (ls, id))
return FcLangEqual;
- id = -id - 1;
best = FcLangDifferentLang;
for (i = id - 1; i >= 0; i--)
{
FcLangResult
FcLangSetCompare (const FcLangSet *lsa, const FcLangSet *lsb)
{
- int i;
+ int i, j;
FcLangResult best, r;
for (i = 0; i < NUM_LANG_SET_MAP; i++)
if (lsa->map[i] & lsb->map[i])
return FcLangEqual;
best = FcLangDifferentLang;
+ for (j = 0; j < NUM_COUNTRY_SET; j++)
+ for (i = 0; i < NUM_LANG_SET_MAP; i++)
+ if ((lsa->map[i] & fcLangCountrySets[j][i]) &&
+ (lsb->map[i] & fcLangCountrySets[j][i]))
+ {
+ best = FcLangDifferentCountry;
+ break;
+ }
if (lsa->extra)
{
r = FcLangSetCompareStrSet (lsb, lsa->extra);
strs.num = 1;
strs.size = 1;
strs.strs = &str;
+ strs.ref = 1;
str = (FcChar8 *) lang;
}
return &ls;
FcLangSet *
FcNameParseLangSet (const FcChar8 *string)
{
- FcChar8 lang[32];
- const FcChar8 *end, *next;
+ FcChar8 lang[32],c;
+ int i;
FcLangSet *ls;
ls = FcLangSetCreate ();
if (!ls)
goto bail0;
- while (string && *string)
+ for(;;)
{
- end = (FcChar8 *) strchr ((char *) string, '|');
- if (!end)
+ for(i = 0; i < 31;i++)
{
- end = string + strlen ((char *) string);
- next = end;
- }
- else
- next = end + 1;
- if (end - string < sizeof (lang) - 1)
- {
- strncpy ((char *) lang, (char *) string, end - string);
- lang[end-string] = '\0';
- if (!FcLangSetAdd (ls, lang))
- goto bail1;
+ c = *string++;
+ if(c == '\0' || c == '|')
+ break; /* end of this code */
+ lang[i] = c;
}
- string = next;
+ lang[i] = '\0';
+ if (!FcLangSetAdd (ls, lang))
+ goto bail1;
+ if(c == '\0')
+ break;
}
return ls;
bail1:
if (!first)
if (!FcStrBufChar (buf, '|'))
return FcFalse;
- if (!FcStrBufString (buf, extra));
+ if (!FcStrBufString (buf, extra))
return FcFalse;
first = FcFalse;
}
return FcStrSetEqual (lsa->extra, lsb->extra);
return FcFalse;
}
+
+static FcBool
+FcLangSetContainsLang (const FcLangSet *ls, const FcChar8 *lang)
+{
+ int id;
+ int i;
+
+ id = FcLangSetIndex (lang);
+ if (id < 0)
+ id = -id - 1;
+ else if (FcLangSetBitGet (ls, id))
+ return FcTrue;
+ /*
+ * search up and down among equal languages for a match
+ */
+ for (i = id - 1; i >= 0; i--)
+ {
+ if (FcLangCompare (fcLangCharSets[i].lang, lang) == FcLangDifferentLang)
+ break;
+ if (FcLangSetBitGet (ls, i) &&
+ FcLangContains (fcLangCharSets[i].lang, lang))
+ return FcTrue;
+ }
+ for (i = id; i < NUM_LANG_CHAR_SET; i++)
+ {
+ if (FcLangCompare (fcLangCharSets[i].lang, lang) == FcLangDifferentLang)
+ break;
+ if (FcLangSetBitGet (ls, i) &&
+ FcLangContains (fcLangCharSets[i].lang, lang))
+ return FcTrue;
+ }
+ if (ls->extra)
+ {
+ FcStrList *list = FcStrListCreate (ls->extra);
+ FcChar8 *extra;
+
+ if (list)
+ {
+ while ((extra = FcStrListNext (list)))
+ {
+ if (FcLangContains (extra, lang))
+ break;
+ }
+ FcStrListDone (list);
+ if (extra)
+ return FcTrue;
+ }
+ }
+ return FcFalse;
+}
+
+/*
+ * return FcTrue if lsa contains every language in lsb
+ */
+FcBool
+FcLangSetContains (const FcLangSet *lsa, const FcLangSet *lsb)
+{
+ int i, j;
+ FcChar32 missing;
+
+ if (FcDebug() & FC_DBG_MATCHV)
+ {
+ printf ("FcLangSet "); FcLangSetPrint (lsa);
+ printf (" contains "); FcLangSetPrint (lsb);
+ printf ("\n");
+ }
+ /*
+ * check bitmaps for missing language support
+ */
+ for (i = 0; i < NUM_LANG_SET_MAP; i++)
+ {
+ missing = lsb->map[i] & ~lsa->map[i];
+ if (missing)
+ {
+ for (j = 0; j < 32; j++)
+ if (missing & (1 << j))
+ {
+ if (!FcLangSetContainsLang (lsa,
+ fcLangCharSets[i*32 + j].lang))
+ {
+ if (FcDebug() & FC_DBG_MATCHV)
+ printf ("\tMissing bitmap %s\n", fcLangCharSets[i*32+j].lang);
+ return FcFalse;
+ }
+ }
+ }
+ }
+ if (lsb->extra)
+ {
+ FcStrList *list = FcStrListCreate (lsb->extra);
+ FcChar8 *extra;
+
+ if (list)
+ {
+ while ((extra = FcStrListNext (list)))
+ {
+ if (!FcLangSetContainsLang (lsa, extra))
+ {
+ if (FcDebug() & FC_DBG_MATCHV)
+ printf ("\tMissing string %s\n", extra);
+ break;
+ }
+ }
+ FcStrListDone (list);
+ if (extra)
+ return FcFalse;
+ }
+ }
+ return FcTrue;
+}
+
+static FcLangSet ** langsets = 0;
+static int langset_bank_count = 0, langset_ptr = 0, langset_count = 0;
+
+void
+FcLangSetNewBank (void)
+{
+ langset_count = 0;
+}
+
+/* ideally, should only write one copy of any particular FcLangSet */
+int
+FcLangSetNeededBytes (const FcLangSet *l)
+{
+ langset_count++;
+ return sizeof (FcLangSet);
+}
+
+static FcBool
+FcLangSetEnsureBank (int bi)
+{
+ if (!langsets || bi >= langset_bank_count)
+ {
+ int new_count = langset_bank_count + 2;
+ int i;
+ FcLangSet** tt;
+ tt = realloc(langsets, new_count * sizeof(FcLangSet *));
+ if (!tt)
+ return FcFalse;
+
+ langsets = tt;
+ for (i = langset_bank_count; i < new_count; i++)
+ langsets[i] = 0;
+ langset_bank_count = new_count;
+ }
+
+ return FcTrue;
+}
+
+void *
+FcLangSetDistributeBytes (FcCache * metadata, void * block_ptr)
+{
+ int bi = FcCacheBankToIndex(metadata->bank);
+ if (!FcLangSetEnsureBank(bi))
+ return 0;
+
+ langsets[bi] = block_ptr;
+ block_ptr = (void *)((char *)block_ptr +
+ langset_count * sizeof(FcLangSet));
+ langset_ptr = 0;
+ metadata->langset_count = langset_count;
+ return block_ptr;
+}
+
+FcLangSet *
+FcLangSetSerialize(int bank, FcLangSet *l)
+{
+ int p = langset_ptr, bi = FcCacheBankToIndex(bank);
+
+ if (!l) return 0;
+
+ langsets[bi][langset_ptr] = *l;
+ langsets[bi][langset_ptr].extra = 0;
+ langset_ptr++;
+ return &langsets[bi][p];
+}
+
+void *
+FcLangSetUnserialize (FcCache metadata, void *block_ptr)
+{
+ int bi = FcCacheBankToIndex(metadata.bank);
+ if (!FcLangSetEnsureBank(bi))
+ return 0;
+
+ FcMemAlloc (FC_MEM_LANGSET, metadata.langset_count * sizeof(FcLangSet));
+ langsets[bi] = (FcLangSet *)block_ptr;
+ block_ptr = (void *)((char *)block_ptr +
+ metadata.langset_count * sizeof(FcLangSet));
+ return block_ptr;
+}
git.wh0rd.org / fontconfig.git / blobdiff