add text_languagedetect to guess article language for better hyphenation

[tt-rss.git] / include / rssfuncs.php

diff --git a/include/rssfuncs.php b/include/rssfuncs.php
index cfb0e7a46e2dcb8ba21a51f86965ffc6e8d51ba4..756ecbfc1d5fcfc22c26970119bd24e253c7e2fc 100644 (file)
--- a/include/rssfuncs.php
+++ b/include/rssfuncs.php
@@ -354,6 +354,11 @@
                        $rss->init();
                }
 
+               require_once "lib/languagedetect/LanguageDetect.php";
+
+               $lang = new Text_LanguageDetect();
+               $lang->setNameMode(2);
+
 //             print_r($rss);
 
                $feed = db_escape_string($feed);
@@ -565,6 +570,15 @@
                                        print "\n";
                                }
 
+                               $entry_language = $lang->detect($entry_content, 1);
+
+                               if (count($entry_language) > 0) {
+                                       $entry_language = array_keys($entry_language);
+                                       $entry_language = db_escape_string($entry_language[0]);
+
+                                       _debug("detected language: $entry_language", $debug_enabled);
+                               }
+
                                $entry_comments = $item->get_comments_url();
                                $entry_author = $item->get_author();
 
@@ -677,6 +691,7 @@
                                                        comments,
                                                        num_comments,
                                                        plugin_data,
+                                                       lang,
                                                        author)
                                                VALUES
                                                        ('$entry_title',
@@ -691,6 +706,7 @@
                                                        '$entry_comments',
                                                        '$num_comments',
                                                        '$entry_plugin_data',
+                                                       '$entry_language',
                                                        '$entry_author')");
 
                                        $article_labels = array();