Merge pull request #13483 from annando/languages

Additional languages / new hook for additional language detection

doc/Addons.md

@@ -221,6 +221,13 @@ Please note: body contents are bbcode - not HTML
 Called when receiving a post from another source. This may also be used to post local activity or system generated messages.
 `$b` is the item array of information to be stored in the database and the item body is bbcode.
 
+### detect_languages
+Called after the language detection. This can be used for alternative language detection methods.
+`$data` is an array:
+
+- **text**: The text that is analyzed.
+- **detected**: (input/output) Array of language codes detected in the related text.
+
 ### addon_settings
 Called when generating the HTML for the addon settings page.
 `$data` is an array containing:
@@ -800,6 +807,7 @@ Here is a complete list of all hook callbacks with file locations (as of 24-Sep-
 
 ### src/Model/Item.php
 
+    Hook::callAll('detect_languages', $item);
     Hook::callAll('post_local', $item);
     Hook::callAll('post_remote', $item);
     Hook::callAll('post_local_end', $posted_item);

doc/de/Addons.md

@@ -103,6 +103,13 @@ Derzeitige Hooks
     $b ist das Item-Array einer Information, die in der Datenbank und im Item gespeichert ist.
         {Bitte beachte: der Seiteninhalt ist bbcode - nicht HTML)
 
+**'detect_languages'**
+Wird nach der Sprachenerkennung aufgerufen.
+Dieser Hook kann dafür verwendet werden, alternative Erkennungsfunktionen einzubinden.
+`$data` ist ein Array:
+        'text' => Der analysierte Text.
+        'detected' => (Eingabe/Ausgabe) Das Array mit den erkannten Sprachen.
+
 **'addon_settings'** - wird aufgerufen, wenn die HTML-Ausgabe der Addon-Einstellungsseite generiert wird.
     $b ist die HTML-Ausgabe (String) der Addon-Einstellungsseite vor dem finalen "</form>"-Tag.
 
@@ -316,6 +323,7 @@ Eine komplette Liste aller Hook-Callbacks mit den zugehörigen Dateien (am 01-Ap
 
 ### src/Model/Item.php
 
+    Hook::callAll('detect_languages', $item);
     Hook::callAll('post_local', $item);
     Hook::callAll('post_remote', $item);
     Hook::callAll('post_local_end', $posted_item);

src/Core/L10n.php

@@ -397,13 +397,31 @@ public function getAvailableLanguages(bool $additional = false): array
 				// See https://github.com/friendica/friendica/issues/10511
 				// Persian is manually added to language detection until a persian translation is provided for the interface, at
 				// which point it will be automatically available through `getAvailableLanguages()` and this should be removed.
-				// Additionally Portuguese, Ukrainian, traditional Chinese and Welsh are added to that list.
+				// Additionally some more languages are added to that list that are used in the Fediverse.
 				$additional_langs = [
-					'cy'      => 'Cymraeg',
-					'uk'      => 'Українська',
-					'pt-PT'   => 'Português',
-					'zh-hant' => '繁體',
-					'fa'      => 'فارسی'
+					'af'         => 'Afrikaans',
+					'cy'         => 'Cymraeg',
+					'el-monoton' => 'Ελληνικά',
+					'eu'         => 'euskara',
+					'fa'         => 'فارسی',
+					'gl'         => 'Galego',
+					'hi'         => 'हिन्दी',
+					'hr'         => 'Hrvatski',
+					'id'         => 'bahasa Indonesia',
+					'ko'         => '한국인',
+					'lt'         => 'lietuvių',
+					'lv'         => 'latviešu',
+					'sk'         => 'slovenský',
+					'sl'         => 'Slovenščina',
+					'sw'         => 'Kiswahili',
+					'th'         => 'แบบไทย',
+					'tl'         => 'Wikang Tagalog',
+					'tr'         => 'Türkçe',
+					'pt-PT'      => 'Português',
+					'uk'         => 'Українська',
+					'uz'         => 'Ўзбек',
+					'vi'         => 'Tiếng Việt',
+					'zh-hant'    => '繁體',
 				];
 				$langs = array_merge($additional_langs, $langs);
 				ksort($langs);

src/Model/Item.php

@@ -2017,7 +2017,7 @@ public static function getLanguageArray(string $body, int $count): array
 		$naked_body = BBCode::toPlaintext($naked_body);
 
 		// Remove possibly remaining links
-		$naked_body = preg_replace(Strings::autoLinkRegEx(), '', $naked_body);
+		$naked_body = trim(preg_replace(Strings::autoLinkRegEx(), '', $naked_body));
 
 		if (empty($naked_body)) {
 			return [];
@@ -2029,7 +2029,17 @@ public static function getLanguageArray(string $body, int $count): array
 		$availableLanguages = DI::l10n()->convertForLanguageDetection($availableLanguages);
 
 		$ld = new Language(array_keys($availableLanguages));
-		return $ld->detect($naked_body)->limit(0, $count)->close() ?: [];
+		$languages = $ld->detect($naked_body)->limit(0, $count)->close() ?: [];
+
+		$data = [
+			'text'     => $naked_body,
+			'detected' => $languages,
+		];
+
+		Hook::callAll('detect_languages', $data);
+		$languages = $data['detected'];
+
+		return $languages;
 	}
 
 	/**