X-Git-Url: https://git.mxchange.org/?a=blobdiff_plain;f=langfilter%2Flangfilter.php;h=6922ac82387cf5a1de16f62b93963098be149e49;hb=c9a8974165c34bb70874d80a2639b46baf533ffc;hp=e569ed37db8646277f540d634be95a58d0e449a6;hpb=c6e5bc3429c9c610895ba826d841cddd8c0f378b;p=friendica-addons.git diff --git a/langfilter/langfilter.php b/langfilter/langfilter.php index e569ed37..6922ac82 100644 --- a/langfilter/langfilter.php +++ b/langfilter/langfilter.php @@ -1,5 +1,4 @@ t("Language Filter"), - '$intro' => t('This addon tries to identify the language of a postings. If it does not match any language spoken by you (see below) the posting will be collapsed. Remember detecting the language is not perfect, especially with short postings.'), - '$enabled' => ['langfilter_enable', t('Use the language filter'), $enable_checked, ''], - '$languages' => ['langfilter_languages', t('I speak'), $languages, t('List of abbreviations (iso2 codes) for languages you speak, comma separated. For example "de,it".')], - '$minconfidence' => ['langfilter_minconfidence', t('Minimum confidence in language detection'), $minconfidence, t('Minimum confidence in language detection being correct, from 0 to 100. Posts will not be filtered when the confidence of language detection is below this percent value.')], - '$minlength' => ['langfilter_minlength', t('Minimum length of message body'), $minlength, t('Minimum length of message body for language filter to be used. Posts shorter than this number of characters will not be filtered.')], - '$submit' => t('Save Settings'), + '$title' => L10n::t("Language Filter"), + '$intro' => L10n::t('This addon tries to identify the language posts are writen in. If it does not match any language specifed below, posts will be hidden by collapsing them.'), + '$enabled' => ['langfilter_enable', L10n::t('Use the language filter'), $enable_checked, ''], + '$languages' => ['langfilter_languages', L10n::t('Able to read'), $languages, L10n::t('List of abbreviations (iso2 codes) for languages you speak, comma separated. For example "de,it".')], + '$minconfidence' => ['langfilter_minconfidence', L10n::t('Minimum confidence in language detection'), $minconfidence, L10n::t('Minimum confidence in language detection being correct, from 0 to 100. Posts will not be filtered when the confidence of language detection is below this percent value.')], + '$minlength' => ['langfilter_minlength', L10n::t('Minimum length of message body'), $minlength, L10n::t('Minimum number of characters in message body for filter to be used. Posts shorter than this will not be filtered. Note: Language detection is unreliable for short content (<200 characters).')], + '$submit' => L10n::t('Save Settings'), ]); return; @@ -78,9 +76,9 @@ function langfilter_addon_settings_post(App $a, &$b) return; } - if ($_POST['langfilter-settings-submit']) { + if (!empty($_POST['langfilter-settings-submit'])) { PConfig::set(local_user(), 'langfilter', 'languages', trim($_POST['langfilter_languages'])); - $enable = ((x($_POST, 'langfilter_enable')) ? intval($_POST['langfilter_enable']) : 0); + $enable = (x($_POST, 'langfilter_enable') ? intval($_POST['langfilter_enable']) : 0); $disable = 1 - $enable; PConfig::set(local_user(), 'langfilter', 'disable', $disable); $minconfidence = 0 + $_POST['langfilter_minconfidence']; @@ -96,12 +94,12 @@ function langfilter_addon_settings_post(App $a, &$b) $minlength = 0 + $_POST['langfilter_minlength']; if (!$minlength) { $minlength = 32; - } elseif ($minlengt8h < 0) { + } elseif ($minlength < 0) { $minlength = 32; } PConfig::set(local_user(), 'langfilter', 'minlength', $minlength); - info(t('Language Filter Settings saved.') . EOL); + info(L10n::t('Language Filter Settings saved.') . EOL); } } @@ -114,7 +112,7 @@ function langfilter_addon_settings_post(App $a, &$b) * expand it again. */ -function langfilter_prepare_body(App $a, &$b) +function langfilter_prepare_body_content_filter(App $a, &$hook_data) { $logged_user = local_user(); if (!$logged_user) { @@ -123,8 +121,8 @@ function langfilter_prepare_body(App $a, &$b) // Never filter own messages // TODO: find a better way to extract this - $logged_user_profile = $a->get_baseurl() . '/profile/' . $a->user['nickname']; - if ($logged_user_profile == $b['item']['author-link']) { + $logged_user_profile = $a->getBaseURL() . '/profile/' . $a->user['nickname']; + if ($logged_user_profile == $hook_data['item']['author-link']) { return; } @@ -133,51 +131,71 @@ function langfilter_prepare_body(App $a, &$b) return; } + $naked_body = BBCode::toPlaintext($hook_data['item']['body'], false); + // Don't filter if body lenght is below minimum - $minlen = PConfig::get(local_user(), 'langfilter', 'minlength'); + $minlen = PConfig::get(local_user(), 'langfilter', 'minlength', 32); if (!$minlen) { $minlen = 32; } - if (strlen($b['item']['body']) < $minlen) { + + if (strlen($naked_body) < $minlen) { return; } - $spoken_config = PConfig::get(local_user(), 'langfilter', 'languages'); + $read_languages_string = PConfig::get(local_user(), 'langfilter', 'languages'); $minconfidence = PConfig::get(local_user(), 'langfilter', 'minconfidence'); // Don't filter if no spoken languages are configured - if (!$spoken_config) + if (!$read_languages_string) { return; - $spoken_languages = explode(',', $spoken_config); + } + $read_languages_array = explode(',', $read_languages_string); // Extract the language of the post - $opts = $b['item']['postopts']; - if (!$opts) { - // no options associated to post - return; - } - if (!preg_match('/\blang=([^;]*);([^:]*)/', $opts, $matches)) { - // no lang options associated to post - return; - } + if (!empty($hook_data['item']['language'])) { + $languages = json_decode($hook_data['item']['language'], true); + if (!is_array($languages)) { + return; + } - $lang = $matches[1]; - $confidence = $matches[2]; + foreach ($languages as $iso2 => $confidence) { + break; + } + + if (empty($iso2)) { + return; + } + + $lang = Text_LanguageDetect_ISO639::code2ToName($iso2); + } else { + $opts = $hook_data['item']['postopts']; + if (!$opts) { + // no options associated to post + return; + } + + if (!preg_match('/\blang=([^;]*);([^:]*)/', $opts, $matches)) { + // no lang options associated to post + return; + } + + $lang = $matches[1]; + $confidence = $matches[2]; + + $iso2 = Text_LanguageDetect_ISO639::nameToCode2($lang); + } // Do not filter if language detection confidence is too low if ($minconfidence && $confidence < $minconfidence) { return; } - $iso2 = Text_LanguageDetect_ISO639::nameToCode2($lang); - if (!$iso2) { return; } - $spoken = in_array($iso2, $spoken_languages); - if (!$spoken) { - $rnd = random_string(8); - $b['html'] = ''; + if (!in_array($iso2, $read_languages_array)) { + $hook_data['filter_reasons'][] = L10n::t('Filtered language: %s', ucfirst($lang)); } }