public function warning_length($original, $translation, $locale)
{
if (in_array($locale->slug, $this->length_exclude_languages)) {
return true;
}
if (gp_startswith($original, 'number_format_')) {
return true;
}
$len_src = gp_strlen($original);
$len_trans = gp_strlen($translation);
if (!($this->length_lower_bound * $len_src < $len_trans && $len_trans < $this->length_upper_bound * $len_src) && (!gp_in('_abbreviation', $original) && !gp_in('_initial', $original))) {
return __('Lengths of source and translation differ too much.', 'glotpress');
}
return true;
}
function map_glossary_entries_to_translations_originals($translations, $glossary)
{
$glossary_entries = GP::$glossary_entry->by_glossary_id($glossary->id);
if (empty($glossary_entries)) {
return $translations;
}
$glossary_entries_terms = array();
//Create array of glossary terms, longest first
foreach ($glossary_entries as $key => $value) {
$glossary_entries_terms[$key] = $value->term;
}
uasort($glossary_entries_terms, function ($a, $b) {
return gp_strlen($a) < gp_strlen($b);
});
foreach ($translations as $key => $t) {
//Save our current singular/plural strings before attempting any markup change. Also escape now, since we're going to add some html.
$translations[$key]->singular_glossary_markup = esc_translation($t->singular);
$translations[$key]->plural_glossary_markup = esc_translation($t->plural);
//Search for glossary terms in our strings
$matching_entries = array();
foreach ($glossary_entries_terms as $i => $term) {
$glossary_entry = $glossary_entries[$i];
if (gp_stripos($t->singular . ' ' . $t->plural, $term) !== false) {
$matching_entries[$term][] = array('translation' => $glossary_entry->translation, 'pos' => $glossary_entry->part_of_speech, 'comment' => $glossary_entry->comment);
}
}
//Replace terms in strings with markup
foreach ($matching_entries as $term => $glossary_data) {
$replacement = '<span class="glossary-word" data-translations="' . htmlspecialchars(gp_json_encode($glossary_data), ENT_QUOTES, 'UTF-8') . '">$1</span>';
$translations[$key]->singular_glossary_markup = preg_replace('/\\b(' . preg_quote($term, '/') . '[es|s]?)(?![^<]*<\\/span>)\\b/iu', $replacement, $translations[$key]->singular_glossary_markup);
if ($t->plural) {
$translations[$key]->plural_glossary_markup = preg_replace('/\\b(' . preg_quote($term, '/') . '[es|s]?)(?![^<]*<\\/span>)\\b/iu', $replacement, $translations[$key]->plural_glossary_markup);
}
}
}
return $translations;
}
function closest_original($input, $other_strings)
{
if (empty($other_strings)) {
return null;
}
$input_length = gp_strlen($input);
$closest_similarity = 0;
foreach ($other_strings as $compared_string) {
$compared_string_length = gp_strlen($compared_string);
$max_length_diff = apply_filters('gp_original_import_max_length_diff', 0.5);
if (abs(($input_length - $compared_string_length) / $input_length) > $max_length_diff) {
continue;
}
$similarity = gp_string_similarity($input, $compared_string);
if ($similarity > $closest_similarity) {
$closest = $compared_string;
$closest_similarity = $similarity;
}
}
if (!isset($closest)) {
return null;
}
$min_score = apply_filters('gp_original_import_min_similarity_diff', 0.8);
$close_enough = $closest_similarity > $min_score;
do_action('post_string_similiary_test', $input, $closest, $closest_similarity, $close_enough);
if ($close_enough) {
return $closest;
} else {
return null;
}
}
public function closest_original($input, $other_strings)
{
if (empty($other_strings)) {
return null;
}
$input_length = gp_strlen($input);
$closest_similarity = 0;
foreach ($other_strings as $compared_string) {
$compared_string_length = gp_strlen($compared_string);
/**
* Filter the maximum length difference allowed when comparing originals for a close match when importing.
*
* @since 1.0.0
*
* @param float $max_length_diff The times compared string length can differ from the input string.
*/
$max_length_diff = apply_filters('gp_original_import_max_length_diff', 0.5);
if (abs(($input_length - $compared_string_length) / $input_length) > $max_length_diff) {
continue;
}
$similarity = gp_string_similarity($input, $compared_string);
if ($similarity > $closest_similarity) {
$closest = $compared_string;
$closest_similarity = $similarity;
}
}
if (!isset($closest)) {
return null;
}
/**
* Filter the minimum allowed similarity to be considered as a close match.
*
* @since 1.0.0
*
* @param float $similarity Minimum allowed similarity.
*/
$min_score = apply_filters('gp_original_import_min_similarity_diff', 0.8);
$close_enough = $closest_similarity > $min_score;
/**
* Fires before determining string similarity.
*
* @since 1.0.0
*
* @param string $input The original string to match against.
* @param string $closest Closest matching string.
* @param float $closest_similarity The similarity between strings that was calculated.
* @param bool $close_enough Whether the closest was be determined as close enough match.
*/
do_action('gp_post_string_similiary_test', $input, $closest, $closest_similarity, $close_enough);
if ($close_enough) {
return $closest;
} else {
return null;
}
}
function gp_string_similarity($str1, $str2)
{
$length1 = gp_strlen($str1);
$length2 = gp_strlen($str2);
$len = min($length1, $length2);
if ($len > 5000) {
//Arbitrary limit on character length for speed purpose.
$distance = $len;
} else {
$distance = gp_levenshtein($str1, $str2, $length1, $length2);
}
$similarity = 1 - $distance * 0.9 / $len;
return $similarity;
}
