Nemcio/Grav
1
0
Fork 0
mirror of https://github.com/getgrav/grav.git synced 2025-10-26 07:56:07 +01:00
Code Issues Packages Projects Releases Activity

wordCount Filter for Grav (#3957)

Browse Source
This commit is contained in:
pmoreno.rodriguez
2025-10-13 21:35:33 +02:00
committed by GitHub
parent 03849923d4
commit c82645a42a
1 changed files with 57 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

57
system/src/Grav/Common/Twig/Extension/GravExtension.php
View File

@@ -140,6 +140,7 @@ class GravExtension extends AbstractExtension implements GlobalsInterface
new TwigFilter('starts_with', [$this, 'startsWithFilter']),
new TwigFilter('truncate', [Utils::class, 'truncate']),
new TwigFilter('truncate_html', [Utils::class, 'truncateHTML']),
new TwigFilter('wordcount', [$this, 'wordCountFilter']),
new TwigFilter('json_decode', [$this, 'jsonDecodeFilter']),
new TwigFilter('array_unique', 'array_unique'),
new TwigFilter('basename', 'basename'),
@@ -578,6 +579,62 @@ class GravExtension extends AbstractExtension implements GlobalsInterface
return $str;
}
/**
* Count words in text with improved accuracy for multiple languages
*
* @param string $text The text to count words from
* @param string $locale Optional locale for language-specific counting (default: 'en')
* @return int Number of words
*/
public function wordCountFilter($text, string $locale = 'en'): int
{
if (empty($text)) {
return 0;
}
// Strip HTML tags and decode entities
$cleanText = html_entity_decode(strip_tags($text), ENT_QUOTES, 'UTF-8');
// Remove extra whitespace and normalize
$cleanText = trim(preg_replace('/\s+/', ' ', $cleanText));
if (empty($cleanText)) {
return 0;
}
// Handle different languages
switch (strtolower($locale)) {
case 'zh':
case 'zh-cn':
case 'zh-tw':
case 'chinese':
// Chinese: count characters (excluding spaces and punctuation)
return mb_strlen(preg_replace('/[\s\p{P}]/u', '', $cleanText), 'UTF-8');
case 'ja':
case 'japanese':
// Japanese: count characters (excluding spaces)
return mb_strlen(preg_replace('/\s/', '', $cleanText), 'UTF-8');
case 'ko':
case 'korean':
// Korean: count characters (excluding spaces)
return mb_strlen(preg_replace('/\s/', '', $cleanText), 'UTF-8');
default:
// Western languages: use improved word counting
// Handle contractions, hyphenated words, and numbers better
$words = preg_split('/\s+/', $cleanText, -1, PREG_SPLIT_NO_EMPTY);
// Filter out pure punctuation
$words = array_filter($words, function($word) {
return preg_match('/\w/', $word);
});
return count($words);
}
}
/**
* Get Cron object for a crontab 'at' format
*