Script (PHP) to extract all translatable strings from Craft CMS templates located in a specific folder

extract-translations.php

<?php

/**
 * @author  Quentin Delcourt <quentin@delcourt.be>
 * @date    2017-06-21
 *
 * This script will take a folder path in argument and will extract
 * all translatable strings from the Craft templates encountered
 * in the given folder. Each one of these will be
 * outputted to stdOut in CSV format.
 *
 * If no folder path is provided, the script will look
 * in the current folder.
 *
 * Usage:
 * php extract-translations.php path/to/templates
 */


/**
 * Recursively explore all .html files in a directory
 * and find every single occurence of a string to be translated.
 *
 * @param $folder Relative path to the folder containing the templates
 * @return array
 */
function findTranslations($folder)
{
    $allMatches = [];

    $filesPattern = '/\.html$/';
    $regex = '/\'((?:[^\\\'\\\\]|\\\\.)*)\'\\|translate/';
    $directory = new RecursiveDirectoryIterator($folder);
    $iterator = new RecursiveIteratorIterator($directory);
    $files = new RegexIterator($iterator, $filesPattern);
    foreach ($files as $template) {
        $content = file_get_contents($template);
        $matches = [];
        preg_match_all($regex, $content, $matches);
        if(count($matches) > 1) {
            foreach($matches[1] as $match) {
                $allMatches[] = $match;
            }
        }
    }

    $allMatches = array_unique($allMatches);
    asort($allMatches);

    return $allMatches;
}

/**
 * Output a list of strings on stdOut in CSV format.
 *
 * @param $matches
 */
function outputCsv($matches)
{
    $stdOut = fopen('php://output', 'w');
    foreach($matches as $match) {
        fputcsv($stdOut, [$match]);
    }
}

$folder = (count($argv) > 1) ? $argv[1] : '.';
$translations = outputCsv(findTranslations($folder));