Fix util/extract.php

- Captures way more L10n::t(t) instances
- Add recursive search into `src/` folder
- Fix string capture with new sprintf-free syntax

diff --git a/util/extract.php b/util/extract.php
old mode 100644 (file)
new mode 100755 (executable)
index 13151bc..18fbb9d
--- a/util/extract.php
+++ b/util/extract.php
@@ -1,63 +1,100 @@
+#!/usr/bin/env php
 <?php
 
-       $arr = [];
+/**
+ * @file util/extract.php
+ *
+ * Extracts translation strings from the Friendica project's files to be exported
+ * to Transifex for translation.
+ *
+ * Outputs a PHP file with language strings used by Friendica
+ */
 
-       $files = ['index.php','boot.php'];
-       $files = array_merge($files,glob('mod/*'),glob('include/*'),glob('addon/*/*'));
+$s = '<?php' . PHP_EOL;
+$s .= '
+function string_plural_select($n){
+       return ($n != 1);
+}
 
+';
+
+$arr = [];
 
-       foreach($files as $file) {
-               $str = file_get_contents($file);
+$files = ['index.php', 'boot.php'];
+$files = array_merge(
+       $files,
+       glob('mod/*'),
+       glob('include/*'),
+       glob('addon/*/*'),
+       glob_recursive('src')
+);
 
-               $pat = '| L10n::t\(([^\)]*)\)|';
-               $patt = '| L10n::tt\(([^\)]*)\)|';
+foreach ($files as $file) {
+       $str = file_get_contents($file);
 
-               preg_match_all($pat,$str,$matches);
-               preg_match_all($patt, $str, $matchestt);
+       $pat = '|L10n::t\(([^\)]*+)[\)]|';
+       $patt = '|L10n::tt\(([^\)]*+)[\)]|';
 
+       preg_match_all($pat, $str, $matches);
+       preg_match_all($patt, $str, $matchestt);
 
-               if(count($matches)){
-                       foreach($matches[1] as $match) {
-                               if(! in_array($match,$arr))
-                                       $arr[] = $match;
+       if (count($matches) || count($matchestt)) {
+               $s .= '// ' . $file . PHP_EOL;
+       }
+
+       if (count($matches)) {
+               foreach ($matches[1] as $long_match) {
+                       $match_arr = preg_split('/(?<=[\'"])\s*,/', $long_match);
+                       $match = $match_arr[0];
+                       if (!in_array($match, $arr)) {
+                               if (substr($match, 0, 1) == '$') {
+                                       continue;
+                               }
+
+                               $arr[] = $match;
+
+                               $s .= '$a->strings[' . $match . '] = ' . $match . ';' . "\n";
                        }
                }
-               if(count($matchestt)){
-                       foreach($matchestt[1] as $match) {
-                               $matchtkns = preg_split("|[ \t\r\n]*,[ \t\r\n]*|",$match);
-                               if (count($matchtkns)==3 && !in_array($matchtkns,$arr)){
-                                       $arr[] = $matchtkns;
+       }
+       if (count($matchestt)) {
+               foreach ($matchestt[1] as $match) {
+                       $matchtkns = preg_split("|[ \t\r\n]*,[ \t\r\n]*|", $match);
+                       if (count($matchtkns) == 3 && !in_array($matchtkns[0], $arr)) {
+                               if (substr($matchtkns[1], 0, 1) == '$') {
+                                       continue;
                                }
+
+                               $arr[] = $matchtkns[0];
+
+                               $s .= '$a->strings[' . $matchtkns[0] . "] = array(\n";
+                               $s .= "\t0 => " . $matchtkns[0] . ",\n";
+                               $s .= "\t1 => " . $matchtkns[1] . ",\n";
+                               $s .= ");\n";
                        }
                }
-
        }
+}
 
-       $s = '<?php' . "\n";
-       $s .= '
-function string_plural_select($n){
-       return ($n != 1);
+$s .= '// Timezones' . PHP_EOL;
+
+$zones = timezone_identifiers_list();
+foreach ($zones as $zone) {
+       $s .= '$a->strings[\'' . $zone . '\'] = \'' . $zone . '\';' . "\n";
 }
 
-';
+echo $s;
+
+function glob_recursive($path) {
+       $dir_iterator = new RecursiveDirectoryIterator($path);
+       $iterator = new RecursiveIteratorIterator($dir_iterator, RecursiveIteratorIterator::SELF_FIRST);
 
-       foreach($arr as $a) {
-               if (is_array($a)){
-                       if(substr($a[1],0,1) == '$')
-                               continue;
-                       $s .= '$a->strings[' . $a[0] . "] = array(\n";
-                       $s .= "\t0 => ". $a[0]. ",\n";
-                       $s .= "\t1 => ". $a[1]. ",\n";
-                       $s .= ");\n";
-               } else {
-                       if(substr($a,0,1) == '$')
-                               continue;
-                       $s .= '$a->strings[' . $a . '] = '. $a . ';' . "\n";
+       $return = [];
+       foreach ($iterator as $file) {
+               if ($file->getBasename() != '.' && $file->getBasename() != '..') {
+                       $return[] = $file->getPathname();
                }
        }
 
-       $zones = timezone_identifiers_list();
-       foreach($zones as $zone)
-               $s .= '$a->strings[\'' . $zone . '\'] = \'' . $zone . '\';' . "\n";
-
-       echo $s;
\ No newline at end of file
+       return $return;
+}