Extract core of qa_q_request into new function, qa_slugify

qa-include/qa-base.php

@@ -1462,43 +1462,24 @@
 	}
 
 
+	/**
+	 * Get Q2A request for a question, and make it search-engine friendly, shortening it if necessary
+	 * by removing shorter words which are generally less meaningful.
+	 * @param  int $questionid The question ID
+	 * @param string $title The question title
+	 * @return  string
+	 */
 	function qa_q_request($questionid, $title)
-/*
-	Return the Q2A request for question $questionid, and make it search-engine friendly based on $title, which is
-	shortened if necessary by removing shorter words which are generally less meaningful.
-*/
 	{
 		if (qa_to_override(__FUNCTION__)) { $args=func_get_args(); return qa_call_override(__FUNCTION__, $args); }
 
-		require_once QA_INCLUDE_DIR.'app/options.php';
-		require_once QA_INCLUDE_DIR.'util/string.php';
-
-		$title=qa_block_words_replace($title, qa_get_block_words_preg());
-
-		$words=qa_string_to_words($title, true, false, false);
-
-		$wordlength=array();
-		foreach ($words as $index => $word)
-			$wordlength[$index]=qa_strlen($word);
-
-		$remaining=qa_opt('q_urls_title_length');
-
-		if (array_sum($wordlength)>$remaining) {
-			arsort($wordlength, SORT_NUMERIC); // sort with longest words first
-
-			foreach ($wordlength as $index => $length) {
-				if ($remaining>0)
-					$remaining-=$length;
-				else
-					unset($words[$index]);
-			}
-		}
+		require_once QA_INCLUDE_DIR . 'app/options.php';
+		require_once QA_INCLUDE_DIR . 'util/string.php';
 
-		$title=implode('-', $words);
-		if (qa_opt('q_urls_remove_accents'))
-			$title=qa_string_remove_accents($title);
+		$title = qa_block_words_replace($title, qa_get_block_words_preg());
+		$slug = qa_slugify($title, qa_opt('q_urls_remove_accents'), qa_opt('q_urls_title_length'));
 
-		return (int)$questionid.'/'.$title;
+		return (int) $questionid . '/' . $slug;
 	}
 
 

qa-include/util/string.php

@@ -450,10 +450,12 @@
 	}
 
 
+	/**
+	 * Convert accents in a UTF-8 string to ASCII.
+	 * @param string $string Input string.
+	 * @return string
+	 */
 	function qa_string_remove_accents($string)
-/*
-	Return UTF-8 input $string with all accents (on Roman characters) removed
-*/
 	{
 		if (qa_to_override(__FUNCTION__)) { $args=func_get_args(); return qa_call_override(__FUNCTION__, $args); }
 
@@ -463,6 +465,52 @@
 	}
 
 
+	/**
+	 * Convert string to an SEO-friendly URL segment.
+	 * @param string $string Input string.
+	 * @param bool $asciiOnly If true, removes all non-ASCII characters that weren't converted.
+	 * @param int|null $maxLength Maximum length the segment should be, or null for no limit.
+	 * @return string
+	 */
+	function qa_slugify($string, $asciiOnly=true, $maxLength=null)
+	{
+		if (qa_to_override(__FUNCTION__)) { $args=func_get_args(); return qa_call_override(__FUNCTION__, $args); }
+
+		$words = qa_string_to_words($string, true, false, false);
+
+		if ($maxLength !== null) {
+			$wordlength = array();
+			foreach ($words as $index => $word) {
+				$wordlength[$index] = qa_strlen($word);
+			}
+
+			$remaining = $maxLength;
+			if (array_sum($wordlength)>$remaining) {
+				arsort($wordlength, SORT_NUMERIC); // sort with longest words first
+
+				foreach ($wordlength as $index => $length) {
+					if ($remaining > 0) {
+						$remaining -= $length;
+					} else {
+						unset($words[$index]);
+					}
+				}
+			}
+		}
+
+		$string = implode('-', $words);
+
+		if ($asciiOnly) {
+			// convert accents to ASCII, then remove all remaining non-ASCII characters
+			$string = qa_string_remove_accents($string);
+			$string = preg_replace('/[^ a-zA-Z0-9-]/', '', $string);
+			$string = preg_replace('/-{2,}/', '-', trim($string, '-'));
+		}
+
+		return $string;
+	}
+
+
 	function qa_tags_to_tagstring($tags)
 /*
 	Convert an array of tags into a string for storage in the database