Index: modules/filter/filter.module
===================================================================
RCS file: /cvs/drupal/drupal/modules/filter/filter.module,v
retrieving revision 1.261
diff -u -p -r1.261 filter.module
--- modules/filter/filter.module 13 Jun 2009 19:37:27 -0000 1.261
+++ modules/filter/filter.module 28 Jun 2009 21:23:01 -0000
@@ -731,24 +731,98 @@ function _filter_url_settings($format) {
}
/**
- * URL filter. Automatically converts text web addresses (URLs, e-mail addresses,
- * ftp links, etc.) into hyperlinks.
+ * URL filter. Automatically converts text into hyperlinks.
+ *
+ * This filter identifies and makes clickable three types of "links".
+ * - URLs like http://example.com.
+ * - E-mail addresses like name@example.com.
+ * - Web addresses without the "http://" protocol defined, like www.example.com.
+ * Each type must be processed separately, as there is no one regular
+ * expression that could possibly match all of the cases in one pass.
*/
function _filter_url($text, $format) {
- // Pass length to regexp callback
- _filter_url_trim(NULL, variable_get('filter_url_length_' . $format, 72));
+ // List of tags - the content of which must be skipped.
+ $ignore_tags = 'a|script|style|code';
- $text = ' ' . $text . ' ';
+ // Create an array which contains the regexps for each type of link.
+ // The key to the regexp is the name of a function that is used as
+ // callback function to process matches of the regexp. The callback function
+ // is to return the replacement for the match. The array is used and
+ // matching/replacement done below inside some loops.
+ $tasks = NULL;
// Match absolute URLs.
- $text = preg_replace_callback("`(|
||[ \n\r\t\(])((http://|https://|ftp://|mailto:|smb://|afp://|file://|gopher://|news://|ssl://|sslv2://|sslv3://|tls://|tcp://|udp://)([a-zA-Z0-9@:%_+*~#?&=.,/;-]*[a-zA-Z0-9@:%_+*~#&=/;-]))([.,?!]*?)(?=(|||[ \n\r\t\)]))`i", '_filter_url_parse_full_links', $text);
+ $protocols = 'http://|https://|ftp://|mailto:|smb://|afp://|file://|gopher://|news://|ssl://|sslv2://|sslv3://|tls://|tcp://|udp://';
+ $url_pattern = "(?:$protocols)(?:[a-zA-Z0-9@:%_+*~#?&=.,/;-]*[a-zA-Z0-9@:%_+*~#&=/;-])";
+ $replacement = "`($url_pattern)([\.\,\?\!]*?)`i";
+ $tasks['_filter_url_parse_full_links'] = $replacement;
// Match e-mail addresses.
- $text = preg_replace("`(|
||[ \n\r\t\(])([A-Za-z0-9._-]+@[A-Za-z0-9._+-]+\.[A-Za-z]{2,4})([.,?!]*?)(?=(|||[ \n\r\t\)]))`i", '\1\2\3', $text);
+ // Note: The ICANN seems to be on track towards accepting more diverse top level domains,
+ // so this pattern has been "future-proofed" to allow for TLD's of length 2-64.
+ $url_pattern = '[A-Za-z0-9._-]+@[A-Za-z0-9._+-]+\.[A-Za-z]{2,64}';
+ $replacement = "`($url_pattern)`i";
+ $tasks['_filter_url_parse_email_links'] = $replacement;
// Match www domains/addresses.
- $text = preg_replace_callback("`(|
|[ \n\r\t\(])(www\.[a-zA-Z0-9@:%_+*~#?&=.,/;-]*[a-zA-Z0-9@:%_+~#\&=/;-])([.,?!]*?)(?=(|||[ \n\r\t\)]))`i", '_filter_url_parse_partial_links', $text);
- $text = substr($text, 1, -1);
+ $url_pattern = 'www\.[a-zA-Z0-9@:%_+*~#?&=.,/;-]*[a-zA-Z0-9@:%_+~#\&=/;-]';
+ $replacement = "`($url_pattern)([\.\,\?\!]*?)`i";
+ $tasks['_filter_url_parse_partial_links'] = $replacement;
+
+
+ // Pass length to regexp callback.
+ _filter_url_trim(NULL, variable_get('filter_url_length_' . $format, 72));
+
+ // We need to process each case of replacement type separately.
+ // The text must be joined and split again after each
+ // replacement, since replacements create new HTML tags and the new
+ // tags must be correctly protected before the next replacement can be done.
+ foreach ($tasks as $task => $replacement) {
+ // Split at all tags.
+ // This ensures that nothing that is a tagname or attribute will be processed.
+ $chunks = preg_split('/(<.+?>)/i', $text, -1, PREG_SPLIT_DELIM_CAPTURE);
+ // Note: PHP ensures the array consists of alternating delimiters and literals
+ // and begins and ends with a literal (inserting NULL as required).
+ // Therefore, first chunk is always text:
+ $chunk_type = 'text';
+ // Tags to ignore are defined in $ignore_tags (see above).
+ // If an ignore_tag is found, it is stored here and removed only when the
+ // closing tag is found. Until the closing tag is found, no replacements are made.
+ $open_tag = '';
+
+ for ($i = 0; $i < count($chunks); $i++) {
+ if ($chunk_type == 'text') {
+ // Only do replacements when there are no unclosed ignore_tags.
+ if ($open_tag == '') {
+ // This is the high point of this function! If there is a match,
+ // a link is created in the callback function named by $task.
+ $chunks[$i] = preg_replace_callback($replacement, $task, $chunks[$i]);
+ }
+ // Done processing text chunk, so next chunk is a tag.
+ $chunk_type = 'tag';
+ }
+ else {
+ if ($open_tag == '') {
+ // No open ignore_tags. Process this tag...
+ if (preg_match("`<($ignore_tags)(?:\s|>)`i", $chunks[$i], $matches)) {
+ // This matches one of the $ignore_tags.
+ // Catch and store the tag in question.
+ $open_tag = $matches[1];
+ }
+ }
+ else {
+ // There is an $ignore_tag open. See if this is a matching closing tag.
+ // Nothing else is done until we find the closing tag.
+ if (preg_match("`<\/$open_tag>`i", $chunks[$i], $matches)) {
+ $open_tag = '';
+ }
+ }
+ // Done processing tag chunk, so next chunk is text.
+ $chunk_type = 'text';
+ }
+ }
+ $text = implode($chunks);
+ }
return $text;
}
@@ -828,23 +902,42 @@ function _filter_htmlcorrector($text) {
}
/**
- * Make links out of absolute URLs.
+ * Callback function. Make links out of absolute URLs.
*/
function _filter_url_parse_full_links($match) {
- $match[2] = decode_entities($match[2]);
- $caption = check_plain(_filter_url_trim($match[2]));
- $match[2] = check_url($match[2]);
- return $match[1] . '' . $caption . '' . $match[5];
+ // The $i:th parenthesis in the regexp contains the URL.
+ $i = 1;
+
+ $match[$i] = decode_entities($match[$i]);
+ $caption = check_plain(_filter_url_trim($match[$i]));
+ $match[$i] = check_url($match[$i]);
+ return '' . $caption . '' . $match[$i+1];
}
/**
- * Make links out of domain names starting with "www."
+ * Callback function. Make links out of e-mail addresses.
+ */
+function _filter_url_parse_email_links($match) {
+ // The $i:th parenthesis in the regexp contains the URL.
+ $i = 0;
+
+ $match[$i] = decode_entities($match[$i]);
+ $caption = check_plain(_filter_url_trim($match[$i]));
+ $match[$i] = check_url($match[$i]);
+ return '' . $caption . '';
+}
+
+/**
+ * Callback function. Make links out of domain names starting with "www.".
*/
function _filter_url_parse_partial_links($match) {
- $match[2] = decode_entities($match[2]);
- $caption = check_plain(_filter_url_trim($match[2]));
- $match[2] = check_plain($match[2]);
- return $match[1] . '' . $caption . '' . $match[3];
+ // The $i:th parenthesis in the regexp contains the URL.
+ $i = 1;
+
+ $match[$i] = decode_entities($match[$i]);
+ $caption = check_plain(_filter_url_trim($match[$i]));
+ $match[$i] = check_plain($match[$i]);
+ return '' . $caption . '' . $match[$i+1];
}
/**
Index: modules/filter/filter.test
===================================================================
RCS file: /cvs/drupal/drupal/modules/filter/filter.test,v
retrieving revision 1.25
diff -u -p -r1.25 filter.test
--- modules/filter/filter.test 28 Jun 2009 18:03:56 -0000 1.25
+++ modules/filter/filter.test 28 Jun 2009 21:23:02 -0000
@@ -582,6 +582,13 @@ class FilterTestCase extends DrupalWebTe
$f = _filter_url('www.example.com/index.php?a=.', 'f');
$this->assertEqual($f, 'www.example.com/index.php?a=.', t('Converting URLs -- do forget about a dot at the end of a query string.'));
+
+ // Confirm that URLs in comments are not processed.
+ $filtered = _filter_url('', 'f');
+ $this->assertEqual($filtered, '', t('Converting URLs -- do not change comments.'));
+
+ $filtered = _filter_url('http://www.example.com
', 'f'); // Not so sure if it should or it shouldn't convert these.
+ $this->assertEqual($filtered, 'http://www.example.com
', t('Converting URLs -- blockquote handling.'));
}
/**