Index: modules/filter/filter.module
===================================================================
RCS file: /cvs/drupal/drupal/modules/filter/filter.module,v
retrieving revision 1.219
diff -u -p -r1.219 filter.module
--- modules/filter/filter.module 14 Aug 2008 11:58:06 -0000 1.219
+++ modules/filter/filter.module 15 Aug 2008 08:59:55 -0000
@@ -306,8 +306,8 @@ function filter_formats($index = NULL) {
$args[] = variable_get('filter_default_format', 1);
}
- $result = db_query($query . ' ORDER by weight', $args);
- while ($format = db_fetch_object($result)) {
+ $replacementsult = db_query($query . ' ORDER by weight', $args);
+ while ($format = db_fetch_object($replacementsult)) {
$formats[$format->format] = $format;
}
}
@@ -370,8 +370,8 @@ function filter_list_format($format) {
if (!isset($filters[$format])) {
$filters[$format] = array();
- $result = db_query("SELECT * FROM {filters} WHERE format = %d ORDER BY weight, module, delta", $format);
- while ($filter = db_fetch_object($result)) {
+ $replacementsult = db_query("SELECT * FROM {filters} WHERE format = %d ORDER BY weight, module, delta", $format);
+ while ($filter = db_fetch_object($replacementsult)) {
$list = module_invoke($filter->module, 'filter', 'list');
if (isset($list) && is_array($list) && isset($list[$filter->delta])) {
$filter->name = $list[$filter->delta];
@@ -714,24 +714,99 @@ function _filter_url_settings($format) {
}
/**
- * URL filter. Automatically converts text web addresses (URLs, e-mail addresses,
- * ftp links, etc.) into hyperlinks.
+ * URL filter. Automatically converts text web addresses (URLs, e-mail
+ * addresses, ftp links, etc.) into hyperlinks.
+ *
+ * This filter identifies and makes clickable three types of "links".
+ * - URLs like http://example.com.
+ * - E-mail addresses like name@example.com.
+ * - Web addresses without the "http://" protocol defined, like www.example.com.
+ * Each type must be processed separately, as there is no one regular
+ * expression that could possibly match all of the cases in one pass.
*/
function _filter_url($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.
+ $ignoretags = '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 $ignoretags (see above).
+ // If an ignoretag 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.
+ $opentag = '';
+
+ for ($i = 0; $i < count($chunks); $i++) {
+ if ($chunk_type == 'text') {
+ // Only do replacements when there are no unclosed ignoretags.
+ if ($opentag == '') {
+ // 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.
+ $chunk_type = 'tag';
+ }
+ else {
+ if ($opentag == '') {
+ // No open ignoretags. Process this tag...
+ if (preg_match("`<($ignoretags)(?:\s|>)`i", $chunks[$i], $matches)) {
+ // This matches one of the $ignoretags.
+ // Catch and store the tag in question.
+ $opentag = $matches[1];
+ }
+ }
+ else {
+ // There is an $ignoretag open. See if this is a matching closing tag.
+ // Nothing else is done until we find the closing tag.
+ if (preg_match("`<\/$opentag>`i", $chunks[$i], $matches)) {
+ $opentag = '';
+ }
+ }
+ // Done processing tag chunk, so next chunk is text.
+ $chunk_type = 'text';
+ }
+ }
+ $text = implode($chunks);
+ }
return $text;
}
@@ -811,23 +886,42 @@ function _filter_htmlcorrector($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 1.6
diff -u -p -r1.6 filter.test
--- modules/filter/filter.test 15 Aug 2008 07:49:42 -0000 1.6
+++ modules/filter/filter.test 15 Aug 2008 08:59:55 -0000
@@ -206,6 +206,122 @@ class FilterTestCase extends DrupalWebTe
}
/**
+ * Test the URL filter
+ */
+ function testUrlFilter() {
+ $url_filter = 2;
+
+ $format = $this->createFormat($url_filter);
+
+$body=<<using www.example4.com the code tag.
+
+
+Test inside blockquote tag www.example5.com. email with person@example6.com. and url http://www.example7.com. And also using www.example8.com the code tag and also inside www.example9.com em tags bla bla.
+
+
+One more simple code tag test? http://www.example10.com abc
+
+Test the really simple cases next:
+
+http://www.example11.com
+www.example12.com
+person@example13.com
+www.example14.com
+
+What about tags that don't exist like x say www.example15.com? And what about tag beginning www.example16.com with p?
+
+Test <br/>: This is a www.example17.com. example with some http://www.example18.com various tags within the paragraph. *
Also it is important www.example19.com to *
test multiple different url's and wwwstrings http://www.example20.com urls in same paragraph. *
I mean it www.example21.com many of them person@example22.com after each http://www.example23.com other *img*![]()
abc. This is just a www.example24.com paragraph with some http://www.example25.com urls thrown in. This is just a www.example26.com paragraph person@example27.com with some http://www.example28.com urls thrown in.
+
+
+
+Again some simple tests inside various tags:
+
+http://www.example30.com
+http://www.example31.com
+http://www.example32.com
+
+And also example ftp URL ftp://ftp.example33.com.
+
+The old URL filter has problems with this kind of link with www address as part of text in title. www.example43.com
+
+
+- www.example44.com
+- http://www.example45.com
+- person@example46.com
+- check www.example47.com
+- this with some text around: http://www.example48.com not so easy person@example49.com now?
+
+
+
+
+hello.... there!
+END;
+
+
+
+ $edit = array();
+ $edit['title'] = $this->randomName();
+ $edit['body'] = $body;
+ $edit['format'] = $format->format;
+ $edit['type'] = 'page';
+ $page = $this->drupalCreateNode($edit);
+
+ $this->drupalGet('node/' . $page->nid);
+ $this->assertRaw('href="http://www.example1.com"', t('Parse simple www-string but not the end-of-sentence period.'));
+ $this->assertRaw('href="mailto:person@example2.com"', t('Parse simple email string but not the end-of-sentence period.'));
+ $this->assertRaw('href="http://www.example3.com"', t('Parse simple HTTP URL but not the end-of-sentence period.'));
+ $this->assertNoRaw('href="http://www.example4.com"', t('Do not parse simple HTTP URL inside code tags.'));
+ $this->assertRaw('href="http://www.example5.com"', t('Parse www-string inside blockquote tag.'));
+ $this->assertRaw('href="mailto:person@example6.com"', t('Parse email string inside blockquote tag.'));
+ $this->assertRaw('href="http://www.example7.com"', t('Parse HTTP URL inside blockquote tag'));
+ $this->assertNoRaw('href="http://www.example8.com"', t('Do not parse simple HTTP URL inside code tags.'));
+ $this->assertNoRaw('href="http://www.example9.com"', t('Do not parse simple HTTP URL inside em nested inside code tags.'));
+ $this->assertNoRaw('href="http://www.example10.com"', t('Do not parse simple HTTP URL inside code tags.'));
+ $this->assertRaw('href="http://www.example11.com"', t('Parse simple HTTP URL.'));
+ $this->assertRaw('href="http://www.example12.com"', t('Parse simple www-string.'));
+ $this->assertRaw('href="mailto:person@example13.com"', t('Parse simple email string.'));
+ $this->assertNoRaw('href="http://www.example14.com"', t('Do not parse simple HTTP URL inside code tags.'));
+ $this->assertRaw('href="http://www.example15.com"', t('Parse www-string inside tag not part of HTML spec ( ).'));
+ $this->assertRaw('href="http://www.example16.com"', t('Parse www-string inside tag not part of HTML spec but beginning with p ( ).'));
+ $this->assertRaw('href="http://www.example17.com"', t('Parse multiple www-strings inside same paragraph.'));
+ $this->assertRaw('href="http://www.example18.com"', t('Parse multiple www-strings inside same paragraph.'));
+ $this->assertRaw('href="http://www.example19.com"', t('Parse multiple www-strings inside same paragraph.'));
+ $this->assertRaw('href="http://www.example20.com"', t('Parse multiple www-strings inside same paragraph limited with
.'));
+ $this->assertRaw('href="http://www.example21.com"', t('Parse multiple www-strings inside same paragraph limited with
.'));
+ $this->assertRaw('href="mailto:person@example22.com"', t('Parse email string with multiple www-strings inside same paragraph limited with
.'));
+ $this->assertRaw('href="http://www.example23.com"', t('Parse multiple www-strings inside same paragraph limited with
.'));
+ $this->assertRaw('href="http://www.example24.com"', t('Parse multiple www-strings inside same paragraph limited with
and ![]()
.'));
+ $this->assertRaw('href="http://www.example25.com"', t('Parse multiple www-strings inside same paragraph limited with
and ![]()
.'));
+ $this->assertRaw('href="http://www.example26.com"', t('Parse multiple www-strings inside same paragraph limited with
and ![]()
.'));
+ $this->assertRaw('href="mailto:person@example27.com"', t('Parse email string with multiple www-strings inside same paragraph limited with
and ![]()
.'));
+ $this->assertRaw('href="http://www.example28.com"', t('Parse multiple www-strings inside same paragraph limited with
and ![]()
.'));
+ $this->assertNoRaw('href="http://www.example29.com"', t('Do not parse URL inside a script element (part of javascript code).'));
+ $this->assertNoRaw('href="http://www.example30.com"', t('Do not parse URL inside an a element.'));
+ $this->assertRaw('href="http://www.example31.com"', t('Parse URL inside strong tag.'));
+ $this->assertRaw('href="http://www.example32.com"', t('Parse URL inside em tag.'));
+ $this->assertRaw('href="ftp://ftp.example33.com"', t('Parse ftp:// URL.'));
+ $this->assertNoRaw('href="http://www.example41.com"', t('Do not parse www-strings inside an a element title attribute.'));
+ $this->assertNoRaw('assertRaw('href="http://www.example44.com"', t('Parse www-string inside dl dt tags.'));
+ $this->assertRaw('href="http://www.example45.com"', t('Parse URL inside dl dd tags.'));
+ $this->assertRaw('href="mailto:person@example46.com"', t('Parse email string inside dl dd tags.'));
+ $this->assertRaw('href="http://www.example47.com"', t('Parse www-string with text inside dl dd tags.'));
+ $this->assertRaw('href="http://www.example48.com"', t('Parse URL with text inside dl dd tags.'));
+ $this->assertRaw('href="mailto:person@example49.com"', t('Parse email string with text inside dl dd tags.'));
+ $this->assertNoRaw('href="http://www.example50.com"', t('Do not parse URL that is inside HTML comment.'));
+ $this->assertRaw('hello.... there!', t('Verify that last part of normal text is preserved intact.'));
+
+
+ $this->deleteFormat($format);
+ }
+
+ /**
* Test the line break filter
*/
function testLineBreakFilter() {