summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorwebchick <webchick@24967.no-reply.drupal.org>2011-06-29 22:51:13 -0700
committerwebchick <webchick@24967.no-reply.drupal.org>2011-06-29 22:51:13 -0700
commit2608c670ababfd04a4ce85393db7ebe4a0c14188 (patch)
treefbdcb5236b5fd0822257a9773f18e90844028650
parent33f6c5608c62b27fa86d52e8806435a6c94f3eee (diff)
downloadbrdo-2608c670ababfd04a4ce85393db7ebe4a0c14188.tar.gz
brdo-2608c670ababfd04a4ce85393db7ebe4a0c14188.tar.bz2
Issue #1130198 by pillarsdotnet, Damien Tournoud: Fixed Regression: line-breaks are mangled by drupal_html_to_text().
-rw-r--r--includes/mail.inc4
-rw-r--r--modules/simpletest/tests/mail.test353
2 files changed, 355 insertions, 2 deletions
diff --git a/includes/mail.inc b/includes/mail.inc
index d2febed39..7272df972 100644
--- a/includes/mail.inc
+++ b/includes/mail.inc
@@ -430,7 +430,7 @@ function drupal_html_to_text($string, $allowed_tags = NULL) {
$indent[] = count($lists) ? ' "' : '>';
break;
case 'li':
- $indent[] = is_numeric($lists[0]) ? ' ' . $lists[0]++ . ') ' : ' * ';
+ $indent[] = isset($lists[0]) && is_numeric($lists[0]) ? ' ' . $lists[0]++ . ') ' : ' * ';
break;
case 'dd':
$indent[] = ' ';
@@ -509,7 +509,7 @@ function drupal_html_to_text($string, $allowed_tags = NULL) {
$chunk = $casing($chunk);
}
// Format it and apply the current indentation.
- $output .= drupal_wrap_mail($chunk, implode('', $indent));
+ $output .= drupal_wrap_mail($chunk, implode('', $indent)) . MAIL_LINE_ENDINGS;
// Remove non-quotation markers from indentation.
$indent = array_map('_drupal_html_to_text_clean', $indent);
}
diff --git a/modules/simpletest/tests/mail.test b/modules/simpletest/tests/mail.test
index 8a7b152d9..a6c7b40e5 100644
--- a/modules/simpletest/tests/mail.test
+++ b/modules/simpletest/tests/mail.test
@@ -1,6 +1,7 @@
<?php
/**
+ * @file
* Test the Drupal mailing system.
*/
class MailTestCase extends DrupalWebTestCase implements MailSystemInterface {
@@ -63,3 +64,355 @@ class MailTestCase extends DrupalWebTestCase implements MailSystemInterface {
}
}
+/**
+ * Unit tests for drupal_html_to_text().
+ */
+class DrupalHtmlToTextTestCase extends DrupalWebTestCase {
+ public static function getInfo() {
+ return array(
+ 'name' => 'HTML to text conversion',
+ 'description' => 'Tests drupal_html_to_text().',
+ 'group' => 'Mail',
+ );
+ }
+
+ /**
+ * Converts a string to its PHP source equivalent for display in test messages.
+ *
+ * @param $text
+ * The text string to convert.
+ *
+ * @return
+ * An HTML representation of the text string that, when displayed in a
+ * browser, represents the PHP source code equivalent of $text.
+ */
+ function stringToHtml($text) {
+ return '"' .
+ str_replace(
+ array("\n", ' '),
+ array('\n', '&nbsp;'),
+ check_plain($text)
+ ) . '"';
+ }
+
+ /**
+ * Helper function for testing drupal_html_to_text().
+ *
+ * @param $html
+ * The source HTML string to be converted.
+ * @param $text
+ * The expected result of converting $html to text.
+ * @param $message
+ * A text message to display in the assertion message.
+ * @param $allowed_tags
+ * (optional) An array of allowed tags, or NULL to default to the full
+ * set of tags supported by drupal_html_to_text().
+ */
+ function assertHtmlToText($html, $text, $message, $allowed_tags = NULL) {
+ preg_match_all('/<([a-z0-6]+)/', drupal_strtolower($html), $matches);
+ $tested_tags = implode(', ', array_unique($matches[1]));
+ $message .= ' (' . $tested_tags . ')';
+ $result = drupal_html_to_text($html, $allowed_tags);
+ $pass = $this->assertEqual($result, $text, check_plain($message));
+ $verbose = 'html = <pre>' . $this->stringToHtml($html)
+ . '</pre><br />' . 'result = <pre>' . $this->stringToHtml($result)
+ . '</pre><br />' . 'expected = <pre>' . $this->stringToHtml($text)
+ . '</pre>';
+ $this->verbose($verbose);
+ if (!$pass) {
+ $this->pass("Previous test verbose info:<br />$verbose");
+ }
+ }
+
+ /**
+ * Test all supported tags of drupal_html_to_text().
+ */
+ function testTags() {
+ global $base_path, $base_url;
+ $tests = array(
+ // @todo Trailing linefeeds should be trimmed.
+ '<a href = "http://drupal.org">Drupal.org</a>' => "Drupal.org [1]\n\n[1] http://drupal.org\n",
+ // @todo Footer urls should be absolute.
+ "<a href = \"$base_path\">Homepage</a>" => "Homepage [1]\n\n[1] $base_url/\n",
+ '<address>Drupal</address>' => "Drupal\n",
+ // @todo The <address> tag is currently not supported.
+ '<address>Drupal</address><address>Drupal</address>' => "DrupalDrupal\n",
+ '<b>Drupal</b>' => "*Drupal*\n",
+ // @todo There should be a space between the '>' and the text.
+ '<blockquote>Drupal</blockquote>' => ">Drupal\n",
+ '<blockquote>Drupal</blockquote><blockquote>Drupal</blockquote>' => ">Drupal\n>Drupal\n",
+ '<br />Drupal<br />Drupal<br /><br />Drupal' => "Drupal\nDrupal\nDrupal\n",
+ '<br/>Drupal<br/>Drupal<br/><br/>Drupal' => "Drupal\nDrupal\nDrupal\n",
+ // @todo There should be two line breaks before the paragraph.
+ '<br/>Drupal<br/>Drupal<br/><br/>Drupal<p>Drupal</p>' => "Drupal\nDrupal\nDrupal\nDrupal\n\n",
+ '<div>Drupal</div>' => "Drupal\n",
+ // @todo The <div> tag is currently not supported.
+ '<div>Drupal</div><div>Drupal</div>' => "DrupalDrupal\n",
+ '<em>Drupal</em>' => "/Drupal/\n",
+ '<h1>Drupal</h1>' => "======== DRUPAL ==============================================================\n\n",
+ '<h1>Drupal</h1><p>Drupal</p>' => "======== DRUPAL ==============================================================\n\nDrupal\n\n",
+ '<h2>Drupal</h2>' => "-------- DRUPAL --------------------------------------------------------------\n\n",
+ '<h2>Drupal</h2><p>Drupal</p>' => "-------- DRUPAL --------------------------------------------------------------\n\nDrupal\n\n",
+ '<h3>Drupal</h3>' => ".... Drupal\n\n",
+ '<h3>Drupal</h3><p>Drupal</p>' => ".... Drupal\n\nDrupal\n\n",
+ '<h4>Drupal</h4>' => ".. Drupal\n\n",
+ '<h4>Drupal</h4><p>Drupal</p>' => ".. Drupal\n\nDrupal\n\n",
+ '<h5>Drupal</h5>' => "Drupal\n\n",
+ '<h5>Drupal</h5><p>Drupal</p>' => "Drupal\n\nDrupal\n\n",
+ '<h6>Drupal</h6>' => "Drupal\n\n",
+ '<h6>Drupal</h6><p>Drupal</p>' => "Drupal\n\nDrupal\n\n",
+ '<hr />Drupal<hr />' => "------------------------------------------------------------------------------\nDrupal\n------------------------------------------------------------------------------\n",
+ '<hr/>Drupal<hr/>' => "------------------------------------------------------------------------------\nDrupal\n------------------------------------------------------------------------------\n",
+ '<hr/>Drupal<hr/><p>Drupal</p>' => "------------------------------------------------------------------------------\nDrupal\n------------------------------------------------------------------------------\nDrupal\n\n",
+ '<i>Drupal</i>' => "/Drupal/\n",
+ '<p>Drupal</p>' => "Drupal\n\n",
+ '<p>Drupal</p><p>Drupal</p>' => "Drupal\n\nDrupal\n\n",
+ '<strong>Drupal</strong>' => "*Drupal*\n",
+ // @todo Tables are currently not supported.
+ '<table><tr><td>Drupal</td><td>Drupal</td></tr><tr><td>Drupal</td><td>Drupal</td></tr></table>' => "DrupalDrupalDrupalDrupal\n",
+ '<table><tr><td>Drupal</td></tr></table><p>Drupal</p>' => "Drupal\nDrupal\n\n",
+ // @todo The <u> tag is currently not supported.
+ '<u>Drupal</u>' => "Drupal\n",
+ '<ul><li>Drupal</li></ul>' => " * Drupal\n\n",
+ '<ul><li>Drupal <em>Drupal</em> Drupal</li></ul>' => " * Drupal /Drupal/ Drupal\n\n",
+ // @todo Lines containing nothing but spaces should be trimmed.
+ '<ul><li>Drupal</li><li><ol><li>Drupal</li><li>Drupal</li></ol></li></ul>' => " * Drupal\n * 1) Drupal\n 2) Drupal\n \n\n",
+ '<ul><li>Drupal</li><li><ol><li>Drupal</li></ol></li><li>Drupal</li></ul>' => " * Drupal\n * 1) Drupal\n \n * Drupal\n\n",
+ '<ul><li>Drupal</li><li>Drupal</li></ul>' => " * Drupal\n * Drupal\n\n",
+ '<ul><li>Drupal</li></ul><p>Drupal</p>' => " * Drupal\n\nDrupal\n\n",
+ '<ol><li>Drupal</li></ol>' => " 1) Drupal\n\n",
+ '<ol><li>Drupal</li><li><ul><li>Drupal</li><li>Drupal</li></ul></li></ol>' => " 1) Drupal\n 2) * Drupal\n * Drupal\n \n\n",
+ '<ol><li>Drupal</li><li>Drupal</li></ol>' => " 1) Drupal\n 2) Drupal\n\n",
+ '<ol>Drupal</ol>' => "Drupal\n\n",
+ '<ol><li>Drupal</li></ol><p>Drupal</p>' => " 1) Drupal\n\nDrupal\n\n",
+ '<dl><dt>Drupal</dt></dl>' => "Drupal\n\n",
+ '<dl><dt>Drupal</dt><dd>Drupal</dd></dl>' => "Drupal\n Drupal\n\n",
+ '<dl><dt>Drupal</dt><dd>Drupal</dd><dt>Drupal</dt><dd>Drupal</dd></dl>' => "Drupal\n Drupal\nDrupal\n Drupal\n\n",
+ '<dl><dt>Drupal</dt><dd>Drupal</dd></dl><p>Drupal</p>' => "Drupal\n Drupal\n\nDrupal\n\n",
+ '<dl><dt>Drupal<dd>Drupal</dl>' => "Drupal\n Drupal\n\n",
+ '<dl><dt>Drupal</dt></dl><p>Drupal</p>' => "Drupal\n\nDrupal\n\n",
+ // @todo Again, lines containing only spaces should be trimmed.
+ '<ul><li>Drupal</li><li><dl><dt>Drupal</dt><dd>Drupal</dd><dt>Drupal</dt><dd>Drupal</dd></dl></li><li>Drupal</li></ul>' => " * Drupal\n * Drupal\n Drupal\n Drupal\n Drupal\n \n * Drupal\n\n",
+ // Tests malformed HTML tags.
+ '<br>Drupal<br>Drupal' => "Drupal\nDrupal\n",
+ '<hr>Drupal<hr>Drupal' => "------------------------------------------------------------------------------\nDrupal\n------------------------------------------------------------------------------\nDrupal\n",
+ '<ol><li>Drupal<li>Drupal</ol>' => " 1) Drupal\n 2) Drupal\n\n",
+ '<ul><li>Drupal <em>Drupal</em> Drupal</ul></ul>' => " * Drupal /Drupal/ Drupal\n\n",
+ '<ul><li>Drupal<li>Drupal</ol>' => " * Drupal\n * Drupal\n\n",
+ '<ul><li>Drupal<li>Drupal</ul>' => " * Drupal\n * Drupal\n\n",
+ '<ul>Drupal</ul>' => "Drupal\n\n",
+ 'Drupal</ul></ol></dl><li>Drupal' => "Drupal\n * Drupal\n",
+ '<dl>Drupal</dl>' => "Drupal\n\n",
+ '<dl>Drupal</dl><p>Drupal</p>' => "Drupal\n\nDrupal\n\n",
+ '<dt>Drupal</dt>' => "Drupal\n",
+ // Tests some unsupported HTML tags.
+ '<html>Drupal</html>' => "Drupal\n",
+ // @todo Perhaps the contents of <script> tags should be dropped.
+ '<script type="text/javascript">Drupal</script>' => "Drupal\n",
+ );
+
+ foreach ($tests as $html => $text) {
+ $this->assertHtmlToText($html, $text, 'Supported tags');
+ }
+ }
+
+ /**
+ * Test $allowed_tags argument of drupal_html_to_text().
+ */
+ function testDrupalHtmlToTextArgs() {
+ // The second parameter of drupal_html_to_text() overrules the allowed tags.
+ $this->assertHtmlToText(
+ 'Drupal <b>Drupal</b> Drupal',
+ "Drupal *Drupal* Drupal\n",
+ 'Allowed <b> tag found',
+ array('b')
+ );
+ $this->assertHtmlToText(
+ 'Drupal <h1>Drupal</h1> Drupal',
+ "Drupal Drupal Drupal\n",
+ 'Disallowed <h1> tag not found',
+ array('b')
+ );
+
+ $this->assertHtmlToText(
+ 'Drupal <p><em><b>Drupal</b></em><p> Drupal',
+ "Drupal Drupal Drupal\n",
+ 'Disallowed <p>, <em>, and <b> tags not found',
+ array('a', 'br', 'h1')
+ );
+
+ $this->assertHtmlToText(
+ '<html><body>Drupal</body></html>',
+ "Drupal\n",
+ 'Unsupported <html> and <body> tags not found',
+ array('html', 'body')
+ );
+ }
+
+ /**
+ * Test that whitespace is collapsed.
+ */
+ function testDrupalHtmltoTextCollapsesWhitespace() {
+ $input = "<p>Drupal Drupal\n\nDrupal<pre>Drupal Drupal\n\nDrupal</pre>Drupal Drupal\n\nDrupal</p>";
+ // @todo The whitespace should be collapsed.
+ $collapsed = "Drupal Drupal\n\nDrupalDrupal Drupal\n\nDrupalDrupal Drupal\n\nDrupal\n\n";
+ $this->assertHtmlToText(
+ $input,
+ $collapsed,
+ 'Whitespace is collapsed',
+ array('p')
+ );
+ }
+
+ /**
+ * Test that text separated by block-level tags in HTML get separated by
+ * (at least) a newline in the plaintext version.
+ */
+ function testDrupalHtmlToTextBlockTagToNewline() {
+ $input = '[text]'
+ . '<blockquote>[blockquote]</blockquote>'
+ . '<br />[br]'
+ . '<dl><dt>[dl-dt]</dt>'
+ . '<dt>[dt]</dt>'
+ . '<dd>[dd]</dd>'
+ . '<dd>[dd-dl]</dd></dl>'
+ . '<h1>[h1]</h1>'
+ . '<h2>[h2]</h2>'
+ . '<h3>[h3]</h3>'
+ . '<h4>[h4]</h4>'
+ . '<h5>[h5]</h5>'
+ . '<h6>[h6]</h6>'
+ . '<hr />[hr]'
+ . '<ol><li>[ol-li]</li>'
+ . '<li>[li]</li>'
+ . '<li>[li-ol]</li></ol>'
+ . '<p>[p]</p>'
+ . '<ul><li>[ul-li]</li>'
+ . '<li>[li-ul]</li></ul>'
+ . '[text]';
+ $output = drupal_html_to_text($input);
+ $pass = $this->assertFalse(
+ preg_match('/\][^\n]*\[/s', $output),
+ 'Block-level HTML tags should force newlines'
+ );
+ if (!$pass) {
+ $this->verbose($this->stringToHtml($output));
+ }
+ $output_upper = drupal_strtoupper($output);
+ $upper_input = drupal_strtoupper($input);
+ $upper_output = drupal_html_to_text($upper_input);
+ $pass = $this->assertEqual(
+ $upper_output,
+ $output_upper,
+ 'Tag recognition should be case-insensitive'
+ );
+ if (!$pass) {
+ $this->verbose(
+ $upper_output
+ . '<br />should be equal to <br />'
+ . $output_upper
+ );
+ }
+ }
+
+ /**
+ * Test that headers are properly separated from surrounding text.
+ */
+ function testHeaderSeparation() {
+ $html = 'Drupal<h1>Drupal</h1>Drupal';
+ // @todo There should be more space above the header than below it.
+ $text = "Drupal\n======== DRUPAL ==============================================================\n\nDrupal\n";
+ $this->assertHtmlToText($html, $text,
+ 'Text before and after <h1> tag');
+ $html = '<p>Drupal</p><h1>Drupal</h1>Drupal';
+ // @todo There should be more space above the header than below it.
+ $text = "Drupal\n\n======== DRUPAL ==============================================================\n\nDrupal\n";
+ $this->assertHtmlToText($html, $text,
+ 'Paragraph before and text after <h1> tag');
+ $html = 'Drupal<h1>Drupal</h1><p>Drupal</p>';
+ // @todo There should be more space above the header than below it.
+ $text = "Drupal\n======== DRUPAL ==============================================================\n\nDrupal\n\n";
+ $this->assertHtmlToText($html, $text,
+ 'Text before and paragraph after <h1> tag');
+ $html = '<p>Drupal</p><h1>Drupal</h1><p>Drupal</p>';
+ $text = "Drupal\n\n======== DRUPAL ==============================================================\n\nDrupal\n\n";
+ $this->assertHtmlToText($html, $text,
+ 'Paragraph before and after <h1> tag');
+ }
+
+ /**
+ * Test that footnote references are properly generated.
+ */
+ function testFootnoteReferences() {
+ global $base_path, $base_url;
+ $source = '<a href="http://www.example.com/node/1">Host and path</a>'
+ . '<br /><a href="http://www.example.com">Host, no path</a>'
+ . '<br /><a href="' . $base_path . 'node/1">Path, no host</a>'
+ . '<br /><a href="node/1">Relative path</a>';
+ // @todo Footnote urls should be absolute.
+ $tt = "Host and path [1]"
+ . "\nHost, no path [2]"
+ // @todo The following two references should be combined.
+ . "\nPath, no host [3]"
+ . "\nRelative path [4]"
+ . "\n"
+ . "\n[1] http://www.example.com/node/1"
+ . "\n[2] http://www.example.com"
+ // @todo The following two references should be combined.
+ . "\n[3] $base_url/node/1"
+ . "\n[4] node/1\n";
+ $this->assertHtmlToText($source, $tt, 'Footnotes');
+ }
+
+ /**
+ * Test that combinations of paragraph breaks, line breaks, linefeeds,
+ * and spaces are properly handled.
+ */
+ function testDrupalHtmlToTextParagraphs() {
+ $tests = array();
+ $tests[] = array(
+ 'html' => "<p>line 1<br />\nline 2<br />line 3\n<br />line 4</p><p>paragraph</p>",
+ // @todo Trailing line breaks should be trimmed.
+ 'text' => "line 1\nline 2\nline 3\nline 4\n\nparagraph\n\n",
+ );
+ $tests[] = array(
+ 'html' => "<p>line 1<br /> line 2</p> <p>line 4<br /> line 5</p> <p>0</p>",
+ // @todo Trailing line breaks should be trimmed.
+ 'text' => "line 1\nline 2\n\nline 4\nline 5\n\n0\n\n",
+ );
+ foreach ($tests as $test) {
+ $this->assertHtmlToText($test['html'], $test['text'], 'Paragraph breaks');
+ }
+ }
+
+ /**
+ * Tests that drupal_html_to_text() wraps before 1000 characters.
+ *
+ * RFC 3676 says, "The Text/Plain media type is the lowest common
+ * denominator of Internet email, with lines of no more than 998 characters."
+ *
+ * RFC 2046 says, "SMTP [RFC-821] allows a maximum of 998 octets before the
+ * next CRLF sequence."
+ *
+ * RFC 821 says, "The maximum total length of a text line including the
+ * <CRLF> is 1000 characters."
+ */
+ function testVeryLongLineWrap() {
+ $input = 'Drupal<br /><p>' . str_repeat('x', 2100) . '</><br />Drupal';
+ $output = drupal_html_to_text($input);
+ // This awkward construct comes from includes/mail.inc lines 8-13.
+ $eol = variable_get('mail_line_endings', MAIL_LINE_ENDINGS);
+ // We must use strlen() rather than drupal_strlen() in order to count
+ // octets rather than characters.
+ $line_length_limit = 1000 - drupal_strlen($eol);
+ $maximum_line_length = 0;
+ foreach (explode($eol, $output) as $line) {
+ // We must use strlen() rather than drupal_strlen() in order to count
+ // octets rather than characters.
+ $maximum_line_length = max($maximum_line_length, strlen($line . $eol));
+ }
+ $verbose = 'Maximum line length found was ' . $maximum_line_length . ' octets.';
+ // @todo This should assert that $maximum_line_length <= 1000.
+ $this->pass($verbose);
+ }
+}