Knah-Tsaeb/MyShaarli
Knah-Tsaeb/MyShaarli
1
0
Fork 0
Code Issues Pull requests Releases Wiki Activity

Merge pull request #1631 from ArthurHoaro/fix/html-extract-quote-fix

Browse source 
Fix an issue truncating extracted metadata content
This commit is contained in:
ArthurHoaro 2020-11-08 14:06:38 +01:00 committed by GitHub
commit 8d8fa898ab
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
2 changed files with 34 additions and 4 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

8
application/bookmark/LinkUtils.php
View file

@ -68,16 +68,16 @@ function html_extract_tag($tag, $html)
$properties = implode('|', $propertiesKey); $properties = implode('|', $propertiesKey);
// We need a OR here to accept either 'property=og:noquote' or 'property="og:unrelated og:my-tag"' // We need a OR here to accept either 'property=og:noquote' or 'property="og:unrelated og:my-tag"'
$orCondition = '["\']?(?:og:)?'. $tag .'["\']?|["\'][^\'"]*?(?:og:)?' . $tag . '[^\'"]*?[\'"]'; $orCondition = '["\']?(?:og:)?'. $tag .'["\']?|["\'][^\'"]*?(?:og:)?' . $tag . '[^\'"]*?[\'"]';
// Try to retrieve OpenGraph image. // Try to retrieve OpenGraph tag.
$ogRegex = '#<meta[^>]+(?:'. $properties .')=(?:'. $orCondition .')[^>]*content=["\'](.*?)["\'].*?>#'; $ogRegex = '#<meta[^>]+(?:'. $properties .')=(?:'. $orCondition .')[^>]*content=(["\'])([^\1]*?)\1.*?>#';
// If the attributes are not in the order property => content (e.g. Github) // If the attributes are not in the order property => content (e.g. Github)
// New regex to keep this readable... more or less. // New regex to keep this readable... more or less.
$ogRegexReverse = '#<meta[^>]+content=["\'](.*?)["\'][^>]+(?:'. $properties .')=(?:'. $orCondition .').*?>#'; $ogRegexReverse = '#<meta[^>]+content=(["\'])([^\1]*?)\1[^>]+(?:'. $properties .')=(?:'. $orCondition .').*?>#';
if (preg_match($ogRegex, $html, $matches) > 0 if (preg_match($ogRegex, $html, $matches) > 0
|| preg_match($ogRegexReverse, $html, $matches) > 0 || preg_match($ogRegexReverse, $html, $matches) > 0
) { ) {
return $matches[1]; return $matches[2];
} }
return false; return false;

30
tests/bookmark/LinkUtilsTest.php
View file

@ -168,6 +168,36 @@ public function testHtmlExtractExistentNameTag()
$this->assertEquals($description, html_extract_tag('description', $html)); $this->assertEquals($description, html_extract_tag('description', $html));
} }
/**
* Test html_extract_tag() with double quoted content containing single quote, and the opposite.
*/
public function testHtmlExtractExistentNameTagWithMixedQuotes(): void
{
$description = 'Bob and Alice share M&M\'s.';
$html = '<meta property="og:description" content="' . $description . '">';
$this->assertEquals($description, html_extract_tag('description', $html));
$html = '<meta tag1="content1" property="og:unrelated1 og:description og:unrelated2" '.
'tag2="content2" content="' . $description . '" tag3="content3">';
$this->assertEquals($description, html_extract_tag('description', $html));
$html = '<meta property="og:description" name="description" content="' . $description . '">';
$this->assertEquals($description, html_extract_tag('description', $html));
$description = 'Bob and Alice share "cookies".';
$html = '<meta property="og:description" content=\'' . $description . '\'>';
$this->assertEquals($description, html_extract_tag('description', $html));
$html = '<meta tag1="content1" property="og:unrelated1 og:description og:unrelated2" '.
'tag2="content2" content=\'' . $description . '\' tag3="content3">';
$this->assertEquals($description, html_extract_tag('description', $html));
$html = '<meta property="og:description" name="description" content=\'' . $description . '\'>';
$this->assertEquals($description, html_extract_tag('description', $html));
}
/** /**
* Test html_extract_tag() when the tag <meta name= is not found. * Test html_extract_tag() when the tag <meta name= is not found.
*/ */