X-Git-Url: https://git.immae.eu/?a=blobdiff_plain;f=application%2Fbookmark%2FLinkUtils.php;h=d65e97ed47ca9163bb8f0f9feef66969b1981166;hb=8a6b7e96b7176e03238bbb1bcaa4c8b0c25e6358;hp=a74fda5795c09e9f0fd6b3ed1e8d98b81226bc5d;hpb=8d8fa898ab09da87d6c5b333ce6c5b7e8d1506d0;p=github%2Fshaarli%2FShaarli.git
diff --git a/application/bookmark/LinkUtils.php b/application/bookmark/LinkUtils.php
index a74fda57..d65e97ed 100644
--- a/application/bookmark/LinkUtils.php
+++ b/application/bookmark/LinkUtils.php
@@ -67,14 +67,15 @@ function html_extract_tag($tag, $html)
$propertiesKey = ['property', 'name', 'itemprop'];
$properties = implode('|', $propertiesKey);
// 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 tag.
- $ogRegex = '#]+(?:'. $properties .')=(?:'. $orCondition .')[^>]*content=(["\'])([^\1]*?)\1.*?>#';
+ $ogRegex = '#]+(?:' . $properties . ')=(?:' . $orCondition . ')[^>]*content=(["\'])([^\1]*?)\1.*?>#';
// If the attributes are not in the order property => content (e.g. Github)
// New regex to keep this readable... more or less.
- $ogRegexReverse = '#]+content=(["\'])([^\1]*?)\1[^>]+(?:'. $properties .')=(?:'. $orCondition .').*?>#';
+ $ogRegexReverse = '#]+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
) {
return $matches[2];
@@ -116,7 +117,7 @@ function hashtag_autolink($description, $indexUrl = '')
* \p{Mn} - any non marking space (accents, umlauts, etc)
*/
$regex = '/(^|\s)#([\p{Pc}\p{N}\p{L}\p{Mn}]+)/mui';
- $replacement = '$1#$2';
+ $replacement = '$1#$2';
return preg_replace($regex, $replacement, $description);
}
@@ -176,3 +177,49 @@ function is_note($linkUrl)
{
return isset($linkUrl[0]) && $linkUrl[0] === '?';
}
+
+/**
+ * Extract an array of tags from a given tag string, with provided separator.
+ *
+ * @param string|null $tags String containing a list of tags separated by $separator.
+ * @param string $separator Shaarli's default: ' ' (whitespace)
+ *
+ * @return array List of tags
+ */
+function tags_str2array(?string $tags, string $separator): array
+{
+ // For whitespaces, we use the special \s regex character
+ $separator = $separator === ' ' ? '\s' : $separator;
+
+ return preg_split('/\s*' . $separator . '+\s*/', trim($tags) ?? '', -1, PREG_SPLIT_NO_EMPTY);
+}
+
+/**
+ * Return a tag string with provided separator from a list of tags.
+ * Note that given array is clean up by tags_filter().
+ *
+ * @param array|null $tags List of tags
+ * @param string $separator
+ *
+ * @return string
+ */
+function tags_array2str(?array $tags, string $separator): string
+{
+ return implode($separator, tags_filter($tags, $separator));
+}
+
+/**
+ * Clean an array of tags: trim + remove empty entries
+ *
+ * @param array|null $tags List of tags
+ * @param string $separator
+ *
+ * @return array
+ */
+function tags_filter(?array $tags, string $separator): array
+{
+ $trimDefault = " \t\n\r\0\x0B";
+ return array_values(array_filter(array_map(function (string $entry) use ($separator, $trimDefault): string {
+ return trim($entry, $trimDefault . $separator);
+ }, $tags ?? [])));
+}