X-Git-Url: https://git.immae.eu/?a=blobdiff_plain;f=application%2Fbookmark%2FLinkUtils.php;h=d65e97ed47ca9163bb8f0f9feef66969b1981166;hb=831e974ea5fa93d689e65313f84d0c80999674c3;hp=68914fcab749a19b1ba15193decd99247158375a;hpb=06f05c923ae59e5daa1aaa8d1ad4c50bd9064bb2;p=github%2Fshaarli%2FShaarli.git
diff --git a/application/bookmark/LinkUtils.php b/application/bookmark/LinkUtils.php
index 68914fca..d65e97ed 100644
--- a/application/bookmark/LinkUtils.php
+++ b/application/bookmark/LinkUtils.php
@@ -26,7 +26,7 @@ function html_extract_title($html)
*/
function header_extract_charset($header)
{
- preg_match('/charset="?([^; ]+)/i', $header, $match);
+ preg_match('/charset=["\']?([^; "\']+)/i', $header, $match);
if (! empty($match[1])) {
return strtolower(trim($match[1]));
}
@@ -66,16 +66,19 @@ function html_extract_tag($tag, $html)
{
$propertiesKey = ['property', 'name', 'itemprop'];
$properties = implode('|', $propertiesKey);
- // Try to retrieve OpenGraph image.
- $ogRegex = '#]+(?:'. $properties .')=["\']?(?:og:)?'. $tag .'["\'\s][^>]*content=["\']?(.*?)["\'/>]#';
+ // We need a OR here to accept either 'property=og:noquote' or 'property="og:unrelated og:my-tag"'
+ $orCondition = '["\']?(?:og:)?' . $tag . '["\']?|["\'][^\'"]*?(?:og:)?' . $tag . '[^\'"]*?[\'"]';
+ // Try to retrieve OpenGraph tag.
+ $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=["\']([^"\']+)[^>]+(?:'. $properties .')=["\']?(?:og)?:'. $tag .'["\'\s/>]#';
+ $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[1];
+ return $matches[2];
}
return false;
@@ -114,7 +117,7 @@ function hashtag_autolink($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, $description);
}
@@ -136,12 +139,17 @@ function space2nbsp($text)
*
* @param string $description shaare's description.
* @param string $indexUrl URL to Shaarli's index.
-
+ * @param bool $autolink Turn on/off automatic linkifications of URLs and hashtags
+ *
* @return string formatted description.
*/
-function format_description($description, $indexUrl = '')
+function format_description($description, $indexUrl = '', $autolink = true)
{
- return nl2br(space2nbsp(hashtag_autolink(text2clickable($description), $indexUrl)));
+ if ($autolink) {
+ $description = hashtag_autolink(text2clickable($description), $indexUrl);
+ }
+
+ return nl2br(space2nbsp($description));
}
/**
@@ -169,3 +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 ?? [])));
+}