X-Git-Url: https://git.immae.eu/?a=blobdiff_plain;f=application%2Fbookmark%2FLinkUtils.php;h=cf97e3b0a904a08d72a5ee7d3467efb1536f5c96;hb=d9d71b10c3bc70a0881d630b37dc4e918c9e812f;hp=03e1b82a7ec4d69d55c066a810c6a8fd15bab2e7;hpb=2cd0509b503332b1989f06da45d569d4d2929be5;p=github%2Fshaarli%2FShaarli.git diff --git a/application/bookmark/LinkUtils.php b/application/bookmark/LinkUtils.php index 03e1b82a..cf97e3b0 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($match[1])); } @@ -68,16 +68,16 @@ function html_extract_tag($tag, $html) $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 . '[^\'"]*?[\'"]'; - // Try to retrieve OpenGraph image. - $ogRegex = '#]+(?:'. $properties .')=(?:'. $orCondition .')[^>]*content=["\'](.*?)["\'].*?>#'; + // 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 .')=(?:'. $orCondition .').*?>#'; + $ogRegexReverse = '#]+content=(["\'])([^\1]*?)\1[^>]+(?:'. $properties .')=(?:'. $orCondition .').*?>#'; if (preg_match($ogRegex, $html, $matches) > 0 || preg_match($ogRegexReverse, $html, $matches) > 0 ) { - return $matches[1]; + return $matches[2]; } return false; @@ -138,12 +138,17 @@ function space2nbsp($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)); } /** @@ -171,3 +176,49 @@ 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 ?? []))); +}