X-Git-Url: https://git.immae.eu/?a=blobdiff_plain;f=application%2Fbookmark%2FLinkUtils.php;h=8fa2953a78d3e8f18bbcba3824801d859169c16e;hb=9ef8555ad298668bcb8537ccdd2ab6560f44177f;hp=988970bdf924854ba06fa352a61a1d3c62da94bd;hpb=5bb384cd2742ff4f982cbe914cee551b74ab4bad;p=github%2Fshaarli%2FShaarli.git
diff --git a/application/bookmark/LinkUtils.php b/application/bookmark/LinkUtils.php
index 988970bd..8fa2953a 100644
--- a/application/bookmark/LinkUtils.php
+++ b/application/bookmark/LinkUtils.php
@@ -1,64 +1,7 @@
+ * - Meta tag:
*
- * @param array|Countable $links Linklist.
+ * @param string $tag Name of the tag to retrieve.
+ * @param string $html HTML content where to look for charset.
*
- * @return int Number of private links.
+ * @return bool|string Charset string if found, false otherwise.
*/
-function count_private($links)
+function html_extract_tag($tag, $html)
{
- $cpt = 0;
- foreach ($links as $link) {
- if ($link['private']) {
- $cpt += 1;
- }
+ $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 . '[^\'"]*?[\'"]';
+ // Support quotes in double quoted content, and the other way around
+ $content = 'content=(["\'])((?:(?!\1).)*)\1';
+ // Try to retrieve OpenGraph tag.
+ $ogRegex = '#]+(?:' . $properties . ')=(?:' . $orCondition . ')[^>]*' . $content . '.*?>#';
+ // 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 . ')=(?:' . $orCondition . ').*?>#';
+
+ if (
+ preg_match($ogRegex, $html, $matches) > 0
+ || preg_match($ogRegexReverse, $html, $matches) > 0
+ ) {
+ return $matches[2];
}
- return $cpt;
+ return false;
}
/**
- * In a string, converts URLs to clickable links.
+ * In a string, converts URLs to clickable bookmarks.
*
* @param string $text input string.
*
- * @return string returns $text with all links converted to HTML links.
+ * @return string returns $text with all bookmarks converted to HTML bookmarks.
*
* @see Function inspired from http://www.php.net/manual/en/function.preg-replace.php#85722
*/
function text2clickable($text)
{
$regex = '!(((?:https?|ftp|file)://|apt:|magnet:)\S+[a-z0-9\(\)]/?)!si';
- return preg_replace($regex, '$1', $text);
+ $format = function (array $match): string {
+ return '' . $match[1] . ''
+ ;
+ };
+
+ return preg_replace_callback($regex, $format, $text);
}
/**
@@ -154,6 +123,9 @@ function text2clickable($text)
*/
function hashtag_autolink($description, $indexUrl = '')
{
+ $tokens = '(?:' . BookmarkDefaultFormatter::SEARCH_HIGHLIGHT_OPEN . ')' .
+ '(?:' . BookmarkDefaultFormatter::SEARCH_HIGHLIGHT_CLOSE . ')'
+ ;
/*
* To support unicode: http://stackoverflow.com/a/35498078/1484919
* \p{Pc} - to match underscore
@@ -161,9 +133,20 @@ function hashtag_autolink($description, $indexUrl = '')
* \p{L} - letter from any language
* \p{Mn} - any non marking space (accents, umlauts, etc)
*/
- $regex = '/(^|\s)#([\p{Pc}\p{N}\p{L}\p{Mn}]+)/mui';
- $replacement = '$1#$2';
- return preg_replace($regex, $replacement, $description);
+ $regex = '/(^|\s)#([\p{Pc}\p{N}\p{L}\p{Mn}' . $tokens . ']+)/mui';
+ $format = function (array $match) use ($indexUrl): string {
+ $cleanMatch = str_replace(
+ BookmarkDefaultFormatter::SEARCH_HIGHLIGHT_OPEN,
+ '',
+ str_replace(BookmarkDefaultFormatter::SEARCH_HIGHLIGHT_CLOSE, '', $match[2])
+ );
+ return $match[1] . '' .
+ '#' . $match[2] .
+ '';
+ };
+
+ return preg_replace_callback($regex, $format, $description);
}
/**
@@ -184,12 +167,17 @@ function space2nbsp($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));
}
/**
@@ -202,5 +190,64 @@ function format_description($description, $indexUrl = '')
*/
function link_small_hash($date, $id)
{
- return smallHash($date->format(LinkDB::LINK_DATE_FORMAT) . $id);
+ return smallHash($date->format(Bookmark::LINK_DATE_FORMAT) . $id);
+}
+
+/**
+ * Returns whether or not the link is an internal note.
+ * Its URL starts by `?` because it's actually a permalink.
+ *
+ * @param string $linkUrl
+ *
+ * @return bool true if internal note, false otherwise.
+ */
+function is_note($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 ?? [])));
}