X-Git-Url: https://git.immae.eu/?a=blobdiff_plain;f=application%2Fbookmark%2FLinkUtils.php;h=8fa2953a78d3e8f18bbcba3824801d859169c16e;hb=9ef8555ad298668bcb8537ccdd2ab6560f44177f;hp=e7af4d552b409b77595985cc635273a6a1d8e448;hpb=ee07b7283faa197fc062ed85f4f96f98e8e77b03;p=github%2Fshaarli%2FShaarli.git

diff --git a/application/bookmark/LinkUtils.php b/application/bookmark/LinkUtils.php
index e7af4d55..8fa2953a 100644
--- a/application/bookmark/LinkUtils.php
+++ b/application/bookmark/LinkUtils.php
@@ -1,6 +1,7 @@
 <?php
 
 use Shaarli\Bookmark\Bookmark;
+use Shaarli\Formatter\BookmarkDefaultFormatter;
 
 /**
  * Extract title from an HTML document.
@@ -66,16 +67,21 @@ function html_extract_tag($tag, $html)
 {
     $propertiesKey = ['property', 'name', 'itemprop'];
     $properties = implode('|', $propertiesKey);
-    // Try to retrieve OpenGraph image.
-    $ogRegex = '#<meta[^>]+(?:'. $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 . '[^\'"]*?[\'"]';
+    // Support quotes in double quoted content, and the other way around
+    $content = 'content=(["\'])((?:(?!\1).)*)\1';
+    // Try to retrieve OpenGraph tag.
+    $ogRegex = '#<meta[^>]+(?:' . $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 = '#<meta[^>]+content=["\']([^"\']+)[^>]+(?:'. $properties .')=["\']?(?:og)?:'. $tag .'["\'\s/>]#';
+    $ogRegexReverse = '#<meta[^>]+' . $content . '[^>]+(?:' . $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;
@@ -93,7 +99,18 @@ function html_extract_tag($tag, $html)
 function text2clickable($text)
 {
     $regex = '!(((?:https?|ftp|file)://|apt:|magnet:)\S+[a-z0-9\(\)]/?)!si';
-    return preg_replace($regex, '<a href="$1">$1</a>', $text);
+    $format = function (array $match): string {
+        return '<a href="' .
+            str_replace(
+                BookmarkDefaultFormatter::SEARCH_HIGHLIGHT_OPEN,
+                '',
+                str_replace(BookmarkDefaultFormatter::SEARCH_HIGHLIGHT_CLOSE, '', $match[1])
+            ) .
+            '">' . $match[1] . '</a>'
+        ;
+    };
+
+    return preg_replace_callback($regex, $format, $text);
 }
 
 /**
@@ -106,6 +123,9 @@ function text2clickable($text)
  */
 function hashtag_autolink($description, $indexUrl = '')
 {
+    $tokens = '(?:' . BookmarkDefaultFormatter::SEARCH_HIGHLIGHT_OPEN . ')' .
+              '(?:' . BookmarkDefaultFormatter::SEARCH_HIGHLIGHT_CLOSE . ')'
+    ;
     /*
      * To support unicode: http://stackoverflow.com/a/35498078/1484919
      * \p{Pc} - to match underscore
@@ -113,9 +133,20 @@ function hashtag_autolink($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<a href="'. $indexUrl .'./add-tag/$2" title="Hashtag $2">#$2</a>';
-    return preg_replace($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] . '<a href="' . $indexUrl . './add-tag/' . $cleanMatch . '"' .
+            ' title="Hashtag ' . $cleanMatch . '">' .
+                '#' . $match[2] .
+        '</a>';
+    };
+
+    return preg_replace_callback($regex, $format, $description);
 }
 
 /**
@@ -136,12 +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));
 }
 
 /**
@@ -169,3 +205,49 @@ 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 ?? [])));
+}