Move getTags to BBCode

move getTags function to BBCode class and update calls.

diff --git a/mod/item.php b/mod/item.php
index bf8c5f7217170b9451dff977e785eb5d4f2cb004..193fa9ed9fba600b28674f3ee113b6547f9c80d5 100644 (file)
--- a/mod/item.php
+++ b/mod/item.php
@@ -348,7 +348,7 @@ function item_post(App $a) {
        $str_tags = '';
        $inform   = '';
 
-       $tags = Strings::getTags($body);
+       $tags = BBCode::getTags($body);
 
        // Add a tag if the parent contact is from ActivityPub or OStatus (This will notify them)
        if ($parent && in_array($thr_parent_contact['network'], [Protocol::OSTATUS, Protocol::ACTIVITYPUB])) {

diff --git a/mod/photos.php b/mod/photos.php
index b5c648d3bd49494d6e2a2fd4d6993496a8dd40ae..8b11e72421cf878e87c3f6c462c1b94c97136e3f 100644 (file)
--- a/mod/photos.php
+++ b/mod/photos.php
@@ -525,7 +525,7 @@ function photos_post(App $a)
                        }
 
                        $taginfo = [];
-                       $tags = Strings::getTags($rawtags);
+                       $tags = BBCode::getTags($rawtags);
 
                        if (count($tags)) {
                                foreach ($tags as $tag) {

diff --git a/src/Content/Text/BBCode.php b/src/Content/Text/BBCode.php
index 778d95cedb8685c30279f8944ef57ffc02d57a44..9c9adec0fa873037bd33bcc462347ceecf185dda 100644 (file)
--- a/src/Content/Text/BBCode.php
+++ b/src/Content/Text/BBCode.php
@@ -1911,4 +1911,78 @@ class BBCode extends BaseObject
 
                return $text;
        }
+
+       /**
+     * @brief Pull out all #hashtags and @person tags from $string.
+     *
+     * We also get @person@domain.com - which would make
+     * the regex quite complicated as tags can also
+     * end a sentence. So we'll run through our results
+     * and strip the period from any tags which end with one.
+     * Returns array of tags found, or empty array.
+     *
+     * @param string $string Post content
+     * 
+     * @return array List of tag and person names
+     */
+    public static function getTags($string)
+    {
+        $ret = [];
+
+        // Convert hashtag links to hashtags
+        $string = preg_replace('/#\[url\=([^\[\]]*)\](.*?)\[\/url\]/ism', '#$2', $string);
+
+        // ignore anything in a code block
+        $string = preg_replace('/\[code\](.*?)\[\/code\]/sm', '', $string);
+
+        // Force line feeds at bbtags
+        $string = str_replace(['[', ']'], ["\n[", "]\n"], $string);
+
+        // ignore anything in a bbtag
+        $string = preg_replace('/\[(.*?)\]/sm', '', $string);
+
+        // Match full names against @tags including the space between first and last
+        // We will look these up afterward to see if they are full names or not recognisable.
+
+        if (preg_match_all('/(@[^ \x0D\x0A,:?]+ [^ \x0D\x0A@,:?]+)([ \x0D\x0A@,:?]|$)/', $string, $matches)) {
+            foreach ($matches[1] as $match) {
+                if (strstr($match, ']')) {
+                    // we might be inside a bbcode color tag - leave it alone
+                    continue;
+                }
+
+                if (substr($match, -1, 1) === '.') {
+                    $ret[] = substr($match, 0, -1);
+                } else {
+                    $ret[] = $match;
+                }
+            }
+        }
+
+        // Otherwise pull out single word tags. These can be @nickname, @first_last
+        // and #hash tags.
+
+        if (preg_match_all('/([!#@][^\^ \x0D\x0A,;:?]+)([ \x0D\x0A,;:?]|$)/', $string, $matches)) {
+            foreach ($matches[1] as $match) {
+                if (strstr($match, ']')) {
+                    // we might be inside a bbcode color tag - leave it alone
+                    continue;
+                }
+                if (substr($match, -1, 1) === '.') {
+                    $match = substr($match,0,-1);
+                }
+                // ignore strictly numeric tags like #1
+                if ((strpos($match, '#') === 0) && ctype_digit(substr($match, 1))) {
+                    continue;
+                }
+                // try not to catch url fragments
+                if (strpos($string, $match) && preg_match('/[a-zA-z0-9\/]/', substr($string, strpos($string, $match) - 1, 1))) {
+                    continue;
+                }
+                $ret[] = $match;
+            }
+        }
+
+        return $ret;
+    }
 }

diff --git a/src/Model/Item.php b/src/Model/Item.php
index a96c77921d7183c51d7e89557fd0850e42bcc509..e12f6b6eedfc0b014512690be1bc2261ee0e804f 100644 (file)
--- a/src/Model/Item.php
+++ b/src/Model/Item.php
@@ -2403,7 +2403,7 @@ class Item extends BaseObject
        public static function setHashtags(&$item)
        {
 
-               $tags = Strings::getTags($item["body"]);
+               $tags = BBCode::getTags($item["body"]);
 
                // No hashtags?
                if (!count($tags)) {

diff --git a/src/Util/Strings.php b/src/Util/Strings.php
index 30f8b7329a0c25400338549920287852d29e2722..5cb1fd6b3a9ac0df10cb7fa84209cb44c17d8a02 100644 (file)
--- a/src/Util/Strings.php
+++ b/src/Util/Strings.php
@@ -158,7 +158,7 @@ class Strings
     {
         if ($network != "") {
             if ($url != "") {
-                $network_name = '<a href="'.$url.'">'.ContactSelector::networkToName($network, $url)."</a>";
+                $network_name = '<a href="' . $url  .'">' . ContactSelector::networkToName($network, $url) . "</a>";
             } else {
                 $network_name = ContactSelector::networkToName($network);
             }
@@ -279,80 +279,6 @@ class Strings
         return base64_decode(strtr($s, '-_', '+/'));
     }
 
-    /**
-     * @brief Pull out all #hashtags and @person tags from $string.
-     *
-     * We also get @person@domain.com - which would make
-     * the regex quite complicated as tags can also
-     * end a sentence. So we'll run through our results
-     * and strip the period from any tags which end with one.
-     * Returns array of tags found, or empty array.
-     *
-     * @param string $string Post content
-     * 
-     * @return array List of tag and person names
-     */
-    public static function getTags($string)
-    {
-        $ret = [];
-
-        // Convert hashtag links to hashtags
-        $string = preg_replace('/#\[url\=([^\[\]]*)\](.*?)\[\/url\]/ism', '#$2', $string);
-
-        // ignore anything in a code block
-        $string = preg_replace('/\[code\](.*?)\[\/code\]/sm', '', $string);
-
-        // Force line feeds at bbtags
-        $string = str_replace(['[', ']'], ["\n[", "]\n"], $string);
-
-        // ignore anything in a bbtag
-        $string = preg_replace('/\[(.*?)\]/sm', '', $string);
-
-        // Match full names against @tags including the space between first and last
-        // We will look these up afterward to see if they are full names or not recognisable.
-
-        if (preg_match_all('/(@[^ \x0D\x0A,:?]+ [^ \x0D\x0A@,:?]+)([ \x0D\x0A@,:?]|$)/', $string, $matches)) {
-            foreach ($matches[1] as $match) {
-                if (strstr($match, ']')) {
-                    // we might be inside a bbcode color tag - leave it alone
-                    continue;
-                }
-
-                if (substr($match, -1, 1) === '.') {
-                    $ret[] = substr($match, 0, -1);
-                } else {
-                    $ret[] = $match;
-                }
-            }
-        }
-
-        // Otherwise pull out single word tags. These can be @nickname, @first_last
-        // and #hash tags.
-
-        if (preg_match_all('/([!#@][^\^ \x0D\x0A,;:?]+)([ \x0D\x0A,;:?]|$)/', $string, $matches)) {
-            foreach ($matches[1] as $match) {
-                if (strstr($match, ']')) {
-                    // we might be inside a bbcode color tag - leave it alone
-                    continue;
-                }
-                if (substr($match, -1, 1) === '.') {
-                    $match = substr($match,0,-1);
-                }
-                // ignore strictly numeric tags like #1
-                if ((strpos($match, '#') === 0) && ctype_digit(substr($match, 1))) {
-                    continue;
-                }
-                // try not to catch url fragments
-                if (strpos($string, $match) && preg_match('/[a-zA-z0-9\/]/', substr($string, strpos($string, $match) - 1, 1))) {
-                    continue;
-                }
-                $ret[] = $match;
-            }
-        }
-
-        return $ret;
-    }
-
     /**
      * @brief Normalize url
      *