spelling: occurred

[friendica.git] / src / Content / Text / HTML.php

diff --git a/src/Content/Text/HTML.php b/src/Content/Text/HTML.php
index 557d62e12ac6d6b8c7e082ce1200b127b04327fb..89a680501f1674761d0c394cf1bc17799a50a6b5 100644 (file)
--- a/src/Content/Text/HTML.php
+++ b/src/Content/Text/HTML.php
@@ -23,6 +23,7 @@ namespace Friendica\Content\Text;
 
 use DOMDocument;
 use DOMXPath;
+use Friendica\Protocol\HTTP\MediaType;
 use Friendica\Content\Widget\ContactBlock;
 use Friendica\Core\Hook;
 use Friendica\Core\Renderer;
@@ -280,9 +281,9 @@ class HTML
                        self::tagToBBCode($doc, 'div', [], "\r", "\r");
                        self::tagToBBCode($doc, 'p', [], "\n", "\n");
 
-                       self::tagToBBCode($doc, 'ul', [], "[list]", "[/list]");
-                       self::tagToBBCode($doc, 'ol', [], "[list=1]", "[/list]");
-                       self::tagToBBCode($doc, 'li', [], "[*]", "");
+                       self::tagToBBCode($doc, 'ul', [], "[ul]", "\n[/ul]");
+                       self::tagToBBCode($doc, 'ol', [], "[ol]", "\n[/ol]");
+                       self::tagToBBCode($doc, 'li', [], "\n[li]", "[/li]");
 
                        self::tagToBBCode($doc, 'hr', [], "[hr]", "");
 
@@ -348,33 +349,6 @@ class HTML
                                $message = str_replace("\n\n\n", "\n\n", $message);
                        } while ($oldmessage != $message);
 
-                       do {
-                               $oldmessage = $message;
-                               $message = str_replace(
-                                       [
-                                               "[/size]\n\n",
-                                               "\n[hr]",
-                                               "[hr]\n",
-                                               "\n[list",
-                                               "[/list]\n",
-                                               "\n[/",
-                                               "[list]\n",
-                                               "[list=1]\n",
-                                               "\n[*]"],
-                                       [
-                                               "[/size]\n",
-                                               "[hr]",
-                                               "[hr]",
-                                               "[list",
-                                               "[/list]",
-                                               "[/",
-                                               "[list]",
-                                               "[list=1]",
-                                               "[*]"],
-                                       $message
-                               );
-                       } while ($message != $oldmessage);
-
                        $message = str_replace(
                                ['[b][b]', '[/b][/b]', '[i][i]', '[/i][/i]'],
                                ['[b]', '[/b]', '[i]', '[/i]'],
@@ -1031,7 +1005,7 @@ class HTML
                // the quotes, e.g.:
                //
                //    concat("'foo'", '"', "bar")
-               return 'concat(' . implode(', \'"\', ', array_map(['self', 'xpathQuote'], explode('"', $value))) . ')';
+               return 'concat(' . implode(', \'"\', ', array_map([self::class, 'xpathQuote'], explode('"', $value))) . ')';
        }
 
        /**
@@ -1055,4 +1029,30 @@ class HTML
 
                return $result !== false && $result->length > 0;
        }
+
+       /**
+        * @param DOMDocument $doc
+        * @return string|null Lowercase charset
+        */
+       public static function extractCharset(DOMDocument $doc): ?string
+       {
+               $xpath = new DOMXPath($doc);
+
+               $expression = "string(//meta[@charset]/@charset)";
+               if ($charset = $xpath->evaluate($expression)) {
+                       return strtolower($charset);
+               }
+
+               try {
+                       // This expression looks for a meta tag with the http-equiv attribute set to "content-type" ignoring case
+                       // whose content attribute contains a "charset" string and returns its value
+                       $expression = "string(//meta[@http-equiv][translate(@http-equiv, 'ABCDEFGHIJKLMNOPQRSTUVWXYZ', 'abcdefghijklmnopqrstuvwxyz') = 'content-type'][contains(translate(@content, 'ABCDEFGHIJKLMNOPQRSTUVWXYZ', 'abcdefghijklmnopqrstuvwxyz'), 'charset')]/@content)";
+                       $mediaType = MediaType::fromContentType($xpath->evaluate($expression));
+                       if (isset($mediaType->parameters['charset'])) {
+                               return strtolower($mediaType->parameters['charset']);
+                       }
+               } catch(\InvalidArgumentException $e) {}
+
+               return null;
+       }
 }