Merge pull request #2566 from marc1706/ticket/11711

[ticket/11711] Inform user of unsupported characters while posting

* marc1706/ticket/11711:
  [ticket/11711] Improve coding and comments of character check
  [ticket/11711] Reduce size of tests and also test topic titles
  [ticket/11711] Improve checks for unsupported characters and check subject
  [ticket/11711] Inform user of unsupported characters while posting

phpBB/includes/message_parser.php

@@ -1194,6 +1194,15 @@ class parse_message extends bbcode_firstpass
 			}
 		}
 
+		// Check for out-of-bounds characters that are currently
+		// not supported by utf8_bin in MySQL
+		if (preg_match_all('/[\x{10000}-\x{10FFFF}]/u', $this->message, $matches))
+		{
+			$character_list = implode('<br />', $matches[0]);
+			$this->warn_msg[] = $user->lang('UNSUPPORTED_CHARACTERS_MESSAGE', $character_list);
+			return $update_this_message ? $this->warn_msg : $return_message;
+		}
+
 		// Check for "empty" message. We do not check here for maximum length, because bbcode, smilies, etc. can add to the length.
 		// The maximum length check happened before any parsings.
 		if ($mode === 'post' && utf8_clean_string($this->message) === '')

phpBB/language/en/posting.php

@@ -256,6 +256,8 @@ $lang = array_merge($lang, array(
 
 	'UNAUTHORISED_BBCODE'		=> 'You cannot use certain BBCodes: %s.',
 	'UNGLOBALISE_EXPLAIN'		=> 'To switch this topic back from being global to a normal topic, you need to select the forum you wish this topic to be displayed.',
+	'UNSUPPORTED_CHARACTERS_MESSAGE'	=> 'Your message contains the following unsupported characters:<br />%s',
+	'UNSUPPORTED_CHARACTERS_SUBJECT'	=> 'Your subject contains the following unsupported characters:<br />%s',
 	'UPDATE_COMMENT'			=> 'Update comment',
 	'URL_INVALID'				=> 'The URL you specified is invalid.',
 	'URL_NOT_FOUND'				=> 'The file specified could not be found.',

phpBB/posting.php

@@ -1028,6 +1028,14 @@ if ($submit || $preview || $refresh)
 		$error[] = $user->lang['EMPTY_SUBJECT'];
 	}
 
+	// Check for out-of-bounds characters that are currently
+	// not supported by utf8_bin in MySQL
+	if (preg_match_all('/[\x{10000}-\x{10FFFF}]/u', $post_data['post_subject'], $matches))
+	{
+		$character_list = implode('<br />', $matches[0]);
+		$error[] = $user->lang('UNSUPPORTED_CHARACTERS_SUBJECT', $character_list);
+	}
+
 	$post_data['poll_last_vote'] = (isset($post_data['poll_last_vote'])) ? $post_data['poll_last_vote'] : 0;
 
 	if ($post_data['poll_option_text'] &&

tests/functional/posting_test.php

@@ -36,4 +36,27 @@ class phpbb_functional_posting_test extends phpbb_functional_test_case
 		$crawler = self::request('GET', "posting.php?mode=quote&f=2&t={$post2['topic_id']}&p={$post2['post_id']}&sid={$this->sid}");
 		$this->assertContains('This is a test post posted by the testing framework.', $crawler->filter('html')->text());
 	}
+
+	public function test_unsupported_characters()
+	{
+		$this->login();
+
+		$this->add_lang('posting');
+
+		self::create_post(2,
+			1,
+			'Unsupported characters',
+			"This is a test with these weird characters: \xF0\x9F\x88\xB3 \xF0\x9F\x9A\xB6",
+			array(),
+			'Your message contains the following unsupported characters'
+		);
+
+		self::create_post(2,
+			1,
+			"Unsupported: \xF0\x9F\x88\xB3 \xF0\x9F\x9A\xB6",
+			'This is a test with emoji characters in the topic title.',
+			array(),
+			'Your subject contains the following unsupported characters'
+		);
+	}
 }

tests/test_framework/phpbb_functional_test_case.php

@@ -1075,10 +1075,18 @@ class phpbb_functional_test_case extends phpbb_test_case
 		$crawler = self::request('POST', $posting_url, $form_data);
 
 		if ($expected !== '')
+		{
+			if (isset($this->lang[$expected]))
 			{
 				$this->assertContainsLang($expected, $crawler->filter('html')->text());
+			}
+			else
+			{
+				$this->assertContains($expected, $crawler->filter('html')->text());
+			}
 			return null;
 		}
+
 		$url = $crawler->selectLink($form_data['subject'])->link()->getUri();
 
 		return array(