From d8484e65bdff720e166f9b230d4b788bea64e7cd Mon Sep 17 00:00:00 2001 From: Hypolite Petovan Date: Wed, 25 Sep 2019 20:25:42 -0400 Subject: [PATCH 1/2] Move autolinker execution earlier in BBCode::convert - Prevents already replaced HTML links (like in [audio] tag) to be doubled - Add test for it --- src/Content/Text/BBCode.php | 7 +++++-- tests/src/Content/Text/BBCodeTest.php | 15 +++++++++++---- 2 files changed, 16 insertions(+), 6 deletions(-) diff --git a/src/Content/Text/BBCode.php b/src/Content/Text/BBCode.php index 02c9551027..c5f9986f1f 100644 --- a/src/Content/Text/BBCode.php +++ b/src/Content/Text/BBCode.php @@ -1466,6 +1466,11 @@ class BBCode extends BaseObject $text = str_replace('[hr]', '
', $text); + if (!$for_plaintext) { + // Autolinker for isolated URLs + $text = preg_replace(Strings::autoLinkRegEx(), '[url]$1[/url]', $text); + } + // This is actually executed in Item::prepareBody() $nosmile = strpos($text, '[nosmile]') !== false; @@ -1648,9 +1653,7 @@ class BBCode extends BaseObject $text = Smilies::replace($text); } - // if the HTML is used to generate plain text, then don't do this search, but replace all URL of that kind to text if (!$for_plaintext) { - $text = preg_replace(Strings::autoLinkRegEx(), '[url]$1[/url]', $text); if (in_array($simple_html, [7, 9])) { $text = preg_replace_callback("/\[url\](.*?)\[\/url\]/ism", 'self::convertUrlForOStatusCallback', $text); $text = preg_replace_callback("/\[url\=(.*?)\](.*?)\[\/url\]/ism", 'self::convertUrlForOStatusCallback', $text); diff --git a/tests/src/Content/Text/BBCodeTest.php b/tests/src/Content/Text/BBCodeTest.php index 3d8beb3930..1ff653db91 100644 --- a/tests/src/Content/Text/BBCodeTest.php +++ b/tests/src/Content/Text/BBCodeTest.php @@ -179,41 +179,48 @@ class BBCodeTest extends MockedTest 'bug-2199-named-size' => [ 'expectedHtml' => 'Test text', 'text' => '[size=xx-large]Test text[/size]', - 'simpleHtml' => 0, ], 'bug-2199-numeric-size' => [ 'expectedHtml' => 'Test text', 'text' => '[size=24]Test text[/size]', - 'simpleHtml' => 0, ], 'bug-2199-diaspora-no-named-size' => [ 'expectedHtml' => 'Test text', 'text' => '[size=xx-large]Test text[/size]', + 'try_oembed' => false, // Triggers the diaspora compatible output 'simpleHtml' => 3, ], 'bug-2199-diaspora-no-numeric-size' => [ 'expectedHtml' => 'Test text', 'text' => '[size=24]Test text[/size]', + 'try_oembed' => false, // Triggers the diaspora compatible output 'simpleHtml' => 3, ], + 'bug-7665-audio-tag' => [ + 'expectedHtml' => '', + 'text' => '[audio]http://www.cendrones.fr/colloque2017/jonathanbocquet.mp3[/audio]', + 'try_oembed' => true, + ], ]; } /** * Test convert bbcodes to HTML + * * @dataProvider dataBBCodes * * @param string $expectedHtml Expected HTML output * @param string $text BBCode text + * @param bool $try_oembed Whether to convert multimedia BBCode tag * @param int $simpleHtml BBCode::convert method $simple_html parameter value, optional. * @param bool $forPlaintext BBCode::convert method $for_plaintext parameter value, optional. * @throws \Friendica\Network\HTTPException\InternalServerErrorException */ - public function testConvert($expectedHtml, $text, $simpleHtml = 0, $forPlaintext = false) + public function testConvert($expectedHtml, $text, $try_oembed = false, $simpleHtml = 0, $forPlaintext = false) { - $actual = BBCode::convert($text, false, $simpleHtml, $forPlaintext); + $actual = BBCode::convert($text, $try_oembed, $simpleHtml, $forPlaintext); $this->assertEquals($expectedHtml, $actual); } From 1f20a9330c54e976df37340f6e0ae532c4016225 Mon Sep 17 00:00:00 2001 From: Hypolite Petovan Date: Wed, 25 Sep 2019 20:44:09 -0400 Subject: [PATCH 2/2] Add new ignoreChildren behavior to HTML::tagToBBCode - Allow discarding HTML tag children for single-value BBCode - Add test for it --- src/Content/Text/HTML.php | 38 +++++++++++++++++++++-------- tests/src/Content/Text/HTMLTest.php | 26 ++++++++++++++++++++ 2 files changed, 54 insertions(+), 10 deletions(-) diff --git a/src/Content/Text/HTML.php b/src/Content/Text/HTML.php index ddc587d72a..eee443a8f3 100644 --- a/src/Content/Text/HTML.php +++ b/src/Content/Text/HTML.php @@ -42,14 +42,32 @@ class HTML return $cleaned; } - private static function tagToBBCode(DOMDocument $doc, $tag, $attributes, $startbb, $endbb) + /** + * Search all instances of a specific HTML tag node in the provided DOM document and replaces them with BBCode text nodes. + * + * @see HTML::tagToBBCodeSub() + */ + private static function tagToBBCode(DOMDocument $doc, string $tag, array $attributes, string $startbb, string $endbb, bool $ignoreChildren = false) { do { - $done = self::tagToBBCodeSub($doc, $tag, $attributes, $startbb, $endbb); + $done = self::tagToBBCodeSub($doc, $tag, $attributes, $startbb, $endbb, $ignoreChildren); } while ($done); } - private static function tagToBBCodeSub(DOMDocument $doc, $tag, $attributes, $startbb, $endbb) + /** + * Search the first specific HTML tag node in the provided DOM document and replaces it with BBCode text nodes. + * + * @param DOMDocument $doc + * @param string $tag HTML tag name + * @param array $attributes Array of attributes to match and optionally use the value from + * @param string $startbb BBCode tag opening + * @param string $endbb BBCode tag closing + * @param bool $ignoreChildren If set to false, the HTML tag children will be appended as text inside the BBCode tag + * Otherwise, they will be entirely ignored. Useful for simple BBCode that draw their + * inner value from an attribute value and disregard the tag children. + * @return bool Whether a replacement was done + */ + private static function tagToBBCodeSub(DOMDocument $doc, string $tag, array $attributes, string $startbb, string $endbb, bool $ignoreChildren = false) { $savestart = str_replace('$', '\x01', $startbb); $replace = false; @@ -98,7 +116,7 @@ class HTML $node->parentNode->insertBefore($StartCode, $node); - if ($node->hasChildNodes()) { + if (!$ignoreChildren && $node->hasChildNodes()) { /** @var \DOMNode $child */ foreach ($node->childNodes as $key => $child) { /* Remove empty text nodes at the start or at the end of the children list */ @@ -296,14 +314,14 @@ class HTML self::tagToBBCode($doc, 'a', ['href' => '/mailto:(.+)/'], '[mail=$1]', '[/mail]'); self::tagToBBCode($doc, 'a', ['href' => '/(.+)/'], '[url=$1]', '[/url]'); - self::tagToBBCode($doc, 'img', ['src' => '/(.+)/', 'alt' => '/(.+)/'], '[img=$1]$2', '[/img]'); - self::tagToBBCode($doc, 'img', ['src' => '/(.+)/', 'width' => '/(\d+)/', 'height' => '/(\d+)/'], '[img=$2x$3]$1', '[/img]'); - self::tagToBBCode($doc, 'img', ['src' => '/(.+)/'], '[img]$1', '[/img]'); + self::tagToBBCode($doc, 'img', ['src' => '/(.+)/', 'alt' => '/(.+)/'], '[img=$1]$2', '[/img]', true); + self::tagToBBCode($doc, 'img', ['src' => '/(.+)/', 'width' => '/(\d+)/', 'height' => '/(\d+)/'], '[img=$2x$3]$1', '[/img]', true); + self::tagToBBCode($doc, 'img', ['src' => '/(.+)/'], '[img]$1', '[/img]', true); - self::tagToBBCode($doc, 'video', ['src' => '/(.+)/'], '[video]$1', '[/video]'); - self::tagToBBCode($doc, 'audio', ['src' => '/(.+)/'], '[audio]$1', '[/audio]'); - self::tagToBBCode($doc, 'iframe', ['src' => '/(.+)/'], '[iframe]$1', '[/iframe]'); + self::tagToBBCode($doc, 'video', ['src' => '/(.+)/'], '[video]$1', '[/video]', true); + self::tagToBBCode($doc, 'audio', ['src' => '/(.+)/'], '[audio]$1', '[/audio]', true); + self::tagToBBCode($doc, 'iframe', ['src' => '/(.+)/'], '[iframe]$1', '[/iframe]', true); self::tagToBBCode($doc, 'key', [], '[code]', '[/code]'); self::tagToBBCode($doc, 'code', [], '[code]', '[/code]'); diff --git a/tests/src/Content/Text/HTMLTest.php b/tests/src/Content/Text/HTMLTest.php index 65ae052498..1d550049a9 100644 --- a/tests/src/Content/Text/HTMLTest.php +++ b/tests/src/Content/Text/HTMLTest.php @@ -50,4 +50,30 @@ class HTMLTest extends MockedTest $this->assertEquals($expected, $output); } + + public function dataHTMLText() + { + return [ + 'bug-7665-audio-tag' => [ + 'expectedBBCode' => '[audio]http://www.cendrones.fr/colloque2017/jonathanbocquet.mp3[/audio]', + 'html' => '', + ], + ]; + } + + /** + * Test convert bbcodes to HTML + * + * @dataProvider dataHTMLText + * + * @param string $expectedBBCode Expected BBCode output + * @param string $html HTML text + * @throws \Friendica\Network\HTTPException\InternalServerErrorException + */ + public function testToBBCode($expectedBBCode, $html) + { + $actual = HTML::toBBCode($html); + + $this->assertEquals($expectedBBCode, $actual); + } }