/**
* This section uses tidy to parse the file
* Specify configuration
*
* @param string $text string to parse.
*/
public function parseString($text)
{
$tidy = new tidy();
$escaped_text = preg_replace(array('/<script([^>]*)>(.*?)<\\/script>/s', '/<link([^>]*)>(.*?)<\\/link>/s', '/<link([^>]*)\\/>/s', '/<style([^>]*)>(.*?)<\\/style>/s'), array('<rs:script$1><![CDATA[$2]]></rs:script>', '<rs:link$1><![CDATA[$2]]></rs:link>', '<rs:link$1/>', '<rs:style$1><![CDATA[$2]]></rs:style>'), $text);
$escaped_text = preg_replace_callback('/<fb:share-button[^>]*>.*<\\/fb:share-button>/s', array('RingsideSocialDslParser', '_replace_meta_and_links'), $escaped_text);
$textToParse = "<rs:social-dsl>{$escaped_text}</rs:social-dsl>";
$tagRegistry = Social_Dsl_TagRegistry::getInstance();
$tagRegistry->scanForNewTags($textToParse);
$tidy->parseString($textToParse, $tagRegistry->getTidyConfiguration());
// error_log("Before tidy parse:");
// error_log($escaped_text);
// error_log("After tidy parse:");
// error_log($tidy);
ob_start();
try {
$this->walk_nodes($tidy->root());
} catch (Exception $e) {
ob_end_clean();
error_log($e->getMessage());
error_log($e->getTraceAsString());
}
$pre_text = ob_get_clean();
if (empty($pre_text)) {
return $pre_text;
}
// We can allow directly <rs:script tags, and replace them with standard script tags
// TODO review this and make this a handler?
$final_text = preg_replace(array('/<rs:script([^>]*)>(.*?)<!\\[CDATA\\[(.*?)]]>(.*?)<\\/rs:script>/s', '/<rs:link([^>]*)>(.*?)<!\\[CDATA\\[(.*?)]]>(.*?)<\\/rs:link>/s', '/<rs:link([^>]*)><\\/rs:link>/s', '/<rs:link([^>]*)\\/>/s', '/<rs:style([^>]*)>(.*?)<!\\[CDATA\\[(.*?)]]>(.*?)<\\/rs:style>/s'), array('<script$1>$2$3$4</script>', '<link$1>$2$3$4</link>', '<link$1/>', '<link$1/>', '<style$1>$2$3$4</style>'), $pre_text);
$result = '';
if (strstr(strtolower($text), '<html>')) {
// If the input has the HTML wrapper, emit the whole result (minus the rs-div div tag)
$result = preg_replace('/<rs:social-dsl>(.*)<\\/rs:social-dsl>/s', '$1', $final_text);
} else {
$matches = array();
preg_match('/<rs:social-dsl>(.*)<\\/rs:social-dsl>/s', $final_text, $matches);
if (!empty($matches) && count($matches) > 1) {
$result = $matches[1];
}
}
// Replace truly-empty rs:social-dsl node (tidy does this)
$result = preg_replace('/<rs:social-dsl *\\/>/', '', $result);
return $result;
}
/**
* Echo the body segments
*
* @return void
*/
public function outputBody()
{
$body = implode('', $this->_body);
if (!$this->_enableTidy) {
echo $body;
} else {
$tidy = new tidy();
$tidy->ParseString($body, $this->_config, $this->_tidyEncoding);
#$tidy->cleanRepair();
if ($this->_echoTidyErrors && $tidy->errorBuffer) {
echo $body;
echo "\n\n\n<!--\n";
echo "tidy detected the following errors:\n";
echo $tidy->errorBuffer;
echo "\n-->";
} else {
echo $tidy->root();
}
}
}
/**
* Get data from path
*
* @param string $path
*
* @return string
*/
public function get($path)
{
/* @var $response \Guzzle\Http\Message\Response */
$response = $this->client->get($path)->send();
if ($response->isError()) {
throw new \RuntimeException('Failed to query the server ' . $this->host);
}
if ($response->getStatusCode() != 200 || !($html = $response->getBody(true))) {
return '';
}
$html = iconv('windows-1251', 'utf-8', $html);
// clean content
$config = array('output-xhtml' => true, 'indent' => true, 'indent-spaces' => 0, 'fix-backslash' => true, 'hide-comments' => true, 'drop-empty-paras' => true, 'wrap' => false);
$this->tidy->parseString($html, $config, 'utf8');
$this->tidy->cleanRepair();
$html = $this->tidy->root()->value;
// ignore blocks
$html = preg_replace('/<noembed>.*?<\\/noembed>/is', '', $html);
$html = preg_replace('/<noindex>.*?<\\/noindex>/is', '', $html);
return $html;
}
/**
* (non-PHPdoc)
* @see PHPCPD_Detector_Tokenizer_AbstractTokenizer::cpd()
*/
public function cpd(PHPCPD_Detector_Strategy $strategy, $file)
{
$buffer = file_get_contents($file);
$this->_iLines = substr_count($buffer, PHP_EOL);
$this->_aTokens = array();
$currentTokenPositions = array();
$currentSignature = '';
$tidy_config = array('clean' => true, 'drop-proprietary-attributes' => true, 'output-xhtml' => true, 'hide-comments' => true, 'show-body-only' => true, 'word-2000' => true, 'wrap' => '0');
$oTidy = new tidy($file, $tidy_config);
$oNode = $oTidy->root();
$this->_tokenHelper($oNode);
unset($oTidy, $oNode, $buffer);
$tokenNr = 0;
foreach ($this->_aTokens as $token) {
if (!isset($this->tokensIgnoreList[$token[0]])) {
$currentTokenPositions[$tokenNr++] = $token[2];
$currentSignature .= chr($token[0] & 255) . pack('N*', crc32($token[1]));
}
}
$strategy->tokenFactor($this->_fTokenFactor, $this->_iMinLines)->processFile($file, $currentTokenPositions, $currentSignature);
return $this;
}
