/**
* @Then /^after merging them I\'m expecting one result with:$/
*/
public function afterMergingThemIMExpectingOneResultWith(TableNode $table)
{
$result = new ParserResult();
foreach ($this->results as $r) {
$result->mergeWithResult($r);
}
$warningsCount = $examplesCount = 0;
foreach ($table->getHash() as $row) {
$file = $row['test-file'];
$warningsCount += count(explode(', ', $row['warning']));
$examplesCount += count(explode(', ', $row['examples']));
assertEquals(explode(', ', $row['warning']), $result->getWarnings($file));
assertEquals(explode(', ', $row['examples']), $result->getExamples($file));
assertEquals($row['result'], $result->getResult($file));
if ($row['skip']) {
assertTrue($result->isSkipped($file));
} else {
assertFalse($result->isSkipped($file));
}
}
assertEquals($warningsCount, $result->countAllWarnings());
assertEquals($examplesCount, $result->countAllExamples());
}
/**
* Process single file.
*
* @param $file Source file
* @param $parseExamples Whether or not include also examples
* @return ParserResult Parse result including all warnings and skipped files
*/
public function processFile($file, $parseExamples, $result = null)
{
if (!$result) {
$result = new ParserResult();
}
$dom = new \DOMDocument();
@$dom->loadHTML(file_get_contents($file));
// Most important object used to traverse HTML DOM structure.
$xpath = new \DOMXPath($dom);
$function = [];
// Parse function name.
$h1 = $xpath->query('//h1[@class="refname"]');
// Check if it managed to find function name.
if ($h1->length == 0) {
$result->addSkipped(basename($file));
return $result;
}
// Function short description.
$description = $xpath->query('//span[@class="dc-title"]');
if ($description->length > 0) {
// some functions don't have any description
$function['desc'] = trim(Utils::simplifyString($description->item(0)->textContent), '.') . '.';
} else {
$function['desc'] = null;
}
// Function long description.
$longDescParagraphs = $xpath->query('//div[contains(@class,"description")]//p[contains(@class,"para") or contains(@class,"simpara")]');
$function['long_desc'] = Utils::extractFormattedText($longDescParagraphs);
if ($function['long_desc']) {
$function['long_desc'] = trim($function['long_desc'], '.') . '.';
}
// PHP version since this function is available.
$version = $xpath->query('//p[@class="verinfo"]');
if ($version->length > 0) {
// check from which PHP version is it available
$function['ver'] = trim($version->item(0)->textContent, '()');
} else {
$function['ver'] = null;
}
// Return value description.
$items = $xpath->query('//div[contains(@class,"returnvalues")]/p');
if ($items->length > 0) {
$function['ret_desc'] = Utils::simplifyString($items->item(0)->textContent);
}
// "See also" methods
$seeAlso = $xpath->query('//div[contains(@class,"seealso")]');
if ($seeAlso->length > 0) {
$function['seealso'] = [];
//$seeAlsoArray = array();
$lis = $xpath->query('.//li', $seeAlso->item(0));
foreach ($lis as $li) {
// store just the name and description without parenthesis
$text = explode('-', $li->textContent);
$name = rtrim(trim($text[0]), '()');
if (strpos($name, ' ') === false) {
$function['seealso'][] = $name;
}
}
}
// filename (url)
$function['filename'] = substr(basename($file), 0, -5);
// All function parameters in an array of arrays because some function have mutliple definitions.
// Note: One function can have multiple parameter count (see http://www.php.net/manual/en/function.strtr.php)
$function['params'] = [];
$funcDescription = $xpath->query('//div[@class="refsect1 description"]/div[@class="methodsynopsis dc-description"]');
foreach ($funcDescription as $index => $description) {
// Function name for this parameter list.
$altName = str_replace('->', '::', $xpath->query('./span[@class="methodname"]', $description)->item(0)->textContent);
$parsedParams = ['list' => [], 'name' => $altName];
// Return value type (boolean, integer, mixed, ... ).
$span = $xpath->query('./span[@class="type"]', $description);
if ($span->length > 0 && !isset($function['return']['type'])) {
$parsedParams['ret_type'] = $span->item(0)->textContent;
}
// Parameter containers.
$params = $xpath->query('span[@class="methodparam"]', $description);
// skip empty parameter list (function declaration that doesn't take any parameter)
if ($params->item(0)->textContent != 'void') {
$optional = substr_count($description->textContent, '[');
$allParameters = $xpath->query('//div[contains(@class,"parameters")]/dl/dt/code[@class="parameter"]');
for ($i = 0; $i < $params->length; $i++) {
$paramNodes = $xpath->query('*', $params->item($i));
$descPattern = './*[self::p or self::ul or self::blockquote or self::table or self::div[@class="methodsynopsis dc-description"]]';
$paramDescriptions = null;
$varName = $paramNodes->item(1)->textContent;
foreach ($allParameters as $index => $paramDesc) {
if (ltrim($varName, '$&') == $paramDesc->textContent) {
$paramDescriptions = $xpath->query('//div[contains(@class,"parameters")]/dl/dd[' . ($index + 1) . ']');
break;
}
}
// Single parameter.
$param = array('type' => $paramNodes->item(0) ? $paramNodes->item(0)->textContent : 'unknown', 'var' => $varName, 'beh' => $params->length - $optional > $i ? 'required' : 'optional', 'desc' => $paramDescriptions ? Utils::extractFormattedText($xpath->query($descPattern, $paramDescriptions->item(0)), $xpath) : null);
// Default value for this parameter
if ($paramNodes->length >= 3) {
$param['default'] = trim($paramNodes->item(2)->textContent, ' =');
}
$parsedParams['list'][] = $param;
}
}
$function['params'][] = $parsedParams;
}
// If parser didn't find any parameters, no short or long description then it's probably not a function in this file
if (!$function['params'] || !isset($function['desc']) && !isset($function['long_desc'])) {
$result->addSkipped(basename($file));
return $result;
}
$funcName = strtolower($function['params'][0]['name']);
foreach ($function['params'] as $index => $param) {
// Use all alternative names just as a reference to the first (primary) name.
$name = strtolower($function['params'][$index]['name']);
$result->setResult($name, $index == 0 ? $function : $funcName);
}
// Parse all examples in this file.
if ($parseExamples != self::SKIP_EXAMPLES) {
// Find all source code containers.
$exampleDiv = $xpath->query('//div[@class="example" or @class="informalexample"]');
for ($i = 0; $i < $exampleDiv->length; $i++) {
$output = null;
// Get source code as pure text, without any syntax highlighting tags.
$sourceCode = $dom->saveXML($xpath->query('.//div[@class="phpcode"]', $exampleDiv->item($i))->item(0));
$outputDiv = $xpath->query('.//div[@class="cdata"]', $exampleDiv->item($i));
if ($outputDiv->length > 0) {
$output = $xpath->query('.//div[@class="cdata"]', $exampleDiv->item($i))->item(0)->textContent;
}
// Example title, strip beginning and ending php tags.
$ps = $xpath->query('p', $exampleDiv->item($i));
if ($ps->length > 0) {
$title = $ps->item(0)->textContent;
// Remove some unnecessary stuff.
$title = trim(preg_replace('/^(Example #\\d+|Beispiel #\\d+|Exemplo #\\d+|Exemple #\\d+|Przykład #\\d+)/', '', $title));
$title = trim(preg_replace('/\\s+/', ' ', $title));
} else {
$title = null;
}
$example = ['title' => $title, 'source' => Utils::clearSourceCode($sourceCode), 'output' => trim($output) ?: null];
// Skip examples with malformed UTF-8 characters.
// @todo: check where's the problem
json_encode($example, JSON_NUMERIC_CHECK | JSON_UNESCAPED_UNICODE);
if (json_last_error()) {
$result->addWarning($funcName, 'Example "' . $title . '": ' . json_last_error_msg());
} else {
$result->addExample($funcName, $example);
}
}
}
return $result;
}
