[DomCrawler] Add support for XPath expression evaluation

This commit is contained in:
Jakub Zalas 2016-07-26 09:49:12 +01:00
parent 695549fdde
commit 3148fad236
2 changed files with 75 additions and 0 deletions

View File

@ -592,6 +592,36 @@ class Crawler implements \Countable, \IteratorAggregate
return $html;
}
/**
* Evaluates an XPath expression.
*
* Since an XPath expression might evaluate to either a simple type or a \DOMDoneList,
* this method will return either an array of simple types or a new Crawler instance.
*
* @param string $xpath An XPath expression
*
* @return array|Crawler An array of evaluation results or a new Crawler instance
*/
public function evaluate($xpath)
{
if (null === $this->document) {
throw new \LogicException('Cannot evaluate the expression on an uninitialized crawler.');
}
$data = array();
$domxpath = $this->createDOMXPath($this->document, $this->findNamespacePrefixes($xpath));
foreach ($this->nodes as $node) {
$data[] = $domxpath->evaluate($xpath, $node);
}
if (isset($data[0]) && $data[0] instanceof \DOMNodeList) {
return $this->createSubCrawler($data);
}
return $data;
}
/**
* Extracts information from the list of nodes.
*

View File

@ -1061,6 +1061,51 @@ HTML;
$this->assertCount(1, $crawler->filter('li:contains("List item 1")'));
}
public function testEvaluateReturnsTypedResultOfXPathExpressionOnADocumentSubset()
{
$crawler = $this->createTestCrawler();
$result = $crawler->filterXPath('//form/input')->evaluate('substring-before(@name, "Name")');
$this->assertSame(array('Text', 'Foo', 'Bar'), $result);
}
public function testEvaluateReturnsTypedResultOfNamespacedXPathExpressionOnADocumentSubset()
{
$crawler = $this->createTestXmlCrawler();
$result = $crawler->filterXPath('//yt:accessControl/@action')->evaluate('string(.)');
$this->assertSame(array('comment', 'videoRespond'), $result);
}
public function testEvaluateReturnsTypedResultOfNamespacedXPathExpression()
{
$crawler = $this->createTestXmlCrawler();
$crawler->registerNamespace('youtube', 'http://gdata.youtube.com/schemas/2007');
$result = $crawler->evaluate('string(//youtube:accessControl/@action)');
$this->assertSame(array('comment'), $result);
}
public function testEvaluateReturnsACrawlerIfXPathExpressionEvaluatesToANode()
{
$crawler = $this->createTestCrawler()->evaluate('//form/input[1]');
$this->assertInstanceOf(Crawler::class, $crawler);
$this->assertCount(1, $crawler);
$this->assertSame('input', $crawler->first()->nodeName());
}
/**
* @expectedException \LogicException
*/
public function testEvaluateThrowsAnExceptionIfDocumentIsEmpty()
{
(new Crawler())->evaluate('//form/input[1]');
}
public function createTestCrawler($uri = null)
{
$dom = new \DOMDocument();