diogo
/
symfony
Archived
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

bug #11624 [DomCrawler] fix the axes handling in a bc way (xabbuh) 

This PR was merged into the 2.3 branch.

Discussion
----------

[DomCrawler] fix the axes handling in a bc way

| Q             | A
| ------------- | ---
| Bug fix?      | yes
| New feature?  | no
| BC breaks?    | no
| Deprecations? | no
| Tests pass?   | yes
| Fixed tickets | #11503
| License       | MIT
| Doc PR        |

The previous fix in #11548 for handling XPath axes was not backward compatible. In previous Symfony versions the Crawler handled nodes by holding a "fake root node". This must be taken into account when evaluating (relativizing) XPath expressions.

Commits
-------

d26040f [DomCrawler] fix the axes handling in a bc way
This commit is contained in:
Christophe Coevoet 2014-08-19 12:43:37 +02:00
parent afab3ee4c2 d26040f356
commit 00aedfc150
2 changed files with 63 additions and 21 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

25
src/Symfony/Component/DomCrawler/Crawler.php
View File

@ -245,7 +245,9 @@ class Crawler extends \SplObjectStorage
public function addNodeList(\DOMNodeList $nodes)
{
foreach ($nodes as $node) {
$this->addNode($node);
if ($node instanceof \DOMNode) {
$this->addNode($node);
}
}
}
@ -834,18 +836,22 @@ class Crawler extends \SplObjectStorage
// BC for Symfony 2.4 and lower were elements were adding in a fake _root parent
if (0 === strpos($expression, '/_root/')) {
$expression = './'.substr($expression, 7);
} elseif (0 === strpos($expression, 'self::*/')) {
$expression = './'.substr($expression, 8);
}
// add prefix before absolute element selector
if (empty($expression)) {
$expression = $nonMatchingExpression;
} elseif (0 === strpos($expression, '//')) {
$expression = 'descendant-or-self::' . substr($expression, 2);
$expression = 'descendant-or-self::'.substr($expression, 2);
} elseif (0 === strpos($expression, './/')) {
$expression = 'descendant-or-self::' . substr($expression, 3);
$expression = 'descendant-or-self::'.substr($expression, 3);
} elseif (0 === strpos($expression, './')) {
$expression = 'self::' . substr($expression, 2);
} elseif ('/' === $expression[0]) {
$expression = 'self::'.substr($expression, 2);
} elseif (0 === strpos($expression, 'child::')) {
$expression = 'self::'.substr($expression, 7);
} elseif ('/' === $expression[0] || 0 === strpos($expression, 'self::')) {
// the only direct child in Symfony 2.4 and lower is _root, which is already handled previously
// so let's drop the expression entirely
$expression = $nonMatchingExpression;
@ -853,9 +859,12 @@ class Crawler extends \SplObjectStorage
// '.' is the fake root element in Symfony 2.4 and lower, which is excluded from results
$expression = $nonMatchingExpression;
} elseif (0 === strpos($expression, 'descendant::')) {
$expression = 'descendant-or-self::' . substr($expression, strlen('descendant::'));
} elseif (!preg_match('/^(ancestor|ancestor-or-self|attribute|child|descendant-or-self|following|following-sibling|parent|preceding|preceding-sibling|self)::/', $expression)) {
$expression = 'self::' .$expression;
$expression = 'descendant-or-self::'.substr($expression, strlen('descendant::'));
} elseif (preg_match('/^(ancestor|ancestor-or-self|attribute|following|following-sibling|namespace|parent|preceding|preceding-sibling)::/', $expression)) {
// the fake root has no parent, preceding or following nodes and also no attributes (even no namespace attributes)
$expression = $nonMatchingExpression;
} elseif (0 !== strpos($expression, 'descendant-or-self::')) {
$expression = 'self::'.$expression;
}
$expressions[] = $parenthesis.$expression;
}

59
src/Symfony/Component/DomCrawler/Tests/CrawlerTest.php
View File

@ -388,8 +388,8 @@ EOF
$this->assertCount(1, $crawler->filterXPath('//body'));
$this->assertCount(1, $crawler->filterXPath('descendant-or-self::body'));
$this->assertCount(1, $crawler->filterXPath('//div[@id="parent"]')->filterXPath('./div'), 'A child selection finds only the current div');
$this->assertCount(2, $crawler->filterXPath('//div[@id="parent"]')->filterXPath('descendant::div'), 'A descendant selector matches the current div and its child');
$this->assertCount(2, $crawler->filterXPath('//div[@id="parent"]')->filterXPath('//div'), 'A descendant selector matches the current div and its child');
$this->assertCount(3, $crawler->filterXPath('//div[@id="parent"]')->filterXPath('descendant::div'), 'A descendant selector matches the current div and its child');
$this->assertCount(3, $crawler->filterXPath('//div[@id="parent"]')->filterXPath('//div'), 'A descendant selector matches the current div and its child');
$this->assertCount(5, $crawler->filterXPath('(//a | //div)//img'));
$this->assertCount(7, $crawler->filterXPath('((//a | //div)//img | //ul)'));
$this->assertCount(7, $crawler->filterXPath('( ( //a | //div )//img | //ul )'));
@ -411,72 +411,104 @@ EOF
$this->assertCount(3, $crawler->filterXPath('//body')->filterXPath('//button')->parents(), '->filterXpath() preserves parents when chained');
}
public function testFilterXPathWithFakeRoot()
{
$crawler = $this->createTestCrawler();
$this->assertCount(0, $crawler->filterXPath('.'), '->filterXPath() returns an empty result if the XPath references the fake root node');
$this->assertCount(0, $crawler->filterXPath('/_root'), '->filterXPath() returns an empty result if the XPath references the fake root node');
$this->assertCount(0, $crawler->filterXPath('self::*'), '->filterXPath() returns an empty result if the XPath references the fake root node');
$this->assertCount(0, $crawler->filterXPath('self::_root'), '->filterXPath() returns an empty result if the XPath references the fake root node');
}
public function testFilterXPathWithAncestorAxis()
{
$crawler = $this->createTestCrawler()->filterXPath('//form');
$this->assertCount(2, $crawler->filterXPath('ancestor::*'));
$this->assertCount(0, $crawler->filterXPath('ancestor::*'), 'The fake root node has no ancestor nodes');
}
public function testFilterXPathWithAncestorOrSelfAxis()
{
$crawler = $this->createTestCrawler()->filterXPath('//form');
$this->assertCount(3, $crawler->filterXPath('ancestor-or-self::*'));
$this->assertCount(0, $crawler->filterXPath('ancestor-or-self::*'), 'The fake root node has no ancestor nodes');
}
public function testFilterXPathWithAttributeAxis()
{
$crawler = $this->createTestCrawler()->filterXPath('//form');
$this->assertCount(2, $crawler->filterXPath('attribute::*'));
$this->assertCount(0, $crawler->filterXPath('attribute::*'), 'The fake root node has no attribute nodes');
}
public function testFilterXPathWithAttributeAxisAfterElementAxis()
{
$this->assertCount(3, $this->createTestCrawler()->filterXPath('//form/button/attribute::*'), '->filterXPath() handles attribute axes properly when they are preceded by an element filtering axis');
}
public function testFilterXPathWithChildAxis()
{
$crawler = $this->createTestCrawler()->filterXPath('//body');
$crawler = $this->createTestCrawler()->filterXPath('//div[@id="parent"]');
$this->assertCount(2, $crawler->filterXPath('child::input'));
$this->assertCount(1, $crawler->filterXPath('child::div'), 'A child selection finds only the current div');
}
public function testFilterXPathWithFollowingAxis()
{
$crawler = $this->createTestCrawler()->filterXPath('//a');
$this->assertCount(3, $crawler->filterXPath('following::div'));
$this->assertCount(0, $crawler->filterXPath('following::div'), 'The fake root node has no following nodes');
}
public function testFilterXPathWithFollowingSiblingAxis()
{
$crawler = $this->createTestCrawler()->filterXPath('//a');
$this->assertCount(2, $crawler->filterXPath('following-sibling::div'));
$this->assertCount(0, $crawler->filterXPath('following-sibling::div'), 'The fake root node has no following nodes');
}
public function testFilterXPathWithNamespaceAxis()
{
$crawler = $this->createTestCrawler()->filterXPath('//button');
$this->assertCount(0, $crawler->filterXPath('namespace::*'), 'The fake root node has no namespace nodes');
}
public function testFilterXPathWithNamespaceAxisAfterElementAxis()
{
$crawler = $this->createTestCrawler()->filterXPath('//div[@id="parent"]/namespace::*');
$this->assertCount(0, $crawler->filterXPath('namespace::*'), 'Namespace axes cannot be requested');
}
public function testFilterXPathWithParentAxis()
{
$crawler = $this->createTestCrawler()->filterXPath('//button');
$this->assertEquals('foo', $crawler->filterXPath('parent::*')->attr('action'));
$this->assertCount(0, $crawler->filterXPath('parent::*'), 'The fake root node has no parent nodes');
}
public function testFilterXPathWithPrecedingAxis()
{
$crawler = $this->createTestCrawler()->filterXPath('//form');
$this->assertCount(13, $crawler->filterXPath('preceding::*'));
$this->assertCount(0, $crawler->filterXPath('preceding::*'), 'The fake root node has no preceding nodes');
}
public function testFilterXPathWithPrecedingSiblingAxis()
{
$crawler = $this->createTestCrawler()->filterXPath('//form');
$this->assertCount(9, $crawler->filterXPath('preceding-sibling::*'));
$this->assertCount(0, $crawler->filterXPath('preceding-sibling::*'), 'The fake root node has no preceding nodes');
}
public function testFilterXPathWithSelfAxes()
{
$this->assertCount(1, $this->createTestCrawler()->filterXPath('self::*'));
$crawler = $this->createTestCrawler()->filterXPath('//a');
$this->assertCount(0, $crawler->filterXPath('self::a'), 'The fake root node has no "real" element name');
$this->assertCount(0, $crawler->filterXPath('self::a/img'), 'The fake root node has no "real" element name');
$this->assertCount(9, $crawler->filterXPath('self::*/a'));
}
/**
@ -844,6 +876,7 @@ HTML;
</ul>
<div id="parent">
<div id="child"></div>
<div id="child2" xmlns:foo="http://example.com"></div>
</div>
<div id="sibling"><img /></div>
</body>