public function match(FilterableUri $uri)
{
if (null !== $uri->getQuery()) {
$uri->setFiltered(true, 'URI with query string');
return true;
}
return false;
}
public function match(FilterableUri $uri)
{
if (null !== $uri->getFragment()) {
$uri->setFiltered(true, 'URI with hash fragment');
return true;
}
return false;
}
/**
* @return bool
*/
public function match(FilterableUri $uri)
{
$scheme = $uri->getScheme();
if (!in_array($scheme, $this->allowedSchemes)) {
$uri->setFiltered(true, 'Scheme not allowed');
return true;
}
return false;
}
public function match(FilterableUri $uri)
{
/*
* if the URI does not contain the seed, it is not allowed
*/
if (false === stripos($uri->toString(), $this->seed->toString())) {
$uri->setFiltered(true, 'Doesn\'t match base URI');
return true;
}
return false;
}
public function match(FilterableUri $uri)
{
$currentHostname = $uri->getHost();
if ($this->allowSubDomains) {
// only use hostname.tld for comparison
$currentHostname = join('.', array_slice(explode('.', $currentHostname), -2));
}
if (!in_array($currentHostname, $this->allowedHosts)) {
$uri->setFiltered(true, 'Hostname not allowed');
return true;
}
return false;
}
/**
* Function that crawls each provided URI
* It applies all processors and listeners set on the Spider
*
* This is a either depth first algorithm as explained here:
* https://en.wikipedia.org/wiki/Depth-first_search#Example
* Note that because we don't do it recursive, but iteratively,
* results will be in a different order from the example, because
* we always take the right-most child first, whereas a recursive
* variant would always take the left-most child first
*
* or
*
* a breadth first algorithm
*
* @return void
*/
private function doCrawl()
{
while (count($this->traversalQueue)) {
/** @var $currentUri Uri */
$currentUri = $this->getNextUriFromQueue();
// Fetch the document
if (!($resource = $this->fetchResource($currentUri))) {
continue;
}
$this->dispatch(SpiderEvents::SPIDER_CRAWL_FILTER_POSTFETCH, new GenericEvent($this, array('document' => $resource)));
if ($this->matchesPostfetchFilter($resource)) {
$this->getStatsHandler()->addToFiltered($resource);
continue;
}
// The document was not filtered, so we add it to the processing queue
$this->dispatch(SpiderEvents::SPIDER_CRAWL_PRE_ENQUEUE, new GenericEvent($this, array('document' => $resource)));
$this->addToProcessQueue($resource);
$nextLevel = $this->alreadySeenUris[$currentUri->toString()] + 1;
if ($nextLevel > $this->maxDepth) {
continue;
}
// Once the document is enqueued, apply the discoverers to look for more links to follow
$discoveredUris = $this->executeDiscoverers($resource);
foreach ($discoveredUris as $uri) {
// normalize the URI
$uri->normalize();
// Decorate the link to make it filterable
$uri = new FilterableUri($uri);
// Always skip nodes we already visited
if (array_key_exists($uri->toString(), $this->alreadySeenUris)) {
continue;
}
$this->dispatch(SpiderEvents::SPIDER_CRAWL_FILTER_PREFETCH, new GenericEvent($this, array('uri' => $uri)));
if ($this->matchesPrefetchFilter($uri)) {
$this->getStatsHandler()->addToFiltered($uri);
} else {
// The URI was not matched by any filter, mark as visited and add to queue
array_push($this->traversalQueue, $uri);
}
$this->alreadySeenUris[$uri->toString()] = $nextLevel;
}
}
}
