/**
* @dataProvider firefoxtests
*/
public function test_from_file($data)
{
$locator = new SimplePie_Locator($data, 0, null, 'MockSimplePie_File', false);
$expected = SimplePie_Misc::get_element('link', $data->body);
$feed = $locator->find(SIMPLEPIE_LOCATOR_ALL, $all);
$this->assertFalse($locator->is_feed($data), 'HTML document not be a feed itself');
$this->assertInstanceOf('MockSimplePie_File', $feed);
$expected = array_map(array(get_class(), 'map_url_attrib'), $expected);
$success = array_filter($expected, array(get_class(), 'filter_success'));
$found = array_map(array(get_class(), 'map_url_file'), $all);
$this->assertEquals($success, $found);
}
function replace_urls($data, $tag, $attributes)
{
//This seems to do nothing at all!?
if (!is_array($this->strip_htmltags) || !in_array($tag, $this->strip_htmltags)) {
$elements = SimplePie_Misc::get_element($tag, $data);
foreach ($elements as $element) {
if (is_array($attributes)) {
foreach ($attributes as $attribute) {
if (isset($element['attribs'][$attribute]['data'])) {
$element['attribs'][$attribute]['data'] = SimplePie_Misc::absolutize_url($element['attribs'][$attribute]['data'], $this->base);
$new_element = SimplePie_Misc::element_implode($element);
$data = str_replace($element['full'], $new_element, $data);
$element['full'] = $new_element;
}
}
} elseif (isset($element['attribs'][$attributes]['data'])) {
$element['attribs'][$attributes]['data'] = SimplePie_Misc::absolutize_url($element['attribs'][$attributes]['data'], $this->base);
$data = str_replace($element['full'], SimplePie_Misc::element_implode($element), $data);
}
}
}
return $data;
}
function result()
{
if ($this->data['file']->url != 'http://diveintomark.org/tests/client/autodiscovery/') {
parent::result();
}
static $done = array();
$links = SimplePie_Misc::get_element('link', $this->data['file']->body);
foreach ($links as $link) {
if (!empty($link['attribs']['href']['data']) && !empty($link['attribs']['rel']['data'])) {
$rel = array_unique(SimplePie_Misc::space_seperated_tokens(strtolower($link['attribs']['rel']['data'])));
$href = SimplePie_Misc::absolutize_url(trim($link['attribs']['href']['data']), $this->data['file']->url);
if (!in_array($href, $done) && in_array('next', $rel)) {
$done[] = $this->data['url'] = $href;
break;
}
}
}
if ($this->data['url']) {
$this->run();
}
}
public function get_links()
{
$links = SimplePie_Misc::get_element('a', $this->file->body);
foreach ($links as $link) {
if (isset($link['attribs']['href']['data'])) {
$href = trim($link['attribs']['href']['data']);
$parsed = SimplePie_Misc::parse_url($href);
if ($parsed['scheme'] === '' || preg_match('/^(http(s)|feed)?$/i', $parsed['scheme'])) {
if ($this->base_location < $link['offset']) {
$href = SimplePie_Misc::absolutize_url(trim($link['attribs']['href']['data']), $this->base);
} else {
$href = SimplePie_Misc::absolutize_url(trim($link['attribs']['href']['data']), $this->http_base);
}
$current = SimplePie_Misc::parse_url($this->file->url);
if ($parsed['authority'] === '' || $parsed['authority'] === $current['authority']) {
$this->local[] = $href;
} else {
$this->elsewhere[] = $href;
}
}
}
}
$this->local = array_unique($this->local);
$this->elsewhere = array_unique($this->elsewhere);
if (!empty($this->local) || !empty($this->elsewhere)) {
return true;
}
return null;
}
function sanitize($data, $attribs, $is_url = false, $force_decode = false)
{
$this->attribs = $attribs;
if (isset($this->feedinfo['type']) && $this->feedinfo['type'] == 'Atom') {
if (!empty($attribs['MODE']) && $attribs['MODE'] == 'base64' || !empty($attribs['TYPE']) && $attribs['TYPE'] == 'application/octet-stream') {
$data = trim($data);
$data = base64_decode($data);
} else {
if (!empty($attribs['MODE']) && $attribs['MODE'] == 'escaped' || !empty($attribs['TYPE']) && ($attribs['TYPE'] == 'html' || $attribs['TYPE'] == 'text/html')) {
$data = $this->entities_decode($data);
}
}
if (!empty($attribs['TYPE']) && ($attribs['TYPE'] == 'xhtml' || $attribs['TYPE'] == 'application/xhtml+xml')) {
if ($this->remove_div) {
$data = preg_replace('/<div( .*)?>/msiU', '', strrev(preg_replace('/>vid\\/</i', '', strrev($data), 1)), 1);
} else {
$data = preg_replace('/<div( .*)?>/msiU', '<div>', $data, 1);
}
$data = $this->convert_entities($data);
}
} else {
$data = $this->convert_entities($data);
}
if ($force_decode) {
$data = $this->entities_decode($data);
}
$data = trim($data);
$data = preg_replace('/<\\!--([^-]|-[^-])*-->/msiU', '', $data);
// If Strip Ads is enabled, strip them.
if ($this->strip_ads) {
$data = preg_replace('/<a (.*)href=(.*)click\\.phdo\\?s=(.*)<\\/a>/msiU', '', $data);
// Pheedo links (tested with Dooce.com)
$data = preg_replace('/<p(.*)>(.*)<a href="http:\\/\\/ad.doubleclick.net\\/jump\\/(.*)<\\/p>/msiU', '', $data);
// Doubleclick links (tested with InfoWorld.com)
$data = preg_replace('/<p><map (.*)name=(.*)google_ad_map(.*)<\\/p>/msiU', '', $data);
// Google AdSense for Feeds (tested with tuaw.com).
// Feedflare, from Feedburner
}
// Replace H1, H2, and H3 tags with the less important H4 tags.
// This is because on a site, the more important headers might make sense,
// but it most likely doesn't fit in the context of RSS-in-a-webpage.
if ($this->replace_headers) {
$data = preg_replace('/<h[1-3]((\\s*((\\w+:)?\\w+)\\s*=\\s*("([^"]*)"|\'([^\']*)\'|(.*)))*)\\s*>/msiU', '<h4\\1>', $data);
$data = preg_replace('/<\\/h[1-3]>/i', '</h4>', $data);
}
if ($is_url) {
$data = $this->replace_urls($data, true);
} else {
$data = preg_replace_callback('/<(\\S+)((\\s*((\\w+:)?\\w+)\\s*=\\s*("([^"]*)"|\'([^\']*)\'|(.*)))*)\\s*(\\/>|>(.*)<\\/\\S+>)/msiU', array(&$this, 'replace_urls'), $data);
}
// If Bypass Image Hotlink is enabled, rewrite all the image tags.
if ($this->bypass_image_hotlink) {
$images = SimplePie_Misc::get_element('img', $data);
foreach ($images as $img) {
if (!empty($img['attribs']['SRC']['data'])) {
$pre = '';
if ($this->bypass_image_hotlink_page) {
$pre = $this->bypass_image_hotlink_page;
}
$pre .= "?{$this->bypass_image_hotlink}=";
$img['attribs']['SRC']['data'] = $pre . rawurlencode(strtr($img['attribs']['SRC']['data'], array_flip(get_html_translation_table(HTML_SPECIALCHARS, ENT_QUOTES))));
$data = str_replace($img['full'], SimplePie_Misc::element_implode($img), $data);
}
}
}
// Strip out HTML tags and attributes that might cause various security problems.
// Based on recommendations by Mark Pilgrim at:
// http://diveintomark.org/archives/2003/06/12/how_to_consume_rss_safely
if ($this->strip_htmltags) {
foreach ($this->strip_htmltags as $tag) {
$data = preg_replace_callback("/<({$tag})((\\s*((\\w+:)?\\w+)(\\s*=\\s*(\"([^\"]*)\"|'([^']*)'|(.*)))?)*)\\s*(\\/>|>(.*)<\\/({$tag})((\\s*((\\w+:)?\\w+)(\\s*=\\s*(\"([^\"]*)\"|'([^']*)'|(.*)))?)*)\\s*>)/msiU", array(&$this, 'do_strip_htmltags'), $data);
}
}
if ($this->strip_attributes) {
foreach ($this->strip_attributes as $attrib) {
$data = preg_replace('/ ' . trim($attrib) . '=("|")(\\w|\\s|=|-|:|;|\\/|\\.|\\?|&|,|#|!|\\(|\\)|\'|'|<|>|\\+|{|})*("|")/i', '', $data);
$data = preg_replace('/ ' . trim($attrib) . '=(\'|')(\\w|\\s|=|-|:|;|\\/|\\.|\\?|&|,|#|!|\\(|\\)|"|"|<|>|\\+|{|})*(\'|')/i', '', $data);
$data = preg_replace('/ ' . trim($attrib) . '=(\\w|\\s|=|-|:|;|\\/|\\.|\\?|&|,|#|!|\\(|\\)|\\+|{|})*/i', '', $data);
}
}
// Convert encoding
$data = SimplePie_Misc::change_encoding($data, $this->input_encoding, $this->output_encoding);
return $data;
}
function replace_urls($data, $tag, $attribute)
{
$elements = SimplePie_Misc::get_element($tag, $data);
foreach ($elements as $element) {
if (isset($element['attribs'][$attribute]['data'])) {
$element['attribs'][$attribute]['data'] = SimplePie_Misc::absolutize_url($element['attribs'][$attribute]['data'], $this->base);
$data = str_replace($element['full'], SimplePie_Misc::element_implode($element), $data);
}
}
return $data;
}
