mirror of
https://github.com/RSS-Bridge/rss-bridge.git
synced 2024-11-28 22:28:51 +03:00
Expanded Sexactu to cover the whole GQ magazine (#861)
The bridge has been expanded to better cover the whole GQ magazine. It should support all countries (provided they all use the same absurdly shitty publication system). It is guaranteed to be only tested with sexactu articles (that I now obtain by loading Maïa Mazaurette author page).
This commit is contained in:
Nicolas Delsaux
LogMANOriginal
parent
5779f641c0
commit
ee158468fa
2 changed files with 119 additions and 88 deletions
119
bridges/GQMagazineBridge.php
Normal file
119
bridges/GQMagazineBridge.php
Normal file
|
|
@ -0,0 +1,119 @@
|
|||
<?php
|
||||
|
||||
/**
|
||||
* An extension of the previous SexactuBridge to cover the whole GQMagazine.
|
||||
* This one taks a page (as an example sexe/news or journaliste/maia-mazaurette) which is to be configured,
|
||||
* reads all the articles visible on that page, and make a stream out of it.
|
||||
* @author nicolas-delsaux
|
||||
*
|
||||
*/
|
||||
class GQMagazineBridge extends BridgeAbstract
|
||||
{
|
||||
|
||||
const MAINTAINER = 'Riduidel';
|
||||
|
||||
const NAME = 'GQMagazine';
|
||||
|
||||
// URI is no more valid, since we can address the whole gq galaxy
|
||||
const URI = 'https://www.gqmagazine.fr';
|
||||
|
||||
const CACHE_TIMEOUT = 7200; // 2h
|
||||
const DESCRIPTION = 'GQMagazine section extractor bridge. This bridge allows you get only a specific section.';
|
||||
|
||||
const PARAMETERS = array( array(
|
||||
'domain' => array(
|
||||
'name' => 'Domain to use',
|
||||
'required' => true,
|
||||
'values' => array(
|
||||
'www.gqmagazine.fr' => 'www.gqmagazine.fr'
|
||||
),
|
||||
'defaultValue' => 'www.gqmagazine.fr'
|
||||
),
|
||||
'page' => array(
|
||||
'name' => 'Initial page to load',
|
||||
'required' => true
|
||||
),
|
||||
));
|
||||
|
||||
const REPLACED_ATTRIBUTES = array(
|
||||
'href' => 'href',
|
||||
'src' => 'src',
|
||||
'data-original' => 'src'
|
||||
);
|
||||
|
||||
private function getDomain() {
|
||||
return $this->getInput('domain');
|
||||
}
|
||||
|
||||
public function getURI()
|
||||
{
|
||||
return $this->getDomain() . '/' . $this->getInput('page');
|
||||
}
|
||||
|
||||
public function collectData()
|
||||
{
|
||||
$html = getSimpleHTMLDOM($this->getURI()) or returnServerError('Could not request ' . $this->getURI());
|
||||
|
||||
// Since GQ don't want simple class scrapping, let's do it the hard way and ... discover content !
|
||||
$main = $html->find('main', 0);
|
||||
foreach ($main->find('a') as $link) {
|
||||
$uri = $link->href;
|
||||
$title = $link->find('h2', 0);
|
||||
$date = $link->find('time', 0);
|
||||
|
||||
$item = array();
|
||||
$author = $link->find('span[itemprop=name]', 0);
|
||||
$item['author'] = $author->plaintext;
|
||||
$item['title'] = $title->plaintext;
|
||||
if(substr($uri, 0, 1) === 'h') { // absolute uri
|
||||
$item['uri'] = $uri;
|
||||
} else if(substr($uri, 0, 1) === '/') { // domain relative url
|
||||
$item['uri'] = $this->getDomain() . $uri;
|
||||
} else {
|
||||
$item['uri'] = $this->getDomain() . '/' . $uri;
|
||||
}
|
||||
|
||||
$article = $this->loadFullArticle($item['uri']);
|
||||
if($article) {
|
||||
$item['content'] = $this->replaceUriInHtmlElement($article);
|
||||
} else {
|
||||
$item['content'] = "<strong>Article body couldn't be loaded</strong>. It must be a bug!";
|
||||
}
|
||||
$short_date = $date->datetime;
|
||||
$item['timestamp'] = strtotime($short_date);
|
||||
$this->items[] = $item;
|
||||
}
|
||||
}
|
||||
|
||||
/**
|
||||
* Loads the full article and returns the contents
|
||||
* @param $uri The article URI
|
||||
* @return The article content
|
||||
*/
|
||||
private function loadFullArticle($uri){
|
||||
$html = getSimpleHTMLDOMCached($uri);
|
||||
// Once again, that generated css classes madness is an obstacle ... which i can go over easily
|
||||
foreach($html->find('div') as $div) {
|
||||
// List the CSS classes of that div
|
||||
$classes = $div->class;
|
||||
// I can't directly lookup that class since GQ since to generate random names like "ArticleBodySection-fkggUW"
|
||||
if(strpos($classes, 'ArticleBodySection') !== false) {
|
||||
return $div;
|
||||
}
|
||||
}
|
||||
return null;
|
||||
}
|
||||
|
||||
/**
|
||||
* Replaces all relative URIs with absolute ones
|
||||
* @param $element A simplehtmldom element
|
||||
* @return The $element->innertext with all URIs replaced
|
||||
*/
|
||||
private function replaceUriInHtmlElement($element){
|
||||
$returned = $element->innertext;
|
||||
foreach (self::REPLACED_ATTRIBUTES as $initial => $final) {
|
||||
$returned = str_replace($initial . '="/', $final . '="' . self::URI . '/', $returned);
|
||||
}
|
||||
return $returned;
|
||||
}
|
||||
}
|
||||
|
|
@ -1,88 +0,0 @@
|
|||
<?php
|
||||
class SexactuBridge extends BridgeAbstract {
|
||||
|
||||
const MAINTAINER = 'Riduidel';
|
||||
const NAME = 'Sexactu';
|
||||
const AUTHOR = 'Maïa Mazaurette';
|
||||
const URI = 'http://www.gqmagazine.fr';
|
||||
const CACHE_TIMEOUT = 7200; // 2h
|
||||
const DESCRIPTION = 'Sexactu via rss-bridge';
|
||||
|
||||
const REPLACED_ATTRIBUTES = array(
|
||||
'href' => 'href',
|
||||
'src' => 'src',
|
||||
'data-original' => 'src'
|
||||
);
|
||||
|
||||
public function getURI(){
|
||||
return self::URI . '/sexactu';
|
||||
}
|
||||
|
||||
public function collectData(){
|
||||
$html = getSimpleHTMLDOM($this->getURI())
|
||||
or returnServerError('Could not request ' . $this->getURI());
|
||||
|
||||
$sexactu = $html->find('.container_sexactu', 0);
|
||||
$rowList = $sexactu->find('.row');
|
||||
foreach($rowList as $row) {
|
||||
// only use first list as second one only contains pages numbers
|
||||
|
||||
$title = $row->find('.title', 0);
|
||||
if($title) {
|
||||
$item = array();
|
||||
$item['author'] = self::AUTHOR;
|
||||
$item['title'] = $title->plaintext;
|
||||
$urlAttribute = 'data-href';
|
||||
$uri = $title->$urlAttribute;
|
||||
if($uri === false)
|
||||
continue;
|
||||
if(substr($uri, 0, 1) === 'h') { // absolute uri
|
||||
$item['uri'] = $uri;
|
||||
} else if(substr($uri, 0, 1) === '/') { // domain relative url
|
||||
$item['uri'] = self::URI . $uri;
|
||||
} else {
|
||||
$item['uri'] = $this->getURI() . $uri;
|
||||
}
|
||||
$article = $this->loadFullArticle($item['uri']);
|
||||
$item['content'] = $this->replaceUriInHtmlElement($article->find('.article_content', 0));
|
||||
|
||||
$publicationDate = $article->find('time[itemprop=datePublished]', 0);
|
||||
$short_date = $publicationDate->datetime;
|
||||
$item['timestamp'] = strtotime($short_date);
|
||||
} else {
|
||||
// Sometimes we get rubbish, ignore.
|
||||
continue;
|
||||
}
|
||||
$this->items[] = $item;
|
||||
}
|
||||
}
|
||||
|
||||
/**
|
||||
* Loads the full article and returns the contents
|
||||
* @param $uri The article URI
|
||||
* @return The article content
|
||||
*/
|
||||
private function loadFullArticle($uri){
|
||||
$html = getSimpleHTMLDOMCached($uri);
|
||||
|
||||
$content = $html->find('#article', 0);
|
||||
if($content) {
|
||||
return $content;
|
||||
}
|
||||
|
||||
return null;
|
||||
}
|
||||
|
||||
/**
|
||||
* Replaces all relative URIs with absolute ones
|
||||
* @param $element A simplehtmldom element
|
||||
* @return The $element->innertext with all URIs replaced
|
||||
*/
|
||||
private function replaceUriInHtmlElement($element){
|
||||
$returned = $element->innertext;
|
||||
foreach (self::REPLACED_ATTRIBUTES as $initial => $final) {
|
||||
$returned = str_replace($initial . '="/', $final . '="' . self::URI . '/', $returned);
|
||||
}
|
||||
return $returned;
|
||||
}
|
||||
}
|
||||
Loading…
Reference in a new issue