From f2f82c7d03cfa8c05717439420a706e011b48265 Mon Sep 17 00:00:00 2001 From: Riduidel Date: Mon, 3 Mar 2014 14:12:24 +0100 Subject: [PATCH 1/8] Prepared Gawker bridge by extracting file cache from initial Liberation bridge --- bridges/Les400Culs.php | 37 +---------------------------- lib/Bridge.php | 54 ++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 55 insertions(+), 36 deletions(-) diff --git a/bridges/Les400Culs.php b/bridges/Les400Culs.php index aea233b..1dd9e3e 100644 --- a/bridges/Les400Culs.php +++ b/bridges/Les400Culs.php @@ -6,7 +6,7 @@ * @update 20/02/2014 */ define("SEXE", "http://sexes.blogs.liberation.fr"); -class Les400Culs extends BridgeAbstract{ +class Les400Culs extends HttpCachingBridgeAbstract{ public function collectData(array $param){ $html = file_get_html($this->getURI()) or $this->returnError('Could not request '.$this->getURI(), 404); @@ -60,39 +60,4 @@ class Les400Culs extends BridgeAbstract{ public function getDescription(){ return "La planète sexe, vue et racontée par Agnès Giard. Et par rss-bridge"; } - - /** - * Maintain locally cached versions of pages to download to avoid multiple doiwnloads. - * A file name is generated by replacing all "/" by "_", and the file is saved below this bridge cache - * @param url url to cache - * @return content of file as string - */ - public function get_cached($url) { - $simplified_url = str_replace(["http://", "https://", "?", "&"], ["", "", "/", "/"], $url); - $filename = __DIR__ . '/../cache/'."pages/".$simplified_url; - if (substr($filename, -1) == '/') { - $filename = $filename."index.html"; - } - if(!file_exists($filename)) { - error_log("we have no local copy of ".$url." Downloading !"); - $dir = substr($filename, 0, strrpos($filename, '/')); - if(!is_dir($dir)) { - mkdir($dir, 0777, true); - } - $this->download_remote($url, $filename); - } - return file_get_contents($filename); - } - - public function download_remote($url , $save_path) { - $f = fopen( $save_path , 'w+'); - $handle = fopen($url , "rb"); - while (!feof($handle)) { - $contents = fread($handle, 8192); - fwrite($f , $contents); - } - fclose($handle); - fclose($f); - } - } diff --git a/lib/Bridge.php b/lib/Bridge.php index dbff16b..3ef4a21 100644 --- a/lib/Bridge.php +++ b/lib/Bridge.php @@ -73,6 +73,60 @@ abstract class BridgeAbstract implements BridgeInterface{ } } +/** + * Extension of BridgeAbstract allowing caching of files downloaded over http files. + * This is specially useful for sites from Gawker or Liberation networks, which allow pages excerpts top be viewed together on index, while full pages have to be downloaded + * separately. + * This class mainly provides a get_cached method which will will download the file from its remote location. + * TODO allow file cache invalidation by touching files on access, and removing files/directories which have not been touched since ... a long time + * After all, rss-bridge is not respaw, isn't it ? + */ +abstract class HttpCachingBridgeAbstract extends BridgeAbstract { + + /** + * Maintain locally cached versions of pages to download to avoid multiple doiwnloads. + * A file name is generated by replacing all "/" by "_", and the file is saved below this bridge cache + * @param url url to cache + * @return content of file as string + */ + public function get_cached($url) { + $simplified_url = str_replace(["http://", "https://", "?", "&"], ["", "", "/", "/"], $url); + // TODO build this from the variable given to Cache + $pageCacheDir = __DIR__ . '/../cache/'."pages/"; + $filename = $pageCacheDir.$simplified_url; + if (substr($filename, -1) == '/') { + $filename = $filename."index.html"; + } + if(file_exists($filename)) { + // TODO touch file and its parent, and try to do neighbour deletion + $currentPath = $filename; + while(!$pageCacheDir==$currentPath) { + touch($currentPath); + $currentPath = dirname($currentPath); + } + } else { + error_log("we have no local copy of ".$url." Downloading !"); + $dir = substr($filename, 0, strrpos($filename, '/')); + if(!is_dir($dir)) { + mkdir($dir, 0777, true); + } + $this->download_remote($url, $filename); + } + return file_get_contents($filename); + } + + public function download_remote($url , $save_path) { + $f = fopen( $save_path , 'w+'); + $handle = fopen($url , "rb"); + while (!feof($handle)) { + $contents = fread($handle, 8192); + fwrite($f , $contents); + } + fclose($handle); + fclose($f); + } +} + class Bridge{ static protected $dirBridge; From fda3e9886e2a3dcd718128799fff7b288cdfe5c9 Mon Sep 17 00:00:00 2001 From: Nicolas Delsaux Date: Wed, 26 Mar 2014 11:16:31 +0100 Subject: [PATCH 2/8] Allowed cache disabling to be performed on a per-query basis (very useful to add bridges to an existing and running installation of rss-bridge) --- index.php | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/index.php b/index.php index a95fcbb..101e014 100644 --- a/index.php +++ b/index.php @@ -39,9 +39,11 @@ try{ // Data retrieval $bridge = Bridge::create($bridge); - $bridge - ->setCache($cache) // Comment this lign for avoid cache use - ->setDatas($_REQUEST); + if(isset($_REQUEST["disable_cache"])) { + } else { + $bridge->setCache($cache); // just add disable cache to your query to disable caching + } + $bridge->setDatas($_REQUEST); // Data transformation $format = Format::create($format); From 954bc4dde0f351eaf54f2833aa0aee9ede2ed48b Mon Sep 17 00:00:00 2001 From: Nicolas Delsaux Date: Wed, 26 Mar 2014 11:20:24 +0100 Subject: [PATCH 3/8] Improved caching behaviour and allowed subclasses to easily use logging 'infrastructure' --- lib/Bridge.php | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-) diff --git a/lib/Bridge.php b/lib/Bridge.php index 3ef4a21..75f470e 100644 --- a/lib/Bridge.php +++ b/lib/Bridge.php @@ -90,7 +90,7 @@ abstract class HttpCachingBridgeAbstract extends BridgeAbstract { * @return content of file as string */ public function get_cached($url) { - $simplified_url = str_replace(["http://", "https://", "?", "&"], ["", "", "/", "/"], $url); + $simplified_url = str_replace(["http://", "https://", "?", "&", "="], ["", "", "/", "/", "/"], $url); // TODO build this from the variable given to Cache $pageCacheDir = __DIR__ . '/../cache/'."pages/"; $filename = $pageCacheDir.$simplified_url; @@ -98,6 +98,7 @@ abstract class HttpCachingBridgeAbstract extends BridgeAbstract { $filename = $filename."index.html"; } if(file_exists($filename)) { + // $this->message("loading cached file from ".$filename." for page at url ".$url); // TODO touch file and its parent, and try to do neighbour deletion $currentPath = $filename; while(!$pageCacheDir==$currentPath) { @@ -105,7 +106,7 @@ abstract class HttpCachingBridgeAbstract extends BridgeAbstract { $currentPath = dirname($currentPath); } } else { - error_log("we have no local copy of ".$url." Downloading !"); + // $this->message("we have no local copy of ".$url." Downloading !"); $dir = substr($filename, 0, strrpos($filename, '/')); if(!is_dir($dir)) { mkdir($dir, 0777, true); @@ -125,6 +126,15 @@ abstract class HttpCachingBridgeAbstract extends BridgeAbstract { fclose($handle); fclose($f); } + + public function message($text) { + $backtrace = debug_backtrace(DEBUG_BACKTRACE_IGNORE_ARGS, 3); + $calling = $backtrace[2]; + $message = $calling["file"].":".$calling["line"] + ." class ".get_class($this)."->".$calling["function"] + ." - ".$text; + error_log($message); + } } class Bridge{ From e564559bdafbb1cfe01849d0932182a31bf6cfdd Mon Sep 17 00:00:00 2001 From: Nicolas Delsaux Date: Wed, 26 Mar 2014 11:22:53 +0100 Subject: [PATCH 4/8] Fixes #42 by adding a working bridge, complete with section specific feeds --- bridges/Freenews.php | 69 ++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 69 insertions(+) create mode 100644 bridges/Freenews.php diff --git a/bridges/Freenews.php b/bridges/Freenews.php new file mode 100644 index 0000000..dad8eb6 --- /dev/null +++ b/bridges/Freenews.php @@ -0,0 +1,69 @@ +uri = RUBRIQUE.$param['id']; + } + $html = file_get_html($this->getURI()) or $this->returnError('Could not request '.$this->getURI(), 404); +// $this->message("loaded HTML from ".$this->getURI()); + // customize name + $this->name = $html->find('title', 0)->innertext; + foreach($html->find('.news_line') as $newsLines) { + $this->parseLine($newsLines); + } + } + + public function parseLine($newsLines) { + foreach($newsLines->find('span') as $newsSpan) { + foreach($newsSpan->find('a') as $newsLink) { + $item = new Item(); + $item->title = trim($newsLink->title); + $item->uri = FREENEWS.$newsLink->href; + // now load that uri from cache + $articlePage = str_get_html($this->get_cached($item->uri)); + $content = $articlePage->find('.chapo', 0); + foreach($content->find('img') as $image) { + $image->src = FREENEWS.$image->src; + } + $redaction = $articlePage->find('.redac', 0); + $rubrique = $redaction->find('a', 0); + $auteur = $redaction->find('a', 1); + $item->content = $content->innertext; + $item->name = $auteur->innertext; + // format should parse 2014-03-25T16:21:20Z. But, according to http://stackoverflow.com/a/10478469, it is not that simple + $item->timestamp = DateTime::createFromFormat('Y-m-d\TH:i:s+', $redaction->title)->getTimestamp(); + $this->items[] = $item; + // return after first link, as there are hidden treasures in those pages + return; + } + } + } + + public function getName(){ + return $this->name; + } + + public function getURI(){ + return $this->uri; + } + + public function getCacheDuration(){ + return 3600; // 2h hours + } + public function getDescription(){ + return "Un site d'actualité pour les freenautes (mais ne parlant pas que de la freebox). par rss-bridge"; + } +} From 0fa32c7cf9d3ab947766c986119372f737769654 Mon Sep 17 00:00:00 2001 From: Nicolas Delsaux Date: Wed, 26 Mar 2014 16:04:10 +0100 Subject: [PATCH 5/8] added a simple method to allow all subclasses to easily relocate image links --- lib/Bridge.php | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/lib/Bridge.php b/lib/Bridge.php index 75f470e..4467da0 100644 --- a/lib/Bridge.php +++ b/lib/Bridge.php @@ -71,6 +71,17 @@ abstract class BridgeAbstract implements BridgeInterface{ return $this; } + + /** + * Set default image SRC attribute to point on given server when none is provided (that's to say when image src starts with '/' + */ + public function defaultImageSrcTo($content, $server) { + foreach($content->find('img') as $image) { + if(strpos($image->src, '/')==0) { + $image->src = $server.$image->src; + } + } + } } /** From 1a12f48e2e872b84212a4273acc86d37073a683a Mon Sep 17 00:00:00 2001 From: Nicolas Delsaux Date: Wed, 26 Mar 2014 16:22:31 +0100 Subject: [PATCH 6/8] Working WorldOfTanks official news bridge, complete with images and categories --- bridges/WorldOfTanks.php | 63 ++++++++++++++++++++++++++++++++++++++++ 1 file changed, 63 insertions(+) create mode 100644 bridges/WorldOfTanks.php diff --git a/bridges/WorldOfTanks.php b/bridges/WorldOfTanks.php new file mode 100644 index 0000000..1a6e625 --- /dev/null +++ b/bridges/WorldOfTanks.php @@ -0,0 +1,63 @@ +lang = $param['lang']; + } + if(empty($param['category'])) { + $this->uri = WORLD_OF_TANKS.$this->lang.NEWS; + } else { + $this->uri = WORLD_OF_TANKS.$this->lang.NEWS.$param['category']."/"; + } + $html = file_get_html($this->getURI()) or $this->returnError('Could not request '.$this->getURI(), 404); + $this->message("loaded HTML from ".$this->getURI()); + // customize name + $this->name = $html->find('title', 0)->innertext; + foreach($html->find('.b-imgblock_ico') as $infoLink) { + $this->parseLine($infoLink); + } + } + + public function parseLine($infoLink) { + $item = new Item(); + $item->uri = WORLD_OF_TANKS.$infoLink->href; + // now load that uri from cache +// $this->message("loading page ".$item->uri); + $articlePage = str_get_html($this->get_cached($item->uri)); + $content = $articlePage->find('.l-content', 0); + $this->defaultImageSrcTo($content, WORLD_OF_TANKS); + $item->title = $content->find('h1', 0)->innertext; + $item->content = $content->find('.b-content', 0)->innertext; +// $item->name = $auteur->innertext; + $item->timestamp = $content->find('.b-statistic_time', 0)->getAttribute("data-timestamp"); + $this->items[] = $item; + } + + public function getName(){ + return $this->name; + } + + public function getURI(){ + return $this->uri; + } + + public function getCacheDuration(){ + return 3600; // 2h hours + } + public function getDescription(){ + return "Toutes les actualités les plus brulantes de ce simulateur de destruction d'acier."; + } +} From fd71ceae821a1e6b597a252e16c86bb682e72570 Mon Sep 17 00:00:00 2001 From: Nicolas Delsaux Date: Wed, 2 Apr 2014 10:55:42 +0100 Subject: [PATCH 7/8] Adding a meta-bridge for all Gawker publications (lifehacker, kotaku, you name it, ...) --- bridges/Gawker.php | 89 ++++++++++++++++++++++++++++++++++++++++++++++ lib/Bridge.php | 47 +++++++++++++++++------- 2 files changed, 123 insertions(+), 13 deletions(-) create mode 100644 bridges/Gawker.php diff --git a/bridges/Gawker.php b/bridges/Gawker.php new file mode 100644 index 0000000..4d13522 --- /dev/null +++ b/bridges/Gawker.php @@ -0,0 +1,89 @@ +uri = $param['site']; + } + $html = file_get_html($this->getURI()) or $this->returnError('Could not request '.$this->getURI(), 404); + $this->message("loaded HTML from ".$this->getURI()); + // customize name + $this->name = $html->find('title', 0)->innertext; + foreach($html->find('.main-column') as $content) { + $this->parseContent($content); + } + } + + public function parseContent($content) { + foreach($content->find('.headline') as $headline) { + foreach($headline->find('a') as $articleLink) { + // notice we only use article from this gawker site (as gawker like to see us visit other sites) + if(strpos($articleLink->href, $this->getURI())>=0) { + $this->parseLink($articleLink); + } + } + } + } + + public function parseLink($infoLink) { + $item = new Item(); + $item->uri = $infoLink->href; + $item->title = $infoLink->innertext; + try { + // now load that uri from cache +// $this->message("loading page ".$item->uri); + $articlePage = str_get_html($this->get_cached($item->uri)); + if(is_object($articlePage)) { + $content = $articlePage->find('.post-content', 0); + $this->defaultImageSrcTo($content, $this->getURI()); + $item->content = $content->innertext; + // http://stackoverflow.com/q/22715928/15619 + $publishtime = $articlePage->find('.publish-time', 0)->getAttribute("data-publishtime"); + // don't know what I'm doing there, but http://www.epochconverter.com/programming/functions-php.php#epoch2date recommends it + $item->timestamp = $this->js_to_unix_timestamp($publishtime); + $vcard = $articlePage->find('.vcard', 0); + if(is_object($vcard)) { + $item->name = $vcard->find('a', 0)->innertext; + } + } else { + throw new Exception("cache content for ".$item->uri." is NOT a Simple DOM parser object !"); + } + } catch(Exception $e) { + $this->message("obtaining ".$item->uri." resulted in exception ".$e->getMessage().". Deleting cached page ..."); + // maybe file is incorrect. it should be discarded from cache + $this->remove_from_cache($item->url); + $item->content = $e->getMessage(); + } + $this->items[] = $item; + } + + function js_to_unix_timestamp($jsTimestamp){ + return $jsTimestamp/1000; + } + + public function getName(){ + return $this->name; + } + + public function getURI(){ + return $this->uri; + } + + public function getCacheDuration(){ + return 3600; // 1h + } + public function getDescription(){ + return "Gawker press blog content."; + } +} diff --git a/lib/Bridge.php b/lib/Bridge.php index 4467da0..aeaba6f 100644 --- a/lib/Bridge.php +++ b/lib/Bridge.php @@ -109,33 +109,54 @@ abstract class HttpCachingBridgeAbstract extends BridgeAbstract { $filename = $filename."index.html"; } if(file_exists($filename)) { - // $this->message("loading cached file from ".$filename." for page at url ".$url); +// $this->message("loading cached file from ".$filename." for page at url ".$url); // TODO touch file and its parent, and try to do neighbour deletion - $currentPath = $filename; - while(!$pageCacheDir==$currentPath) { - touch($currentPath); - $currentPath = dirname($currentPath); - } + $this->refresh_in_cache($pageCacheDir, $filename); } else { - // $this->message("we have no local copy of ".$url." Downloading !"); +// $this->message("we have no local copy of ".$url." Downloading to ".$filename); $dir = substr($filename, 0, strrpos($filename, '/')); if(!is_dir($dir)) { +// $this->message("creating directories for ".$dir); mkdir($dir, 0777, true); } $this->download_remote($url, $filename); } return file_get_contents($filename); } + + private function refresh_in_cache($pageCacheDir, $filename) { + $currentPath = $filename; + while(!$pageCacheDir==$currentPath) { + touch($currentPath); + $currentPath = dirname($currentPath); + } + } public function download_remote($url , $save_path) { $f = fopen( $save_path , 'w+'); - $handle = fopen($url , "rb"); - while (!feof($handle)) { - $contents = fread($handle, 8192); - fwrite($f , $contents); + if($f) { + $handle = fopen($url , "rb"); + if($handle) { + while (!feof($handle)) { + $contents = fread($handle, 8192); + if($contents) { + fwrite($f , $contents); + } + } + fclose($handle); + } + fclose($f); } - fclose($handle); - fclose($f); + } + + public function remove_from_cache($url) { + $simplified_url = str_replace(["http://", "https://", "?", "&", "="], ["", "", "/", "/", "/"], $url); + // TODO build this from the variable given to Cache + $pageCacheDir = __DIR__ . '/../cache/'."pages/"; + $filename = realpath($pageCacheDir.$simplified_url); + $this->message("removing from cache \"".$filename."\" WELL, NOT REALLY"); + // filename is NO GOOD +// unlink($filename); } public function message($text) { From ce28cfad221329f645f328379b88f46995b395ec Mon Sep 17 00:00:00 2001 From: Nicolas Delsaux Date: Wed, 18 Mar 2015 17:42:55 +0100 Subject: [PATCH 8/8] Rewritten those bridge to use a new mechanism ! RSS expander will get data from the RSS feed and expand the truncated entries to their full values. --- bridges/Freenews.php | 82 +++++++++++++---------------------------- bridges/Gawker.php | 79 +++++++++++++-------------------------- bridges/Les400Culs.php | 73 ++++++++++++++---------------------- bridges/RssExpander.php | 68 ++++++++++++++++++++++++++++++++++ bridges/Sexactu.php | 2 +- 5 files changed, 147 insertions(+), 157 deletions(-) create mode 100644 bridges/RssExpander.php diff --git a/bridges/Freenews.php b/bridges/Freenews.php index dad8eb6..caaf769 100644 --- a/bridges/Freenews.php +++ b/bridges/Freenews.php @@ -1,69 +1,37 @@ uri = RUBRIQUE.$param['id']; - } - $html = file_get_html($this->getURI()) or $this->returnError('Could not request '.$this->getURI(), 404); -// $this->message("loaded HTML from ".$this->getURI()); - // customize name - $this->name = $html->find('title', 0)->innertext; - foreach($html->find('.news_line') as $newsLines) { - $this->parseLine($newsLines); - } + $param['url'] = RSS; + parent::collectData($param); } - public function parseLine($newsLines) { - foreach($newsLines->find('span') as $newsSpan) { - foreach($newsSpan->find('a') as $newsLink) { - $item = new Item(); - $item->title = trim($newsLink->title); - $item->uri = FREENEWS.$newsLink->href; - // now load that uri from cache - $articlePage = str_get_html($this->get_cached($item->uri)); - $content = $articlePage->find('.chapo', 0); - foreach($content->find('img') as $image) { - $image->src = FREENEWS.$image->src; - } - $redaction = $articlePage->find('.redac', 0); - $rubrique = $redaction->find('a', 0); - $auteur = $redaction->find('a', 1); - $item->content = $content->innertext; - $item->name = $auteur->innertext; - // format should parse 2014-03-25T16:21:20Z. But, according to http://stackoverflow.com/a/10478469, it is not that simple - $item->timestamp = DateTime::createFromFormat('Y-m-d\TH:i:s+', $redaction->title)->getTimestamp(); - $this->items[] = $item; - // return after first link, as there are hidden treasures in those pages - return; - } - } - } + protected function parseRSSItem($newsItem) { + $item = new Item(); + $item->title = trim($newsItem->title); +// $this->message("item has for title \"".$item->title."\""); + if(empty($newsItem->guid)) { + $item->uri = $newsItem->link; + } else { + $item->uri = $newsItem->guid; + } + // now load that uri from cache +// $this->message("now loading page ".$item->uri); + $articlePage = str_get_html($this->get_cached($item->uri)); - public function getName(){ - return $this->name; - } - - public function getURI(){ - return $this->uri; - } - - public function getCacheDuration(){ - return 3600; // 2h hours - } - public function getDescription(){ - return "Un site d'actualité pour les freenautes (mais ne parlant pas que de la freebox). par rss-bridge"; + $content = $articlePage->find('.post-container', 0); + $item->content = $content->innertext; + $item->name = $articlePage->find('a[rel=author]', 0)->innertext; + // format should parse 2014-03-25T16:21:20Z. But, according to http://stackoverflow.com/a/10478469, it is not that simple + $item->timestamp = $this->RSS_2_0_time_to_timestamp($newsItem); + return $item; } } diff --git a/bridges/Gawker.php b/bridges/Gawker.php index 4d13522..f8b484f 100644 --- a/bridges/Gawker.php +++ b/bridges/Gawker.php @@ -2,44 +2,36 @@ /** * * @name Gawker media -* @description A bridge allowing access to any of the numerous Gawker media blogs (Lifehacker, deadspin, Kotaku, Jezebel, and so on +* @description A bridge allowing access to any of the numerous Gawker media blogs (Lifehacker, deadspin, Kotaku, Jezebel, and so on. Notice you have to give its id to find the RSS stream in gawker maze * @update 27/03/2014 -* @use1(site="site") +* @use1(site="site id to put in uri between feeds.gawker.com and /full .. which is obviously not full AT ALL") */ -class Gawker extends HttpCachingBridgeAbstract{ - private $uri; - private $name; +require_once 'bridges/RssExpander.php'; +define("RSS_PREFIX", "http://feeds.gawker.com/"); +define("RSS_SUFFIX", "/full"); +class Gawker extends RssExpander{ + + private function toURI($name) { + return RSS_PREFIX.$name.RSS_SUFFIX; + } public function collectData(array $param){ if (empty($param['site'])) { trigger_error("If no site is provided, nothing is gonna happen", E_USER_ERROR); } else { - $this->uri = $param['site']; + $this->name = $param['site']; + $param['url'] = $this->toURI(strtolower($param['site'])); } - $html = file_get_html($this->getURI()) or $this->returnError('Could not request '.$this->getURI(), 404); - $this->message("loaded HTML from ".$this->getURI()); - // customize name - $this->name = $html->find('title', 0)->innertext; - foreach($html->find('.main-column') as $content) { - $this->parseContent($content); - } +// $this->message("loading feed from ".$this->getURI()); + parent::collectData($param); } - - public function parseContent($content) { - foreach($content->find('.headline') as $headline) { - foreach($headline->find('a') as $articleLink) { - // notice we only use article from this gawker site (as gawker like to see us visit other sites) - if(strpos($articleLink->href, $this->getURI())>=0) { - $this->parseLink($articleLink); - } - } - } - } - public function parseLink($infoLink) { + protected function parseRSSItem($newsItem) { $item = new Item(); - $item->uri = $infoLink->href; - $item->title = $infoLink->innertext; + $item->uri = trim($newsItem->link); + $item->title = trim($newsItem->title); + $item->timestamp = $this->RSS_2_0_time_to_timestamp($newsItem); +// $this->message("///////////////////////////////////////////////////////////////////////////////////////\nprocessing item ".var_export($item, true)."\n\n\nbuilt from\n\n\n".var_export($newsItem, true)); try { // now load that uri from cache // $this->message("loading page ".$item->uri); @@ -47,15 +39,15 @@ class Gawker extends HttpCachingBridgeAbstract{ if(is_object($articlePage)) { $content = $articlePage->find('.post-content', 0); $this->defaultImageSrcTo($content, $this->getURI()); - $item->content = $content->innertext; - // http://stackoverflow.com/q/22715928/15619 - $publishtime = $articlePage->find('.publish-time', 0)->getAttribute("data-publishtime"); - // don't know what I'm doing there, but http://www.epochconverter.com/programming/functions-php.php#epoch2date recommends it - $item->timestamp = $this->js_to_unix_timestamp($publishtime); $vcard = $articlePage->find('.vcard', 0); if(is_object($vcard)) { - $item->name = $vcard->find('a', 0)->innertext; + $authorLink = $vcard->find('a', 0); + $item->name = $authorLink->innertext; + // TODO use author link href to fill the feed info } +// $this->message("item quite loaded : ".var_export($item, true)); + // I set item content as last element, for easier var_export reading + $item->content = $content->innertext; } else { throw new Exception("cache content for ".$item->uri." is NOT a Simple DOM parser object !"); } @@ -65,25 +57,6 @@ class Gawker extends HttpCachingBridgeAbstract{ $this->remove_from_cache($item->url); $item->content = $e->getMessage(); } - $this->items[] = $item; - } - - function js_to_unix_timestamp($jsTimestamp){ - return $jsTimestamp/1000; - } - - public function getName(){ - return $this->name; - } - - public function getURI(){ - return $this->uri; - } - - public function getCacheDuration(){ - return 3600; // 1h - } - public function getDescription(){ - return "Gawker press blog content."; + return $item; } } diff --git a/bridges/Les400Culs.php b/bridges/Les400Culs.php index 1dd9e3e..e6dee73 100644 --- a/bridges/Les400Culs.php +++ b/bridges/Les400Culs.php @@ -5,59 +5,40 @@ * @description La planète sexe vue par Agnès Girard via rss-bridge * @update 20/02/2014 */ +require_once 'bridges/RssExpander.php'; define("SEXE", "http://sexes.blogs.liberation.fr"); -class Les400Culs extends HttpCachingBridgeAbstract{ +define("RSS", "http://sexes.blogs.liberation.fr/feeds/"); +/** + * As it seems that Les 400 culs currently offer a full feed, we won't change it content here. + * But I'm ready for the day where it will ... again ... provide some truncated content + */ +class Les400Culs extends RssExpander{ public function collectData(array $param){ - $html = file_get_html($this->getURI()) or $this->returnError('Could not request '.$this->getURI(), 404); - - foreach($html->find('#alpha-inner') as $articles) { - foreach($articles->find('div.entry') as $article) { - $header = $article->find('h3.entry-header a', 0); - $content = $article->find('div.entry-content', 0); - - - $item = new Item(); - $item->title = trim($header->innertext); - $item->uri = $header->href; - $item->name = "Agnès Girard"; - // date is stored outside this node ! - $dateHeader = $article->prev_sibling(); - // http://stackoverflow.com/a/6239199/15619 (strtotime is typical amercian bullshit) - $item->timestamp = DateTime::createFromFormat('d/m/Y', $dateHeader->innertext)->getTimestamp(); - - - $linkForMore = $content->find('p.entry-more-link a',0); - if($linkForMore==null) { - $item->content = $content->innertext; - } else { - $pageAddress = $linkForMore->href; - $articlePage = str_get_html($this->get_cached($linkForMore->href)); - if($articlePage==null) { - $item->content = $content->innertext."\n

".$linkForMore->outertext."

"; - } else { - // TODO use some caching there ! - $fullContent = $articlePage->find('div.entry-content', 0); - $item->content = $fullContent->innertext; - } - } - $this->items[] = $item; - } - } + $param['url'] = RSS; + parent::collectData($param); } + + protected function parseRSSItem($newsItem) { + $item = new Item(); + $item->title = trim($newsItem->title); +// $this->message("browsing item ".var_export($newsItem, true)); + if(empty($newsItem->guid)) { + $item->uri = $newsItem->link; + } else { + $item->uri = $newsItem->guid; + } + // now load that uri from cache +// $this->message("now loading page ".$item->uri); +// $articlePage = str_get_html($this->get_cached($item->uri)); - public function getName(){ - return 'Les 400 Culs'; +// $content = $articlePage->find('.post-container', 0); + $item->content = $newsItem->description; + $item->name = $newsItem->author; + $item->timestamp = $this->RSS_2_0_time_to_timestamp($newsItem); + return $item; } - - public function getURI(){ - return SEXE; - } - public function getCacheDuration(){ return 7200; // 2h hours } - public function getDescription(){ - return "La planète sexe, vue et racontée par Agnès Giard. Et par rss-bridge"; - } } diff --git a/bridges/RssExpander.php b/bridges/RssExpander.php new file mode 100644 index 0000000..07268ea --- /dev/null +++ b/bridges/RssExpander.php @@ -0,0 +1,68 @@ +returnError('There is no $param[\'url\'] for this RSS expander', 404); + } + // $this->message("Loading from ".$param['url']); + // Notice WE DO NOT use cache here on purpose : we want a fresh view of the RSS stream each time + $rssContent = simplexml_load_file($param['url']) or $this->returnError('Could not request '.$param['url'], 404); +// $this->message("loaded RSS from ".$param['url']); + // TODO insert RSS format detection + // we suppose for now, we have some RSS 2.0 + $this->collect_RSS_2_0_data($rssContent); + } + + private function collect_RSS_2_0_data($rssContent) { + $rssContent = $rssContent->channel[0]; +// $this->message("RSS content is ===========\n".var_export($rssContent, true)."==========="); + $this->load_RSS_2_0_feed_data($rssContent); + foreach($rssContent->item as $item) { +// $this->message("parsing item ".var_export($item, true)); + $this->items[] = $this->parseRSSItem($item); + } + } + + protected function RSS_2_0_time_to_timestamp($item) { + return DateTime::createFromFormat('D, d M Y H:i:s e', $item->pubDate)->getTimestamp(); + } + + // TODO set title, link, description, language, and so on + protected function load_RSS_2_0_feed_data($rssContent) { + $this->name = trim($rssContent->title); + $this->uri = trim($rssContent->link); + $this->description = trim($rssContent->description); + } + + /** + * Method should return, from a source RSS item given by lastRSS, one of our Items objects + * @param $item the input rss item + * @return a RSS-Bridge Item, with (hopefully) the whole content) + */ + abstract protected function parseRSSItem($item); + + + public function getName(){ + return $this->name; + } + + public function getURI(){ + return $this->uri; + } + + public function getDescription() { + return $this->description; + } +} \ No newline at end of file diff --git a/bridges/Sexactu.php b/bridges/Sexactu.php index 20ba5f9..edc173f 100644 --- a/bridges/Sexactu.php +++ b/bridges/Sexactu.php @@ -1,6 +1,6 @@