Merge remote-tracking branch 'origin/master' into fix1

* [FeedMergeBridge] Add new bridge

Here is a bridge that merges two or more feeds into one.

Co-authored-by: Bocki <henning@bocklage.com>
Co-authored-by: Dag <me@dvikan.no>

bridges/FeedMergeBridge.php

@@ -0,0 +1,54 @@
+<?php
+
+class FeedMergeBridge extends FeedExpander {
+	const MAINTAINER = 'dvikan';
+	const NAME = 'FeedMerge';
+	const URI = 'https://github.com/RSS-Bridge/rss-bridge';
+	const DESCRIPTION = <<<'TEXT'
+This bridge merges two or more feeds into a single feed. Max 10 items are fetched from each feed.
+TEXT;
+
+	const PARAMETERS = [
+		[
+			'feed_name' => [
+				'name' => 'Feed name',
+				'type' => 'text',
+				'exampleValue' => 'rss-bridge/FeedMerger',
+			],
+			'feed_1' => [
+				'name' => 'Feed url',
+				'type' => 'text',
+				'required' => true,
+				'exampleValue' => 'https://lorem-rss.herokuapp.com/feed?unit=day'
+			],
+			'feed_2' => ['name' => 'Feed url', 'type' => 'text'],
+			'feed_3' => ['name' => 'Feed url', 'type' => 'text'],
+			'feed_4' => ['name' => 'Feed url', 'type' => 'text'],
+			'feed_5' => ['name' => 'Feed url', 'type' => 'text'],
+		]
+	];
+
+	public function collectData() {
+		$limit = 10;
+		$feeds = [
+			$this->getInput('feed_1'),
+			$this->getInput('feed_2'),
+			$this->getInput('feed_3'),
+			$this->getInput('feed_4'),
+			$this->getInput('feed_5'),
+		];
+		// Remove empty values
+		$feeds = array_filter($feeds);
+		foreach ($feeds as $feed) {
+			$this->collectExpandableDatas($feed, $limit);
+		}
+	}
+
+	public function getIcon() {
+		return 'https://cdn.jsdelivr.net/npm/famfamfam-silk@1.0.0/dist/png/folder_feed.png';
+	}
+
+	public function getName() {
+		return $this->getInput('feed_name') ?: 'rss-bridge/FeedMerger';
+	}
+}

bridges/GatesNotesBridge.php

@@ -0,0 +1,54 @@
+<?php
+class GatesNotesBridge extends FeedExpander {
+
+	const MAINTAINER = 'corenting';
+	const NAME = 'Gates Notes';
+	const URI = 'https://www.gatesnotes.com';
+	const DESCRIPTION = 'Returns the newest articles.';
+	const CACHE_TIMEOUT = 21600; // 6h
+
+	protected function parseItem($item){
+		$item = parent::parseItem($item);
+
+		$article_html = getSimpleHTMLDOMCached($item['uri']);
+		if(!$article_html) {
+			$item['content'] .= '<p><em>Could not request ' . $this->getName() . ': ' . $item['uri'] . '</em></p>';
+			return $item;
+		}
+		$article_html = defaultLinkTo($article_html, $this->getURI());
+
+		$top_description = '<p>' . $article_html->find('div.article_top_description', 0)->innertext . '</p>';
+		$hero_image = '<img src=' . $article_html->find('img.article_top_DMT_Image', 0)->getAttribute('data-src') . '>';
+
+		$article_body = $article_html->find('div.TGN_Article_ReadTimeSection', 0);
+		// Convert iframe of Youtube videos to link
+		foreach($article_body->find('iframe') as $found) {
+
+			$iframeUrl = $found->getAttribute('src');
+
+			if ($iframeUrl) {
+				$text = 'Embedded Youtube video, click here to watch on Youtube.com';
+				$found->outertext = '<p><a href="' . $iframeUrl . '">' . $text . '</a></p>';
+			}
+		}
+		// Remove <link> CSS ressources
+		foreach($article_body->find('link') as $found) {
+
+			$linkedRessourceUrl = $found->getAttribute('href');
+
+			if (str_ends_with($linkedRessourceUrl, '.css')) {
+				$found->outertext = '';
+			}
+		}
+		$article_body = sanitize($article_body->innertext);
+
+		$item['content'] = $top_description . $hero_image . $article_body;
+
+		return $item;
+	}
+
+	public function collectData(){
+		$feed = static::URI . '/rss';
+		$this->collectExpandableDatas($feed);
+	}
+}

config.default.ini.php

@@ -13,6 +13,7 @@
 timezone = "UTC"
 
 [http]
+timeout = 60
 useragent = "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/74.0.3729.169 Safari/537.36"
 
 [cache]

docs/03_For_Hosts/08_Custom_Configuration.md

@@ -17,6 +17,7 @@ __Notice__: If a parameter is not specified in your `config.ini.php` RSS-Bridge
 The configuration file is split into sections:
 
 * [system](#system)
+* [http client](#http client)
 * [cache](#cache)
 * [proxy](#proxy)
 * [authentication](#authentication)
@@ -25,6 +26,8 @@ The configuration file is split into sections:
 
 _System_: This section specifies system specific parameters
 
+_Http client_: This section has http client options
+
 _Cache_: This section is all about the caching behavior of RSS-Bridge
 
 _Proxy_: This section can be used to specify a proxy server for RSS-Bridge to utilize for fetching contents
@@ -98,6 +101,21 @@ Allow users to disable proxy usage for specific requests.
 
 `false` = disabled (default)
 
+## Http client
+
+This section provides the following parameters:
+
+- timeout
+- useragent
+
+### timeout
+
+Default network timeout.
+
+### useragent
+
+Default user agent.
+
 ## Authentication
 
 This section provides following parameters:

lib/FeedExpander.php

@@ -95,10 +95,6 @@ abstract class FeedExpander extends BridgeAbstract {
 			or returnServerError('Could not request ' . $url);
 		$rssContent = simplexml_load_string(trim($content));
 
-		if ($rssContent === false) {
-			throw new \Exception('Unable to parse string as xml');
-		}
-
 		Debug::log('Detecting feed format/version');
 		switch(true) {
 		case isset($rssContent->item[0]):

lib/contents.php

@@ -128,104 +128,74 @@ function getContents($url, $header = array(), $opts = array(), $returnHeader = f
 		'content' => '',
 	);
 
-	// Use file_get_contents() if curl module is not installed
-	if(! function_exists('curl_version')) {
+	$ch = curl_init($url);
+	curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
+	curl_setopt($ch, CURLOPT_FOLLOWLOCATION, true);
 
-		$httpHeaders = '';
+	if(is_array($header) && count($header) !== 0) {
 
-		foreach ($header as $headerL) {
-			$httpHeaders .= $headerL . "\r\n";
-		}
+		Debug::log('Setting headers: ' . json_encode($header));
+		curl_setopt($ch, CURLOPT_HTTPHEADER, $header);
 
-		$ctx = stream_context_create(array(
-			'http' => array(
-				'header' => $httpHeaders
-			)
-		));
-
-		$data = @file_get_contents($url, 0, $ctx);
-
-		if($data === false) {
-			$errorCode = 500;
-		} else {
-			$errorCode = 200;
-			$retVal['header'] = implode("\r\n", $http_response_header);
-		}
-
-		$curlError = '';
-		$curlErrno = '';
-		$headerSize = 0;
-		$finalHeader = array();
-	} else {
-		$ch = curl_init($url);
-		curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
-		curl_setopt($ch, CURLOPT_FOLLOWLOCATION, true);
-
-		if(is_array($header) && count($header) !== 0) {
-
-			Debug::log('Setting headers: ' . json_encode($header));
-			curl_setopt($ch, CURLOPT_HTTPHEADER, $header);
-
-		}
-
-		curl_setopt($ch, CURLOPT_USERAGENT, Configuration::getConfig('http', 'useragent'));
-		curl_setopt($ch, CURLOPT_ENCODING, '');
-		curl_setopt($ch, CURLOPT_PROTOCOLS, CURLPROTO_HTTP | CURLPROTO_HTTPS);
-
-		if(is_array($opts) && count($opts) !== 0) {
-
-			Debug::log('Setting options: ' . json_encode($opts));
-
-			foreach($opts as $key => $value) {
-				curl_setopt($ch, $key, $value);
-			}
-
-		}
-
-		if(defined('PROXY_URL') && !defined('NOPROXY')) {
-
-			Debug::log('Setting proxy url: ' . PROXY_URL);
-			curl_setopt($ch, CURLOPT_PROXY, PROXY_URL);
-
-		}
-
-		// We always want the response header as part of the data!
-		curl_setopt($ch, CURLOPT_HEADER, true);
-
-		// Build "If-Modified-Since" header
-		if(!Debug::isEnabled() && $time = $cache->getTime()) { // Skip if cache file doesn't exist
-			Debug::log('Adding If-Modified-Since');
-			curl_setopt($ch, CURLOPT_TIMEVALUE, $time);
-			curl_setopt($ch, CURLOPT_TIMECONDITION, CURL_TIMECOND_IFMODSINCE);
-		}
-
-		// Enables logging for the outgoing header
-		curl_setopt($ch, CURLINFO_HEADER_OUT, true);
-
-		$data = curl_exec($ch);
-		$errorCode = curl_getinfo($ch, CURLINFO_HTTP_CODE);
-
-		$curlError = curl_error($ch);
-		$curlErrno = curl_errno($ch);
-		$curlInfo = curl_getinfo($ch);
-
-		Debug::log('Outgoing header: ' . json_encode($curlInfo));
-
-		if($data === false)
-			Debug::log('Cant\'t download ' . $url . ' cUrl error: ' . $curlError . ' (' . $curlErrno . ')');
-
-		$headerSize = curl_getinfo($ch, CURLINFO_HEADER_SIZE);
-		$header = substr($data, 0, $headerSize);
-		$retVal['header'] = $header;
-
-		Debug::log('Response header: ' . $header);
-
-		$headers = parseResponseHeader($header);
-		$finalHeader = end($headers);
-
-		curl_close($ch);
 	}
 
+	curl_setopt($ch, CURLOPT_USERAGENT, Configuration::getConfig('http', 'useragent'));
+	curl_setopt($ch, CURLOPT_TIMEOUT, Configuration::getConfig('http', 'timeout'));
+	curl_setopt($ch, CURLOPT_ENCODING, '');
+	curl_setopt($ch, CURLOPT_PROTOCOLS, CURLPROTO_HTTP | CURLPROTO_HTTPS);
+
+	if(is_array($opts) && count($opts) !== 0) {
+
+		Debug::log('Setting options: ' . json_encode($opts));
+
+		foreach($opts as $key => $value) {
+			curl_setopt($ch, $key, $value);
+		}
+	}
+
+	if(defined('PROXY_URL') && !defined('NOPROXY')) {
+
+		Debug::log('Setting proxy url: ' . PROXY_URL);
+		curl_setopt($ch, CURLOPT_PROXY, PROXY_URL);
+
+	}
+
+	// We always want the response header as part of the data!
+	curl_setopt($ch, CURLOPT_HEADER, true);
+
+	// Build "If-Modified-Since" header
+	if(!Debug::isEnabled() && $time = $cache->getTime()) { // Skip if cache file doesn't exist
+		Debug::log('Adding If-Modified-Since');
+		curl_setopt($ch, CURLOPT_TIMEVALUE, $time);
+		curl_setopt($ch, CURLOPT_TIMECONDITION, CURL_TIMECOND_IFMODSINCE);
+	}
+
+	// Enables logging for the outgoing header
+	curl_setopt($ch, CURLINFO_HEADER_OUT, true);
+
+	$data = curl_exec($ch);
+	$errorCode = curl_getinfo($ch, CURLINFO_HTTP_CODE);
+
+	$curlError = curl_error($ch);
+	$curlErrno = curl_errno($ch);
+	$curlInfo = curl_getinfo($ch);
+
+	Debug::log('Outgoing header: ' . json_encode($curlInfo));
+
+	if($data === false)
+		Debug::log('Cant\'t download ' . $url . ' cUrl error: ' . $curlError . ' (' . $curlErrno . ')');
+
+	$headerSize = curl_getinfo($ch, CURLINFO_HEADER_SIZE);
+	$header = substr($data, 0, $headerSize);
+	$retVal['header'] = $header;
+
+	Debug::log('Response header: ' . $header);
+
+	$headers = parseResponseHeader($header);
+	$finalHeader = end($headers);
+
+	curl_close($ch);
+
 	$finalHeader = array_change_key_case($finalHeader, CASE_LOWER);
 
 	switch($errorCode) {