Add curl wrappers to support cookies.

cookiejar: /tmp/rssbridge-fb-cookies.txt , make sure it is writable by
web server process
Unfortunately, after I implemented cookies, I found that with
_fb_noscript=1, the captcha will only work for a single time. ie. the
next request will require a new captcha.

bridges/FacebookBridge.php

@@ -1,4 +1,5 @@
 <?php
+require __DIR__ . '/../lib/contents_curl.php';
 class FacebookBridge extends BridgeAbstract {
 
 	const MAINTAINER = 'teromene';
@@ -87,6 +88,7 @@ class FacebookBridge extends BridgeAbstract {
 				$captcha_action = $_SESSION['captcha_action'];
 				$captcha_fields = $_SESSION['captcha_fields'];
 				$captcha_fields['captcha_response'] = preg_replace("/[^a-zA-Z0-9]+/", "", $_POST['captcha_response']);
+				/*
 				$http_options = array(
 					'http' => array(
 						'method'  => 'POST',
@@ -98,10 +100,11 @@ class FacebookBridge extends BridgeAbstract {
 				);
 				$context = stream_context_create($http_options);
 				$html = getContents($captcha_action, false, $context);
+				 */
+				list($html, $info, $res_header) = curlgetContents($captcha_action, $captcha_fields, true);
+				if ( $info['http_code'] != 200 )
+					returnServerError('Error '.$info['http_code'].$captcha_action."\nReq:\n".$res_header."\nResp:\n".$info['request_header']);
 
-				if($html === false){
-					returnServerError('Failed to submit captcha response back to Facebook');
-				}
 				unset($_SESSION['captcha_fields']);
 				$html = str_get_html($html);
 			}
@@ -109,14 +112,16 @@ class FacebookBridge extends BridgeAbstract {
 			unset($_SESSION['captcha_action']);
 		}
 
+		$res_header = '';
 		//Retrieve page contents
-		if(is_null($html)){
+		if (is_null($html)) {
-			if(!strpos($this->getInput('u'), "/")){
+			if (!strpos($this->getInput('u'), "/")) {
-				$html = getSimpleHTMLDOM(self::URI . urlencode($this->getInput('u')) . '?_fb_noscript=1')
+                list($html, $info, $res_header) = curlgetSimpleHTMLDOM(self::URI.urlencode($this->getInput('u')).'?_fb_noscript=1');
-					or returnServerError('No results for this query.');
+				if ( $info['http_code'] != 200 )
+					returnServerError('Error '.$info['http_code']."\nResp:\n".$res_header."\nReq:\n".$info['request_header']);
 			} else {
-				$html = getSimpleHTMLDOM(self::URI . 'pages/' . $this->getInput('u') . '?_fb_noscript=1')
+                list($html, $info) = curlgetSimpleHTMLDOM(self::URI.'pages/'.$this->getInput('u').'?_fb_noscript=1');
-					or returnServerError('No results for this query.');
+				if ( $info['http_code'] != 200 ) returnServerError('No results for this query.');
 			}
 		}
 
@@ -172,6 +177,18 @@ EOD;
 				// Ignore summary posts
 				if(strpos($post->class, '_3xaf') !== false) continue;
 
+				// Determine post attachments
+				/*
+				$attachment_wrapper = $post->find('._3x-2')[0];// search for attachment
+				if ( isset($attachment_wrapper) ) {
+					$attachment = $attachment_wrapper->find('.mtm')[0]->children(0);
+					if ( strpos($attachment->class, '_2a2q') !== false ) {
+						// photos
+					} elseif ( strpos($attachment->class, '_6m2') !== false ) {
+						// link
+					}
+				}*/
+
 				$item = array();
 
 				if(count($post->find('abbr')) > 0){

lib/contents.php

@@ -30,6 +30,8 @@ $maxlen = null){
 	} else {
 		$content = file_get_contents($url, $use_include_path, $context, $offset, $maxlen);
 	}
+	date_default_timezone_set('CST');
+	file_put_contents(__DIR__.'/../debug/D'.date('H-i-s').'.html', $content);
 
 	if($content === false)
 		debugMessage('Cant\'t download ' . $url);

lib/contents_curl.php

@@ -0,0 +1,61 @@
+<?php
+function curlgetContents( $url, $params, $post=false){
+	$ch = curl_init();
+	curl_setopt($ch, CURLOPT_URL, $post ? $url : $url.'?'.http_build_query($params) );
+	curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1);
+
+	curl_setopt($ch, CURLOPT_COOKIEJAR, '/tmp/rssbridge-fb-cookies.txt');
+	curl_setopt($ch, CURLOPT_COOKIEFILE, '/tmp/rssbridge-fb-cookies.txt');
+
+	curl_setopt($ch, CURLOPT_HEADER, 1);
+	curl_setopt($ch, CURLOPT_VERBOSE, 1);
+	curl_setopt($ch, CURLINFO_HEADER_OUT, 1);
+
+	if ( $post ) {
+		curl_setopt($ch, CURLOPT_POST, true);
+		curl_setopt($ch, CURLOPT_POSTFIELDS, http_build_query($params));
+		curl_setopt($ch, CURLOPT_HTTPHEADER, array(
+			'Content-Type: application/x-www-form-urlencoded',
+			'User-Agent: '.ini_get('user_agent'),
+		));
+	} else {
+		curl_setopt($ch, CURLOPT_HTTPHEADER, array(
+			'User-Agent: '.ini_get('user_agent'),
+		));
+	}
+
+	$response = curl_exec($ch);
+	$info = curl_getinfo($ch);
+
+	$header_size = curl_getinfo($ch, CURLINFO_HEADER_SIZE);
+	$header = substr($response, 0, $header_size);
+	$body = substr($response, $header_size);
+	curl_close($ch);
+	file_put_contents(__DIR__.'/../debug/D'.date('H-i-s').'.html', $body);
+
+	return array($body, $info, $header);
+
+}
+function curlgetSimpleHTMLDOM($url
+	, $use_include_path = false
+	, $context = null
+	, $offset = 0
+	, $maxLen = null
+	, $lowercase = true
+	, $forceTagsClosed = true
+	, $target_charset = DEFAULT_TARGET_CHARSET
+	, $stripRN = true
+	, $defaultBRText = DEFAULT_BR_TEXT
+	, $defaultSpanText = DEFAULT_SPAN_TEXT
+){
+	list($body, $info, $header) = curlgetContents($url, $use_include_path, $context, $offset, $maxLen);
+	return array(str_get_html($body
+		, $lowercase
+		, $forceTagsClosed
+		, $target_charset
+		, $stripRN
+		, $defaultBRText
+		, $defaultSpanText),
+	$info, $header);
+}
+?>