diff --git a/bridges/TwitterBridge.php b/bridges/TwitterBridge.php index 0d8b0243..6372def6 100644 --- a/bridges/TwitterBridge.php +++ b/bridges/TwitterBridge.php @@ -2,6 +2,7 @@ class TwitterBridge extends BridgeAbstract { const NAME = 'Twitter Bridge'; const URI = 'https://twitter.com/'; + const API_URI = 'https://api.twitter.com'; const CACHE_TIMEOUT = 300; // 5min const DESCRIPTION = 'returns tweets'; const MAINTAINER = 'pmaziere'; @@ -168,6 +169,27 @@ EOD } } + public function getApiURI() { + switch($this->queriedContext) { + case 'By keyword or hashtag': + return self::API_URI + . '/2/search/adaptive.json?q=' + . urlencode($this->getInput('q')) + . '&tweet_mode=extended'; + case 'By username': + return self::API_URI + . '/2/timeline/profile/' + . $this->getRestId($this->getInput('u')) + . '.json?tweet_mode=extended'; + case 'By list': + return self::API_URI + . '/2/timeline/list.json?list_id=' + . $this->getListId($this->getInput('user'), $this->getInput('list')) + . '&tweet_mode=extended'; + default: returnServerError('Invalid query context !'); + } + } + public function collectData(){ $html = ''; $page = $this->getURI(); @@ -176,14 +198,9 @@ EOD 'User-Agent: Mozilla/5.0 (Windows NT 9.0; WOW64; Trident/7.0; rv:11.0) like Gecko' ); - if(php_sapi_name() === 'cli' && empty(ini_get('curl.cainfo'))) { - $cookies = $this->getCookies($page); - $html = getSimpleHTMLDOM($page, array_merge($header, array("Cookie: $cookies"))); - } else { - $html = getSimpleHTMLDOM($page, $header, array(CURLOPT_COOKIEFILE => '')); - } + $data = json_decode($this->getApiContents($this->getApiURI())); - if(!$html) { + if(!$data) { switch($this->queriedContext) { case 'By keyword or hashtag': returnServerError('No results for this query.'); @@ -196,75 +213,33 @@ EOD $hidePictures = $this->getInput('nopic'); - foreach($html->find('div.js-stream-tweet') as $tweet) { + foreach($data->globalObjects->tweets as $tweet) { // Skip retweets? if($this->getInput('noretweet') - && $tweet->find('div.context span.js-retweet-text a', 0)) { - continue; - } - - // remove 'invisible' content - foreach($tweet->find('.invisible') as $invisible) { - $invisible->outertext = ''; - } - - // Skip protmoted tweets - $heading = $tweet->previousSibling(); - if(!is_null($heading) && - $heading->getAttribute('class') === 'promoted-tweet-heading' - ) { + && isset($tweet->retweeted_status_id_str)) { continue; } $item = array(); // extract username and sanitize - $item['username'] = htmlspecialchars_decode($tweet->getAttribute('data-screen-name'), ENT_QUOTES); - // extract fullname (pseudonym) - $item['fullname'] = htmlspecialchars_decode($tweet->getAttribute('data-name'), ENT_QUOTES); - // get author + $user_info = $this->getUserInformation($tweet->user_id_str, $data->globalObjects); + + $item['username'] = $user_info->name; + $item['fullname'] = $user_info->screen_name; $item['author'] = $item['fullname'] . ' (@' . $item['username'] . ')'; - if($rt = $tweet->find('div.context span.js-retweet-text a', 0)) { - $item['author'] .= ' RT: @' . $rt->plaintext; - } - // get avatar link - $item['avatar'] = $tweet->find('img', 0)->src; - // get TweetID - $item['id'] = $tweet->getAttribute('data-tweet-id'); - // get tweet link - $item['uri'] = self::URI . substr($tweet->find('a.js-permalink', 0)->getAttribute('href'), 1); + $item['avatar'] = $user_info->profile_image_url_https; + + $item['id'] = $tweet->id_str; + $item['uri'] = self::URI . $tweet->user_id_str . '/status/' . $item['id']; // extract tweet timestamp - $item['timestamp'] = $tweet->find('span.js-short-timestamp', 0)->getAttribute('data-time'); + $item['timestamp'] = $tweet->created_at; + // generate the title - $item['title'] = strip_tags($this->fixAnchorSpacing(htmlspecialchars_decode( - $tweet->find('p.js-tweet-text', 0), ENT_QUOTES), '')); + $item['title'] = $tweet->full_text; + $cleanedTweet = $tweet->full_text; - switch($this->queriedContext) { - case 'By list': - // Check if filter applies to list (using raw content) - if($this->getInput('filter')) { - if(stripos($tweet->find('p.js-tweet-text', 0)->plaintext, $this->getInput('filter')) === false) { - continue 2; // switch + for-loop! - } - } - break; - default: - } - - $this->processContentLinks($tweet); - $this->processEmojis($tweet); - - // get tweet text - $cleanedTweet = str_replace( - 'href="/', - 'href="' . self::URI, - $tweet->find('p.js-tweet-text', 0)->innertext - ); - - // fix anchors missing spaces in-between - $cleanedTweet = $this->fixAnchorSpacing($cleanedTweet); - - // Add picture to content + // Add avatar $picture_html = ''; if(!$hidePictures) { $picture_html = <<getImageURI($tweet); - if(!$this->getInput('noimg') && !is_null($images)) { - - foreach ($images as $image) { - - // Set image scaling - $image_orig = $this->getInput('noimgscaling') ? $image : $image . ':orig'; - $image_thumb = $this->getInput('noimgscaling') ? $image : $image . ':thumb'; - + if(isset($tweet->extended_entities->media) && !$this->getInput('noimg')) { + foreach($tweet->extended_entities->media as $media) { + $image = $media->media_url_https; + $display_image = $media->display_url; // add enclosures - $item['enclosures'][] = $image_orig; + $item['enclosures'][] = $image; $image_html .= << + + src="{$display_image}" /> EOD; } } - // add content + switch($this->queriedContext) { + case 'By list': + // Check if filter applies to list (using raw content) + if($this->getInput('filter')) { + if(stripos($cleanedTweet, $this->getInput('filter')) === false) { + continue 2; // switch + for-loop! + } + } + break; + default: + } + $item['content'] = << {$picture_html} @@ -315,151 +296,85 @@ EOD; EOD; - // add quoted tweet - $quotedTweet = $tweet->find('div.QuoteTweet', 0); - if($quotedTweet) { - // get tweet text - $cleanedQuotedTweet = str_replace( - 'href="/', - 'href="' . self::URI, - $quotedTweet->find('div.tweet-text', 0)->innertext - ); - - $this->processContentLinks($quotedTweet); - $this->processEmojis($quotedTweet); - - // Add embeded image to content - $quotedImage_html = ''; - $quotedImages = $this->getQuotedImageURI($tweet); - - if(!$this->getInput('noimg') && !is_null($quotedImages)) { - - foreach ($quotedImages as $image) { - - // Set image scaling - $image_orig = $this->getInput('noimgscaling') ? $image : $image . ':orig'; - $image_thumb = $this->getInput('noimgscaling') ? $image : $image . ':thumb'; - - // add enclosures - $item['enclosures'][] = $image_orig; - - $quotedImage_html .= << - - -EOD; - } - } - - $item['content'] = << -
-
{$cleanedQuotedTweet}
-
-
-
{$quotedImage_html}
-
-EOD; - } $item['content'] = htmlspecialchars_decode($item['content'], ENT_QUOTES); // put out $this->items[] = $item; } + + usort($this->items, array('TwitterBridge', 'compareTweetId')); } - private function processEmojis($tweet){ - // process emojis (reduce size) - foreach($tweet->find('img.Emoji') as $img) { - $img->style .= ' height: 1em;'; + private static function compareTweetId($tweet1, $tweet2) { + return (intval($tweet1['id']) < intval($tweet2['id']) ? 1 : -1); + } + + //The aim of this function is to get an API key and a guest token + //This function takes 2 requests, and therefore is cached + private function getApiKey() { + + $cacheFac = new CacheFactory(); + $cacheFac->setWorkingDir(PATH_LIB_CACHES); + $cache = $cacheFac->create(Configuration::getConfig('cache', 'type')); + $cache->setScope(get_called_class()); + $cache->setKey(array('api_key')); + $data = $cache->loadData(); + + if($data === null) { + $twitterPage = getContents('https://twitter.com'); + $jsMainRegex = '/(https:\/\/abs\.twimg\.com\/responsive-web\/web\/main\.[^\.]+\.js)/m'; + preg_match_all($jsMainRegex, $twitterPage, $jsMainMatches, PREG_SET_ORDER, 0); + $jsLink = $jsMainMatches[0][0]; + $guestTokenRegex = '/gt=([0-9]*)/m'; + preg_match_all($guestTokenRegex, $twitterPage, $guestTokenMatches, PREG_SET_ORDER, 0); + $guestToken = $guestTokenMatches[0][1]; + + $jsContent = getContents($jsLink); + $apiKeyRegex = '/([a-zA-Z0-9]{59}%[a-zA-Z0-9]{44})/m'; + preg_match_all($apiKeyRegex, $jsContent, $apiKeyMatches, PREG_SET_ORDER, 0); + $apiKey = $apiKeyMatches[0][0]; + $cache->saveData(array($apiKey, $guestToken)); + return array($apiKey, $guestToken); } + + return $data; + } - private function processContentLinks($tweet){ - // processing content links - foreach($tweet->find('a') as $link) { - if($link->hasAttribute('data-expanded-url')) { - $link->href = $link->getAttribute('data-expanded-url'); - } - $link->removeAttribute('data-expanded-url'); - $link->removeAttribute('data-query-source'); - $link->removeAttribute('rel'); - $link->removeAttribute('class'); - $link->removeAttribute('target'); - $link->removeAttribute('title'); - } + private function getApiContents($uri) { + $apiKeys = $this->getApiKey(); + $headers = array('authorization: Bearer ' . $apiKeys[0], + 'x-guest-token: ' . $apiKeys[1], + ); + return getContents($uri, $headers); } - private function fixAnchorSpacing($content){ - // fix anchors missing spaces in-between - return str_replace( - 'find('img', 0)) { - foreach ($container->find('img') as $img) { - $images[] = $img->src; + private function getUserInformation($userId, $apiData) { + foreach($apiData->users as $user) { + if($user->id_str == $userId) { + return $user; } } - - if (!empty($images)) { - return $images; - } - - return null; } - private function getQuotedImageURI($tweet){ - // Find media in tweet - $images = array(); - - $container = $tweet->find('div.QuoteMedia-container', 0); - - if($container && $container->find('img', 0)) { - foreach ($container->find('img') as $img) { - $images[] = $img->src; - } - } - - if (!empty($images)) { - return $images; - } - - return null; - } - - private function getCookies($pageURL){ - - $ctx = stream_context_create(array( - 'http' => array( - 'follow_location' => false - ) - ) - ); - $a = file_get_contents($pageURL, 0, $ctx); - - //First request to get the cookie - $cookies = ''; - foreach($http_response_header as $hdr) { - if(stripos($hdr, 'Set-Cookie') !== false) { - $cLine = explode(':', $hdr)[1]; - $cLine = explode(';', $cLine)[0]; - $cookies .= ';' . $cLine; - } - } - - return substr($cookies, 2); - } } diff --git a/cache/pages/.gitkeep b/cache/pages/.gitkeep deleted file mode 100644 index e69de29b..00000000 diff --git a/cache/server/.gitkeep b/cache/server/.gitkeep deleted file mode 100644 index e69de29b..00000000