X-Git-Url: https://git.immae.eu/?a=blobdiff_plain;f=inc%2F3rdparty%2Fmakefulltextfeed.php;h=135964f18477d4ad9ffe2a0d678f033b6664b4c7;hb=2a6440c134ee2d90a84b3e3a2e039d8fb4d83d04;hp=7104bc73b4b3b38a43028d46fa3096a1d540eaa4;hpb=7f667839764621b5aa01c9db8ce5dde2a29ef18f;p=github%2Fwallabag%2Fwallabag.git diff --git a/inc/3rdparty/makefulltextfeed.php b/inc/3rdparty/makefulltextfeed.php old mode 100644 new mode 100755 index 7104bc73..135964f1 --- a/inc/3rdparty/makefulltextfeed.php +++ b/inc/3rdparty/makefulltextfeed.php @@ -55,42 +55,8 @@ if (get_magic_quotes_gpc()) { // set include path set_include_path(realpath(dirname(__FILE__).'/libraries').PATH_SEPARATOR.get_include_path()); -// Autoloading of classes allows us to include files only when they're -// needed. If we've got a cached copy, for example, only Zend_Cache is loaded. -function autoload($class_name) { - static $dir = null; - if ($dir === null) $dir = dirname(__FILE__).'/libraries/'; - static $mapping = array( - // Include FeedCreator for RSS/Atom creation - 'FeedWriter' => 'feedwriter/FeedWriter.php', - 'FeedItem' => 'feedwriter/FeedItem.php', - // Include ContentExtractor and Readability for identifying and extracting content from URLs - 'ContentExtractor' => 'content-extractor/ContentExtractor.php', - 'SiteConfig' => 'content-extractor/SiteConfig.php', - 'Readability' => 'readability/Readability.php', - // Include Humble HTTP Agent to allow parallel requests and response caching - 'HumbleHttpAgent' => 'humble-http-agent/HumbleHttpAgent.php', - 'SimplePie_HumbleHttpAgent' => 'humble-http-agent/SimplePie_HumbleHttpAgent.php', - 'CookieJar' => 'humble-http-agent/CookieJar.php', - // Include Zend Cache to improve performance (cache results) - 'Zend_Cache' => 'Zend/Cache.php', - // Language detect - 'Text_LanguageDetect' => 'language-detect/LanguageDetect.php', - // HTML5 Lib - 'HTML5_Parser' => 'html5/Parser.php', - // htmLawed - used if XSS filter is enabled (xss_filter) - 'htmLawed' => 'htmLawed/htmLawed.php' - ); - if (isset($mapping[$class_name])) { - debug("** Loading class $class_name ({$mapping[$class_name]})"); - require $dir.$mapping[$class_name]; - return true; - } else { - return false; - } -} -spl_autoload_register('autoload'); -require dirname(__FILE__).'/libraries/simplepie/autoloader.php'; + +require_once dirname(__FILE__).'/makefulltextfeedHelpers.php'; //////////////////////////////// // Load config file @@ -415,6 +381,7 @@ if (!$debug_mode) { ////////////////////////////////// // Set up HTTP agent ////////////////////////////////// +global $http; $http = new HumbleHttpAgent(); $http->debug = $debug_mode; $http->userAgentMap = $options->user_agents; @@ -424,6 +391,7 @@ $http->rewriteUrls = $options->rewrite_url; ////////////////////////////////// // Set up Content Extractor ////////////////////////////////// +global $extractor; $extractor = new ContentExtractor(dirname(__FILE__).'/site_config/custom', dirname(__FILE__).'/site_config/standard'); $extractor->debug = $debug_mode; SiteConfig::$debug = $debug_mode; @@ -477,29 +445,6 @@ if ($html_only || !$result) { $isDummyFeed = true; unset($feed, $result); // create single item dummy feed object - class DummySingleItemFeed { - public $item; - function __construct($url) { $this->item = new DummySingleItem($url); } - public function get_title() { return ''; } - public function get_description() { return 'Content extracted from '.$this->item->url; } - public function get_link() { return $this->item->url; } - public function get_language() { return false; } - public function get_image_url() { return false; } - public function get_items($start=0, $max=1) { return array(0=>$this->item); } - } - class DummySingleItem { - public $url; - function __construct($url) { $this->url = $url; } - public function get_permalink() { return $this->url; } - public function get_title() { return null; } - public function get_date($format='') { return false; } - public function get_author($key=0) { return null; } - public function get_authors() { return null; } - public function get_description() { return ''; } - public function get_enclosure($key=0, $prefer=null) { return null; } - public function get_enclosures() { return null; } - public function get_categories() { return null; } - } $feed = new DummySingleItemFeed($url); } @@ -902,294 +847,3 @@ if (!$debug_mode) { if ($callback) echo ');'; } -/////////////////////////////// -// HELPER FUNCTIONS -/////////////////////////////// - -function url_allowed($url) { - global $options; - if (!empty($options->allowed_urls)) { - $allowed = false; - foreach ($options->allowed_urls as $allowurl) { - if (stristr($url, $allowurl) !== false) { - $allowed = true; - break; - } - } - if (!$allowed) return false; - } else { - foreach ($options->blocked_urls as $blockurl) { - if (stristr($url, $blockurl) !== false) { - return false; - } - } - } - return true; -} - -////////////////////////////////////////////// -// Convert $html to UTF8 -// (uses HTTP headers and HTML to find encoding) -// adapted from http://stackoverflow.com/questions/910793/php-detect-encoding-and-make-everything-utf-8 -////////////////////////////////////////////// -function convert_to_utf8($html, $header=null) -{ - $encoding = null; - if ($html || $header) { - if (is_array($header)) $header = implode("\n", $header); - if (!$header || !preg_match_all('/^Content-Type:\s+([^;]+)(?:;\s*charset=["\']?([^;"\'\n]*))?/im', $header, $match, PREG_SET_ORDER)) { - // error parsing the response - debug('Could not find Content-Type header in HTTP response'); - } else { - $match = end($match); // get last matched element (in case of redirects) - if (isset($match[2])) $encoding = trim($match[2], "\"' \r\n\0\x0B\t"); - } - // TODO: check to see if encoding is supported (can we convert it?) - // If it's not, result will be empty string. - // For now we'll check for invalid encoding types returned by some sites, e.g. 'none' - // Problem URL: http://facta.co.jp/blog/archives/20111026001026.html - if (!$encoding || $encoding == 'none') { - // search for encoding in HTML - only look at the first 50000 characters - // Why 50000? See, for example, http://www.lemonde.fr/festival-de-cannes/article/2012/05/23/deux-cretes-en-goguette-sur-la-croisette_1705732_766360.html - // TODO: improve this so it looks at smaller chunks first - $html_head = substr($html, 0, 50000); - if (preg_match('/^<\?xml\s+version=(?:"[^"]*"|\'[^\']*\')\s+encoding=("[^"]*"|\'[^\']*\')/s', $html_head, $match)) { - $encoding = trim($match[1], '"\''); - } elseif (preg_match('/]+)/i', $html_head, $match)) { - $encoding = trim($match[1]); - } elseif (preg_match_all('/]+)>/i', $html_head, $match)) { - foreach ($match[1] as $_test) { - if (preg_match('/charset=["\']?([^"\']+)/i', $_test, $_m)) { - $encoding = trim($_m[1]); - break; - } - } - } - } - if (isset($encoding)) $encoding = trim($encoding); - // trim is important here! - if (!$encoding || (strtolower($encoding) == 'iso-8859-1')) { - // replace MS Word smart qutoes - $trans = array(); - $trans[chr(130)] = '‚'; // Single Low-9 Quotation Mark - $trans[chr(131)] = 'ƒ'; // Latin Small Letter F With Hook - $trans[chr(132)] = '„'; // Double Low-9 Quotation Mark - $trans[chr(133)] = '…'; // Horizontal Ellipsis - $trans[chr(134)] = '†'; // Dagger - $trans[chr(135)] = '‡'; // Double Dagger - $trans[chr(136)] = 'ˆ'; // Modifier Letter Circumflex Accent - $trans[chr(137)] = '‰'; // Per Mille Sign - $trans[chr(138)] = 'Š'; // Latin Capital Letter S With Caron - $trans[chr(139)] = '‹'; // Single Left-Pointing Angle Quotation Mark - $trans[chr(140)] = 'Œ'; // Latin Capital Ligature OE - $trans[chr(145)] = '‘'; // Left Single Quotation Mark - $trans[chr(146)] = '’'; // Right Single Quotation Mark - $trans[chr(147)] = '“'; // Left Double Quotation Mark - $trans[chr(148)] = '”'; // Right Double Quotation Mark - $trans[chr(149)] = '•'; // Bullet - $trans[chr(150)] = '–'; // En Dash - $trans[chr(151)] = '—'; // Em Dash - $trans[chr(152)] = '˜'; // Small Tilde - $trans[chr(153)] = '™'; // Trade Mark Sign - $trans[chr(154)] = 'š'; // Latin Small Letter S With Caron - $trans[chr(155)] = '›'; // Single Right-Pointing Angle Quotation Mark - $trans[chr(156)] = 'œ'; // Latin Small Ligature OE - $trans[chr(159)] = 'Ÿ'; // Latin Capital Letter Y With Diaeresis - $html = strtr($html, $trans); - } - if (!$encoding) { - debug('No character encoding found, so treating as UTF-8'); - $encoding = 'utf-8'; - } else { - debug('Character encoding: '.$encoding); - if (strtolower($encoding) != 'utf-8') { - debug('Converting to UTF-8'); - $html = SimplePie_Misc::change_encoding($html, $encoding, 'utf-8'); - /* - if (function_exists('iconv')) { - // iconv appears to handle certain character encodings better than mb_convert_encoding - $html = iconv($encoding, 'utf-8', $html); - } else { - $html = mb_convert_encoding($html, 'utf-8', $encoding); - } - */ - } - } - } - return $html; -} - -function makeAbsolute($base, $elem) { - $base = new SimplePie_IRI($base); - // remove '//' in URL path (used to prevent URLs from resolving properly) - // TODO: check if this is still the case - if (isset($base->path)) $base->path = preg_replace('!//+!', '/', $base->path); - foreach(array('a'=>'href', 'img'=>'src') as $tag => $attr) { - $elems = $elem->getElementsByTagName($tag); - for ($i = $elems->length-1; $i >= 0; $i--) { - $e = $elems->item($i); - //$e->parentNode->replaceChild($articleContent->ownerDocument->createTextNode($e->textContent), $e); - makeAbsoluteAttr($base, $e, $attr); - } - if (strtolower($elem->tagName) == $tag) makeAbsoluteAttr($base, $elem, $attr); - } -} -function makeAbsoluteAttr($base, $e, $attr) { - if ($e->hasAttribute($attr)) { - // Trim leading and trailing white space. I don't really like this but - // unfortunately it does appear on some sites. e.g. - $url = trim(str_replace('%20', ' ', $e->getAttribute($attr))); - $url = str_replace(' ', '%20', $url); - if (!preg_match('!https?://!i', $url)) { - if ($absolute = SimplePie_IRI::absolutize($base, $url)) { - $e->setAttribute($attr, $absolute); - } - } - } -} -function makeAbsoluteStr($base, $url) { - $base = new SimplePie_IRI($base); - // remove '//' in URL path (causes URLs not to resolve properly) - if (isset($base->path)) $base->path = preg_replace('!//+!', '/', $base->path); - if (preg_match('!^https?://!i', $url)) { - // already absolute - return $url; - } else { - if ($absolute = SimplePie_IRI::absolutize($base, $url)) { - return $absolute; - } - return false; - } -} -// returns single page response, or false if not found -function getSinglePage($item, $html, $url) { - global $http, $extractor; - debug('Looking for site config files to see if single page link exists'); - $site_config = $extractor->buildSiteConfig($url, $html); - $splink = null; - if (!empty($site_config->single_page_link)) { - $splink = $site_config->single_page_link; - } elseif (!empty($site_config->single_page_link_in_feed)) { - // single page link xpath is targeted at feed - $splink = $site_config->single_page_link_in_feed; - // so let's replace HTML with feed item description - $html = $item->get_description(); - } - if (isset($splink)) { - // Build DOM tree from HTML - $readability = new Readability($html, $url); - $xpath = new DOMXPath($readability->dom); - // Loop through single_page_link xpath expressions - $single_page_url = null; - foreach ($splink as $pattern) { - $elems = @$xpath->evaluate($pattern, $readability->dom); - if (is_string($elems)) { - $single_page_url = trim($elems); - break; - } elseif ($elems instanceof DOMNodeList && $elems->length > 0) { - foreach ($elems as $item) { - if ($item instanceof DOMElement && $item->hasAttribute('href')) { - $single_page_url = $item->getAttribute('href'); - break 2; - } elseif ($item instanceof DOMAttr && $item->value) { - $single_page_url = $item->value; - break 2; - } - } - } - } - // If we've got URL, resolve against $url - if (isset($single_page_url) && ($single_page_url = makeAbsoluteStr($url, $single_page_url))) { - // check it's not what we have already! - if ($single_page_url != $url) { - // it's not, so let's try to fetch it... - $_prev_ref = $http->referer; - $http->referer = $single_page_url; - if (($response = $http->get($single_page_url, true)) && $response['status_code'] < 300) { - $http->referer = $_prev_ref; - return $response; - } - $http->referer = $_prev_ref; - } - } - } - return false; -} - -// based on content-type http header, decide what to do -// param: HTTP headers string -// return: array with keys: 'mime', 'type', 'subtype', 'action', 'name' -// e.g. array('mime'=>'image/jpeg', 'type'=>'image', 'subtype'=>'jpeg', 'action'=>'link', 'name'=>'Image') -function get_mime_action_info($headers) { - global $options; - // check if action defined for returned Content-Type - $info = array(); - if (preg_match('!^Content-Type:\s*(([-\w]+)/([-\w\+]+))!im', $headers, $match)) { - // look for full mime type (e.g. image/jpeg) or just type (e.g. image) - // match[1] = full mime type, e.g. image/jpeg - // match[2] = first part, e.g. image - // match[3] = last part, e.g. jpeg - $info['mime'] = strtolower(trim($match[1])); - $info['type'] = strtolower(trim($match[2])); - $info['subtype'] = strtolower(trim($match[3])); - foreach (array($info['mime'], $info['type']) as $_mime) { - if (isset($options->content_type_exc[$_mime])) { - $info['action'] = $options->content_type_exc[$_mime]['action']; - $info['name'] = $options->content_type_exc[$_mime]['name']; - break; - } - } - } - return $info; -} - -function remove_url_cruft($url) { - // remove google analytics for the time being - // regex adapted from http://navitronic.co.uk/2010/12/removing-google-analytics-cruft-from-urls/ - // https://gist.github.com/758177 - return preg_replace('/(\?|\&)utm_[a-z]+=[^\&]+/', '', $url); -} - -function make_substitutions($string) { - if ($string == '') return $string; - global $item, $effective_url; - $string = str_replace('{url}', htmlspecialchars($item->get_permalink()), $string); - $string = str_replace('{effective-url}', htmlspecialchars($effective_url), $string); - return $string; -} - -function get_cache() { - global $options, $valid_key; - static $cache = null; - if ($cache === null) { - $frontendOptions = array( - 'lifetime' => 10*60, // cache lifetime of 10 minutes - 'automatic_serialization' => false, - 'write_control' => false, - 'automatic_cleaning_factor' => $options->cache_cleanup, - 'ignore_user_abort' => false - ); - $backendOptions = array( - 'cache_dir' => ($valid_key) ? $options->cache_dir.'/rss-with-key/' : $options->cache_dir.'/rss/', // directory where to put the cache files - 'file_locking' => false, - 'read_control' => true, - 'read_control_type' => 'strlen', - 'hashed_directory_level' => $options->cache_directory_level, - 'hashed_directory_perm' => 0777, - 'cache_file_perm' => 0664, - 'file_name_prefix' => 'ff' - ); - // getting a Zend_Cache_Core object - $cache = Zend_Cache::factory('Core', 'File', $frontendOptions, $backendOptions); - } - return $cache; -} - -function debug($msg) { - global $debug_mode; - if ($debug_mode) { - echo '* ',$msg,"\n"; - ob_flush(); - flush(); - } -} \ No newline at end of file