// Author: Keyvan Minoukadeh\r
// Copyright (c) 2013 Keyvan Minoukadeh\r
// License: AGPLv3\r
-// Version: 3.1\r
-// Date: 2013-03-05\r
+// Version: 3.2\r
+// Date: 2013-05-13\r
// More info: http://fivefilters.org/content-only/\r
// Help: http://help.fivefilters.org\r
\r
\r
// Usage\r
// -----\r
-// Request this file passing it your feed in the querystring: makefulltextfeed.php?url=mysite.org\r
-// The following options can be passed in the querystring:\r
-// * URL: url=[feed or website url] (required, should be URL-encoded - in php: urlencode($url))\r
-// * URL points to HTML (not feed): html=true (optional, by default it's automatically detected)\r
-// * API key: key=[api key] (optional, refer to config.php)\r
-// * Max entries to process: max=[max number of items] (optional)\r
+// Request this file passing it a web page or feed URL in the querystring: makefulltextfeed.php?url=example.org/article\r
+// For more request parameters, see http://help.fivefilters.org/customer/portal/articles/226660-usage\r
\r
error_reporting(E_ALL ^ E_NOTICE);\r
ini_set("display_errors", 1);\r
////////////////////////////////\r
// Check if service is enabled\r
////////////////////////////////\r
-if (!$options->enabled) { \r
- die('The full-text RSS service is currently disabled'); \r
+if (!$options->enabled) {\r
+ die('The full-text RSS service is currently disabled');\r
}\r
\r
////////////////////////////////\r
////////////////////////////////\r
// Check for feed URL\r
////////////////////////////////\r
-if (!isset($_GET['url'])) { \r
- die('No URL supplied'); \r
+if (!isset($_GET['url'])) {\r
+ die('No URL supplied');\r
}\r
$url = trim($_GET['url']);\r
if (strtolower(substr($url, 0, 7)) == 'feed://') {\r
if (isset($_GET['links'])) $redirect .= '&links='.urlencode($_GET['links']);\r
if (isset($_GET['exc'])) $redirect .= '&exc='.urlencode($_GET['exc']);\r
if (isset($_GET['format'])) $redirect .= '&format='.urlencode($_GET['format']);\r
- if (isset($_GET['callback'])) $redirect .= '&callback='.urlencode($_GET['callback']); \r
+ if (isset($_GET['callback'])) $redirect .= '&callback='.urlencode($_GET['callback']);\r
if (isset($_GET['l'])) $redirect .= '&l='.urlencode($_GET['l']);\r
if (isset($_GET['xss'])) $redirect .= '&xss';\r
if (isset($_GET['use_extracted_title'])) $redirect .= '&use_extracted_title';\r
+ if (isset($_GET['content'])) $redirect .= '&content='.urlencode($_GET['content']);\r
+ if (isset($_GET['summary'])) $redirect .= '&summary='.urlencode($_GET['summary']);\r
if (isset($_GET['debug'])) $redirect .= '&debug';\r
if ($debug_mode) {\r
debug('Redirecting to hide access key, follow URL below to continue');\r
\r
///////////////////////////////////////////////\r
// Set timezone.\r
-// Prevents warnings, but needs more testing - \r
+// Prevents warnings, but needs more testing -\r
// perhaps if timezone is set in php.ini we\r
// don't need to set it at all...\r
///////////////////////////////////////////////\r
}\r
$key_index = ($valid_key) ? (int)$_GET['key'] : 0;\r
if (!$valid_key && $options->key_required) {\r
- die('A valid key must be supplied'); \r
+ die('A valid key must be supplied');\r
}\r
if (!$valid_key && isset($_GET['key']) && $_GET['key'] != '') {\r
die('The entered key is invalid');\r
$favour_feed_titles = $options->favour_feed_titles;\r
}\r
\r
+///////////////////////////////////////////////\r
+// Include full content in output?\r
+///////////////////////////////////////////////\r
+if ($options->content === 'user') {\r
+ if (isset($_GET['content']) && $_GET['content'] === '0') {\r
+ $options->content = false;\r
+ } else {\r
+ $options->content = true;\r
+ }\r
+}\r
+\r
+///////////////////////////////////////////////\r
+// Include summaries in output?\r
+///////////////////////////////////////////////\r
+if ($options->summary === 'user') {\r
+ if (isset($_GET['summary']) && $_GET['summary'] === '1') {\r
+ $options->summary = true;\r
+ } else {\r
+ $options->summary = false;\r
+ }\r
+}\r
+\r
///////////////////////////////////////////////\r
// Exclude items if extraction fails\r
///////////////////////////////////////////////\r
$detect_language = $options->detect_language;\r
}\r
\r
-if ($detect_language >= 2) {\r
- $language_codes = array('albanian' => 'sq','arabic' => 'ar','azeri' => 'az','bengali' => 'bn','bulgarian' => 'bg',\r
- 'cebuano' => 'ceb', // ISO 639-2\r
- 'croatian' => 'hr','czech' => 'cs','danish' => 'da','dutch' => 'nl','english' => 'en','estonian' => 'et','farsi' => 'fa','finnish' => 'fi','french' => 'fr','german' => 'de','hausa' => 'ha',\r
- 'hawaiian' => 'haw', // ISO 639-2 \r
- 'hindi' => 'hi','hungarian' => 'hu','icelandic' => 'is','indonesian' => 'id','italian' => 'it','kazakh' => 'kk','kyrgyz' => 'ky','latin' => 'la','latvian' => 'lv','lithuanian' => 'lt','macedonian' => 'mk','mongolian' => 'mn','nepali' => 'ne','norwegian' => 'no','pashto' => 'ps',\r
- 'pidgin' => 'cpe', // ISO 639-2 \r
- 'polish' => 'pl','portuguese' => 'pt','romanian' => 'ro','russian' => 'ru','serbian' => 'sr','slovak' => 'sk','slovene' => 'sl','somali' => 'so','spanish' => 'es','swahili' => 'sw','swedish' => 'sv','tagalog' => 'tl','turkish' => 'tr','ukrainian' => 'uk','urdu' => 'ur','uzbek' => 'uz','vietnamese' => 'vi','welsh' => 'cy');\r
-}\r
$use_cld = extension_loaded('cld') && (version_compare(PHP_VERSION, '5.3.0') >= 0);\r
\r
/////////////////////////////////////\r
//////////////////////////////////\r
if ($options->caching) {\r
debug('Caching is enabled...');\r
- $cache_id = md5($max.$url.$valid_key.$links.$favour_feed_titles.$xss_filter.$exclude_on_fail.$format.$detect_language.(int)isset($_GET['pubsub']));\r
+ $cache_id = md5($max.$url.(int)$valid_key.$links.(int)$favour_feed_titles.(int)$options->content.(int)$options->summary.(int)$xss_filter.(int)$exclude_on_fail.$format.$detect_language.(int)isset($_GET['pubsub']));\r
$check_cache = true;\r
if ($options->apc && $options->smart_cache) {\r
apc_add("cache.$cache_id", 0, 10*60);\r
////////////////////////////////////////////\r
// Loop through feed items\r
////////////////////////////////////////////\r
-$items = $feed->get_items(0, $max); \r
+$items = $feed->get_items(0, $max);\r
// Request all feed items in parallel (if supported)\r
$urls_sanitized = array();\r
$urls = array();\r
$is_single_page = false;\r
if ($single_page_response = getSinglePage($item, $html, $effective_url)) {\r
$is_single_page = true;\r
- $html = $single_page_response['body'];\r
- // remove strange things\r
- $html = str_replace('</[>', '', $html); \r
- $html = convert_to_utf8($html, $single_page_response['headers']);\r
$effective_url = $single_page_response['effective_url'];\r
- debug("Retrieved single-page view from $effective_url");\r
+ // check if action defined for returned Content-Type\r
+ $mime_info = get_mime_action_info($single_page_response['headers']);\r
+ if (isset($mime_info['action'])) {\r
+ if ($mime_info['action'] == 'exclude') {\r
+ continue; // skip this feed item entry\r
+ } elseif ($mime_info['action'] == 'link') {\r
+ if ($mime_info['type'] == 'image') {\r
+ $html = "<a href=\"$effective_url\"><img src=\"$effective_url\" alt=\"{$mime_info['name']}\" /></a>";\r
+ } else {\r
+ $html = "<a href=\"$effective_url\">Download {$mime_info['name']}</a>";\r
+ }\r
+ $extracted_title = $mime_info['name'];\r
+ $do_content_extraction = false;\r
+ }\r
+ }\r
+ if ($do_content_extraction) {\r
+ $html = $single_page_response['body'];\r
+ // remove strange things\r
+ $html = str_replace('</[>', '', $html);\r
+ $html = convert_to_utf8($html, $single_page_response['headers']);\r
+ debug("Retrieved single-page view from $effective_url");\r
+ }\r
unset($single_page_response);\r
}\r
+ }\r
+ if ($do_content_extraction) {\r
debug('--------');\r
debug('Attempting to extract content');\r
$extract_result = $extractor->process($html, $effective_url);\r
$readability = $extractor->readability;\r
- $content_block = ($extract_result) ? $extractor->getContent() : null; \r
+ $content_block = ($extract_result) ? $extractor->getContent() : null;\r
$extracted_title = ($extract_result) ? $extractor->getTitle() : '';\r
// Deal with multi-page articles\r
//die('Next: '.$extractor->getNextPageUrl());\r
$is_multi_page = (!$is_single_page && $extract_result && $extractor->getNextPageUrl());\r
- if ($options->multipage && $is_multi_page) {\r
+ if ($options->multipage && $is_multi_page && $options->content) {\r
debug('--------');\r
debug('Attempting to process multi-page article');\r
$multi_page_urls = array();\r
// check it's not what we have already!\r
if (!in_array($next_page_url, $multi_page_urls)) {\r
// it's not, so let's attempt to fetch it\r
- $multi_page_urls[] = $next_page_url; \r
+ $multi_page_urls[] = $next_page_url;\r
$_prev_ref = $http->referer;\r
if (($response = $http->get($next_page_url, true)) && $response['status_code'] < 300) {\r
// make sure mime type is not something with a different action associated\r
// did we successfully deal with this multi-page article?\r
if (empty($multi_page_content)) {\r
debug('Failed to extract all parts of multi-page article, so not going to include them');\r
- $multi_page_content[] = $readability->dom->createElement('p')->innerHTML = '<em>This article appears to continue on subsequent pages which we could not extract</em>';\r
+ $_page = $readability->dom->createElement('p');\r
+ $_page->innerHTML = '<em>This article appears to continue on subsequent pages which we could not extract</em>';\r
+ $multi_page_content[] = $_page;\r
}\r
foreach ($multi_page_content as $_page) {\r
$_page = $content_block->ownerDocument->importNode($_page, true);\r
$content_block->appendChild($_page);\r
}\r
- unset($multi_page_urls, $multi_page_content, $page_mime_info, $next_page_url);\r
+ unset($multi_page_urls, $multi_page_content, $page_mime_info, $next_page_url, $_page);\r
}\r
}\r
// use extracted title for both feed and item title if we're using single-item dummy feed\r
} else {\r
$html = $content_block->ownerDocument->saveXML($content_block); // essentially outerHTML\r
}\r
- unset($content_block);\r
+ //unset($content_block);\r
// post-processing cleanup\r
$html = preg_replace('!<p>[\s\h\v]*</p>!u', '', $html);\r
if ($links == 'remove') {\r
}\r
}\r
\r
- if ($valid_key && isset($_GET['pubsub'])) { // used only on fivefilters.org at the moment\r
- $newitem->addElement('guid', 'http://fivefilters.org/content-only/redirect.php?url='.urlencode($item->get_permalink()), array('isPermaLink'=>'false'));\r
+ if ($valid_key && isset($_GET['pubsub'])) { // used only on fivefilters.org at the moment\r
+ $newitem->addElement('guid', 'http://fivefilters.org/content-only/redirect.php?url='.urlencode($item->get_permalink()), array('isPermaLink'=>'false'));\r
+ } else {\r
+ $newitem->addElement('guid', $item->get_permalink(), array('isPermaLink'=>'true'));\r
+ }\r
+ // filter xss?\r
+ if ($xss_filter) {\r
+ debug('Filtering HTML to remove XSS');\r
+ $html = htmLawed::hl($html, array('safe'=>1, 'deny_attribute'=>'style', 'comment'=>1, 'cdata'=>1));\r
+ }\r
+\r
+ // add content\r
+ if ($options->summary === true) {\r
+ // get summary\r
+ $summary = '';\r
+ if (!$do_content_extraction) {\r
+ $summary = $html;\r
} else {\r
- $newitem->addElement('guid', $item->get_permalink(), array('isPermaLink'=>'true'));\r
- }\r
- // filter xss?\r
- if ($xss_filter) {\r
- debug('Filtering HTML to remove XSS');\r
- $html = htmLawed::hl($html, array('safe'=>1, 'deny_attribute'=>'style', 'comment'=>1, 'cdata'=>1));\r
- }\r
- $newitem->setDescription($html);\r
- \r
- // set date\r
- if ((int)$item->get_date('U') > 0) {\r
- $newitem->setDate((int)$item->get_date('U'));\r
- } elseif ($extractor->getDate()) {\r
- $newitem->setDate($extractor->getDate());\r
- }\r
- \r
- // add authors\r
- if ($authors = $item->get_authors()) {\r
- foreach ($authors as $author) {\r
- // for some feeds, SimplePie stores author's name as email, e.g. http://feeds.feedburner.com/nymag/intel\r
- if ($author->get_name() !== null) {\r
- $newitem->addElement('dc:creator', $author->get_name());\r
- } elseif ($author->get_email() !== null) {\r
- $newitem->addElement('dc:creator', $author->get_email());\r
+ // Try to get first few paragraphs\r
+ if (isset($content_block) && ($content_block instanceof DOMElement)) {\r
+ $_paras = $content_block->getElementsByTagName('p');\r
+ foreach ($_paras as $_para) {\r
+ $summary .= preg_replace("/[\n\r\t ]+/", ' ', $_para->textContent).' ';\r
+ if (strlen($summary) > 200) break;\r
}\r
+ } else {\r
+ $summary = $html;\r
}\r
- } elseif ($authors = $extractor->getAuthors()) {\r
- //TODO: make sure the list size is reasonable\r
- foreach ($authors as $author) {\r
- // TODO: xpath often selects authors from other articles linked from the page.\r
- // for now choose first item\r
- $newitem->addElement('dc:creator', $author);\r
- break;\r
+ }\r
+ unset($_paras, $_para);\r
+ $summary = get_excerpt($summary);\r
+ $newitem->setDescription($summary);\r
+ if ($options->content) $newitem->setElement('content:encoded', $html);\r
+ } else {\r
+ if ($options->content) $newitem->setDescription($html);\r
+ }\r
+\r
+ // set date\r
+ if ((int)$item->get_date('U') > 0) {\r
+ $newitem->setDate((int)$item->get_date('U'));\r
+ } elseif ($extractor->getDate()) {\r
+ $newitem->setDate($extractor->getDate());\r
+ }\r
+\r
+ // add authors\r
+ if ($authors = $item->get_authors()) {\r
+ foreach ($authors as $author) {\r
+ // for some feeds, SimplePie stores author's name as email, e.g. http://feeds.feedburner.com/nymag/intel\r
+ if ($author->get_name() !== null) {\r
+ $newitem->addElement('dc:creator', $author->get_name());\r
+ } elseif ($author->get_email() !== null) {\r
+ $newitem->addElement('dc:creator', $author->get_email());\r
}\r
}\r
- \r
- // add language\r
- if ($detect_language) {\r
- $language = $extractor->getLanguage();\r
- if (!$language) $language = $feed->get_language();\r
- if (($detect_language == 3 || (!$language && $detect_language == 2)) && $text_sample) {\r
- try {\r
- if ($use_cld) {\r
- // Use PHP-CLD extension\r
- $php_cld = 'CLD\detect'; // in quotes to prevent PHP 5.2 parse error\r
- $res = $php_cld($text_sample);\r
- if (is_array($res) && count($res) > 0) {\r
- $language = $res[0]['code'];\r
- } \r
- } else {\r
- //die('what');\r
- // Use PEAR's Text_LanguageDetect\r
- if (!isset($l)) {\r
- $l = new Text_LanguageDetect('libraries/language-detect/lang.dat', 'libraries/language-detect/unicode_blocks.dat');\r
- }\r
- $l_result = $l->detect($text_sample, 1);\r
- if (count($l_result) > 0) {\r
- $language = $language_codes[key($l_result)];\r
- }\r
+ } elseif ($authors = $extractor->getAuthors()) {\r
+ //TODO: make sure the list size is reasonable\r
+ foreach ($authors as $author) {\r
+ // TODO: xpath often selects authors from other articles linked from the page.\r
+ // for now choose first item\r
+ $newitem->addElement('dc:creator', $author);\r
+ break;\r
+ }\r
+ }\r
+\r
+ // add language\r
+ if ($detect_language) {\r
+ $language = $extractor->getLanguage();\r
+ if (!$language) $language = $feed->get_language();\r
+ if (($detect_language == 3 || (!$language && $detect_language == 2)) && $text_sample) {\r
+ try {\r
+ if ($use_cld) {\r
+ // Use PHP-CLD extension\r
+ $php_cld = 'CLD\detect'; // in quotes to prevent PHP 5.2 parse error\r
+ $res = $php_cld($text_sample);\r
+ if (is_array($res) && count($res) > 0) {\r
+ $language = $res[0]['code'];\r
+ }\r
+ } else {\r
+ //die('what');\r
+ // Use PEAR's Text_LanguageDetect\r
+ if (!isset($l)) {\r
+ $l = new Text_LanguageDetect();\r
+ $l->setNameMode(2); // return ISO 639-1 codes (e.g. "en")\r
+ }\r
+ $l_result = $l->detect($text_sample, 1);\r
+ if (count($l_result) > 0) {\r
+ $language = key($l_result);\r
}\r
- } catch (Exception $e) {\r
- //die('error: '.$e); \r
- // do nothing\r
}\r
- }\r
- if ($language && (strlen($language) < 7)) { \r
- $newitem->addElement('dc:language', $language);\r
+ } catch (Exception $e) {\r
+ //die('error: '.$e);\r
+ // do nothing\r
}\r
}\r
- \r
- // add MIME type (if it appeared in our exclusions lists)\r
- if (isset($mime_info['mime'])) $newitem->addElement('dc:format', $mime_info['mime']);\r
- // add effective URL (URL after redirects)\r
- if (isset($effective_url)) {\r
- //TODO: ensure $effective_url is valid witout - sometimes it causes problems, e.g.\r
- //http://www.siasat.pk/forum/showthread.php?108883-Pakistan-Chowk-by-Rana-Mubashir-�-25th-March-2012-Special-Program-from-Liari-(Karachi)\r
- //temporary measure: use utf8_encode()\r
- $newitem->addElement('dc:identifier', remove_url_cruft(utf8_encode($effective_url)));\r
- } else {\r
- $newitem->addElement('dc:identifier', remove_url_cruft($item->get_permalink()));\r
+ if ($language && (strlen($language) < 7)) {\r
+ $newitem->addElement('dc:language', $language);\r
}\r
- \r
- // add categories\r
- if ($categories = $item->get_categories()) {\r
- foreach ($categories as $category) {\r
- if ($category->get_label() !== null) {\r
- $newitem->addElement('category', $category->get_label());\r
- }\r
+ }\r
+\r
+ // add MIME type (if it appeared in our exclusions lists)\r
+ if (isset($mime_info['mime'])) $newitem->addElement('dc:format', $mime_info['mime']);\r
+ // add effective URL (URL after redirects)\r
+ if (isset($effective_url)) {\r
+ //TODO: ensure $effective_url is valid witout - sometimes it causes problems, e.g.\r
+ //http://www.siasat.pk/forum/showthread.php?108883-Pakistan-Chowk-by-Rana-Mubashir-�-25th-March-2012-Special-Program-from-Liari-(Karachi)\r
+ //temporary measure: use utf8_encode()\r
+ $newitem->addElement('dc:identifier', remove_url_cruft(utf8_encode($effective_url)));\r
+ } else {\r
+ $newitem->addElement('dc:identifier', remove_url_cruft($item->get_permalink()));\r
+ }\r
+\r
+ // add categories\r
+ if ($categories = $item->get_categories()) {\r
+ foreach ($categories as $category) {\r
+ if ($category->get_label() !== null) {\r
+ $newitem->addElement('category', $category->get_label());\r
}\r
}\r
- \r
- // check for enclosures\r
- if ($options->keep_enclosures) {\r
- if ($enclosures = $item->get_enclosures()) {\r
- foreach ($enclosures as $enclosure) {\r
- // thumbnails\r
- foreach ((array)$enclosure->get_thumbnails() as $thumbnail) {\r
- $newitem->addElement('media:thumbnail', '', array('url'=>$thumbnail));\r
- }\r
- if (!$enclosure->get_link()) continue;\r
- $enc = array();\r
- // Media RSS spec ($enc): http://search.yahoo.com/mrss\r
- // SimplePie methods ($enclosure): http://simplepie.org/wiki/reference/start#methods4\r
- $enc['url'] = $enclosure->get_link();\r
- if ($enclosure->get_length()) $enc['fileSize'] = $enclosure->get_length();\r
- if ($enclosure->get_type()) $enc['type'] = $enclosure->get_type();\r
- if ($enclosure->get_medium()) $enc['medium'] = $enclosure->get_medium();\r
- if ($enclosure->get_expression()) $enc['expression'] = $enclosure->get_expression();\r
- if ($enclosure->get_bitrate()) $enc['bitrate'] = $enclosure->get_bitrate();\r
- if ($enclosure->get_framerate()) $enc['framerate'] = $enclosure->get_framerate();\r
- if ($enclosure->get_sampling_rate()) $enc['samplingrate'] = $enclosure->get_sampling_rate();\r
- if ($enclosure->get_channels()) $enc['channels'] = $enclosure->get_channels();\r
- if ($enclosure->get_duration()) $enc['duration'] = $enclosure->get_duration();\r
- if ($enclosure->get_height()) $enc['height'] = $enclosure->get_height();\r
- if ($enclosure->get_width()) $enc['width'] = $enclosure->get_width();\r
- if ($enclosure->get_language()) $enc['lang'] = $enclosure->get_language();\r
- $newitem->addElement('media:content', '', $enc);\r
+ }\r
+\r
+ // check for enclosures\r
+ if ($options->keep_enclosures) {\r
+ if ($enclosures = $item->get_enclosures()) {\r
+ foreach ($enclosures as $enclosure) {\r
+ // thumbnails\r
+ foreach ((array)$enclosure->get_thumbnails() as $thumbnail) {\r
+ $newitem->addElement('media:thumbnail', '', array('url'=>$thumbnail));\r
}\r
+ if (!$enclosure->get_link()) continue;\r
+ $enc = array();\r
+ // Media RSS spec ($enc): http://search.yahoo.com/mrss\r
+ // SimplePie methods ($enclosure): http://simplepie.org/wiki/reference/start#methods4\r
+ $enc['url'] = $enclosure->get_link();\r
+ if ($enclosure->get_length()) $enc['fileSize'] = $enclosure->get_length();\r
+ if ($enclosure->get_type()) $enc['type'] = $enclosure->get_type();\r
+ if ($enclosure->get_medium()) $enc['medium'] = $enclosure->get_medium();\r
+ if ($enclosure->get_expression()) $enc['expression'] = $enclosure->get_expression();\r
+ if ($enclosure->get_bitrate()) $enc['bitrate'] = $enclosure->get_bitrate();\r
+ if ($enclosure->get_framerate()) $enc['framerate'] = $enclosure->get_framerate();\r
+ if ($enclosure->get_sampling_rate()) $enc['samplingrate'] = $enclosure->get_sampling_rate();\r
+ if ($enclosure->get_channels()) $enc['channels'] = $enclosure->get_channels();\r
+ if ($enclosure->get_duration()) $enc['duration'] = $enclosure->get_duration();\r
+ if ($enclosure->get_height()) $enc['height'] = $enclosure->get_height();\r
+ if ($enclosure->get_width()) $enc['width'] = $enclosure->get_width();\r
+ if ($enclosure->get_language()) $enc['lang'] = $enclosure->get_language();\r
+ $newitem->addElement('media:content', '', $enc);\r
}\r
}\r
- /* } */\r
+ }\r
$output->addItem($newitem);\r
unset($html);\r
$item_count++;\r
projects / github / wallabag / wallabag.git / blobdiff