diff --git a/CREDITS.md b/CREDITS.md index a6dedce4b..6046a6aac 100644 --- a/CREDITS.md +++ b/CREDITS.md @@ -1,5 +1,6 @@ poche is based on : * PHP Readability https://bitbucket.org/fivefilters/php-readability +* Full Text RSS http://code.fivefilters.org/full-text-rss/src * Encoding https://github.com/neitanod/forceutf8 * logo by Brightmix http://www.iconfinder.com/icondetails/43256/128/jeans_monotone_pocket_icon * icons http://icomoon.io diff --git a/inc/3rdparty/content-extractor/ContentExtractor.php b/inc/3rdparty/content-extractor/ContentExtractor.php new file mode 100644 index 000000000..db371c6a3 --- /dev/null +++ b/inc/3rdparty/content-extractor/ContentExtractor.php @@ -0,0 +1,612 @@ + true, + 'output-xhtml' => true, + 'logical-emphasis' => true, + 'show-body-only' => false, + 'new-blocklevel-tags' => 'article, aside, footer, header, hgroup, menu, nav, section, details, datagrid', + 'new-inline-tags' => 'mark, time, meter, progress, data', + 'wrap' => 0, + 'drop-empty-paras' => true, + 'drop-proprietary-attributes' => false, + 'enclose-text' => true, + 'enclose-block-text' => true, + 'merge-divs' => true, + 'merge-spans' => true, + 'char-encoding' => 'utf8', + 'hide-comments' => true + ); + protected $html; + protected $config; + protected $title; + protected $author = array(); + protected $language; + protected $date; + protected $body; + protected $success = false; + public $fingerprints = array(); + public $readability; + public $debug = false; + + function __construct($path, $fallback=null) { + SiteConfig::set_config_path($path, $fallback); + } + + protected function debug($msg) { + if ($this->debug) { + $mem = round(memory_get_usage()/1024, 2); + $memPeak = round(memory_get_peak_usage()/1024, 2); + echo '* ',$msg; + echo ' - mem used: ',$mem," (peak: $memPeak)\n"; + ob_flush(); + flush(); + } + } + + public function reset() { + $this->html = null; + $this->readability = null; + $this->config = null; + $this->title = null; + $this->body = null; + $this->author = array(); + $this->language = null; + $this->date = null; + $this->success = false; + } + + public function findHostUsingFingerprints($html) { + $this->debug('Checking fingerprints...'); + $head = substr($html, 0, 8000); + foreach ($this->fingerprints as $_fp => $_fphost) { + $lookin = 'html'; + if (is_array($_fphost)) { + if (isset($_fphost['head']) && $_fphost['head']) { + $lookin = 'head'; + } + $_fphost = $_fphost['hostname']; + } + if (strpos($$lookin, $_fp) !== false) { + $this->debug("Found match: $_fphost"); + return $_fphost; + } + } + return false; + } + + // returns true on success, false on failure + // $smart_tidy indicates that if tidy is used and no results are produced, we will + // try again without it. Tidy helps us deal with PHP's patchy HTML parsing most of the time + // but it has problems of its own which we try to avoid with this option. + public function process($html, $url, $smart_tidy=true) { + $this->reset(); + // extract host name + $host = @parse_url($url, PHP_URL_HOST); + if (!($this->config = SiteConfig::build($host))) { + // no match, check HTML for fingerprints + if (!empty($this->fingerprints) && ($_fphost = $this->findHostUsingFingerprints($html))) { + $this->config = SiteConfig::build($_fphost); + } + unset($_fphost); + if (!$this->config) { + // no match, so use defaults + $this->config = new SiteConfig(); + } + } + // store copy of config in our static cache array in case we need to process another URL + SiteConfig::add_to_cache($host, $this->config); + + // do string replacements + foreach ($this->config->replace_string as $_repl) { + $html = str_replace($_repl[0], $_repl[1], $html); + } + unset($_repl); + + // use tidy (if it exists)? + // This fixes problems with some sites which would otherwise + // trouble DOMDocument's HTML parsing. (Although sometimes it + // makes matters worse, which is why you can override it in site config files.) + $tidied = false; + if ($this->config->tidy && function_exists('tidy_parse_string') && $smart_tidy) { + $this->debug('Using Tidy'); + $tidy = tidy_parse_string($html, self::$tidy_config, 'UTF8'); + if (tidy_clean_repair($tidy)) { + $original_html = $html; + $tidied = true; + // $html = $tidy->value; + } + $body = $tidy->body(); + if (preg_replace('/\s+/', '', $body->value) !== "
") { + $html = $tidy->value; + } + unset($tidy); + } + + // load and parse html + $this->readability = new Readability($html, $url); + + // we use xpath to find elements in the given HTML document + // see http://en.wikipedia.org/wiki/XPath_1.0 + $xpath = new DOMXPath($this->readability->dom); + + // try to get title + foreach ($this->config->title as $pattern) { + $elems = @$xpath->evaluate($pattern, $this->readability->dom); + if (is_string($elems)) { + $this->debug('Title expression evaluated as string'); + $this->title = trim($elems); + break; + } elseif ($elems instanceof DOMNodeList && $elems->length > 0) { + $this->debug('Title matched'); + $this->title = $elems->item(0)->textContent; + // remove title from document + try { + $elems->item(0)->parentNode->removeChild($elems->item(0)); + } catch (DOMException $e) { + // do nothing + } + break; + } + } + + // try to get author (if it hasn't already been set) + if (empty($this->author)) { + foreach ($this->config->author as $pattern) { + $elems = @$xpath->evaluate($pattern, $this->readability->dom); + if (is_string($elems)) { + $this->debug('Author expression evaluated as string'); + if (trim($elems) != '') { + $this->author[] = trim($elems); + break; + } + } elseif ($elems instanceof DOMNodeList && $elems->length > 0) { + foreach ($elems as $elem) { + if (!isset($elem->parentNode)) continue; + $this->author[] = trim($elem->textContent); + } + if (!empty($this->author)) break; + } + } + } + + // try to get language + $_lang_xpath = array('//html[@lang]/@lang', '//meta[@name="DC.language"]/@content'); + foreach ($_lang_xpath as $pattern) { + $elems = @$xpath->evaluate($pattern, $this->readability->dom); + if (is_string($elems)) { + if (trim($elems) != '') { + $this->language = trim($elems); + break; + } + } elseif ($elems instanceof DOMNodeList && $elems->length > 0) { + foreach ($elems as $elem) { + if (!isset($elem->parentNode)) continue; + $this->language = trim($elem->textContent); + } + if ($this->language) break; + } + } + + // try to get date + foreach ($this->config->date as $pattern) { + $elems = @$xpath->evaluate($pattern, $this->readability->dom); + if (is_string($elems)) { + $this->debug('Date expression evaluated as string'); + $this->date = strtotime(trim($elems, "; \t\n\r\0\x0B")); + } elseif ($elems instanceof DOMNodeList && $elems->length > 0) { + $this->debug('Date matched'); + $this->date = $elems->item(0)->textContent; + $this->date = strtotime(trim($this->date, "; \t\n\r\0\x0B")); + // remove date from document + // $elems->item(0)->parentNode->removeChild($elems->item(0)); + } + if (!$this->date) { + $this->date = null; + } else { + break; + } + } + + // strip elements (using xpath expressions) + foreach ($this->config->strip as $pattern) { + $elems = @$xpath->query($pattern, $this->readability->dom); + // check for matches + if ($elems && $elems->length > 0) { + $this->debug('Stripping '.$elems->length.' elements (strip)'); + for ($i=$elems->length-1; $i >= 0; $i--) { + $elems->item($i)->parentNode->removeChild($elems->item($i)); + } + } + } + + // strip elements (using id and class attribute values) + foreach ($this->config->strip_id_or_class as $string) { + $string = strtr($string, array("'"=>'', '"'=>'')); + $elems = @$xpath->query("//*[contains(@class, '$string') or contains(@id, '$string')]", $this->readability->dom); + // check for matches + if ($elems && $elems->length > 0) { + $this->debug('Stripping '.$elems->length.' elements (strip_id_or_class)'); + for ($i=$elems->length-1; $i >= 0; $i--) { + $elems->item($i)->parentNode->removeChild($elems->item($i)); + } + } + } + + // strip images (using src attribute values) + foreach ($this->config->strip_image_src as $string) { + $string = strtr($string, array("'"=>'', '"'=>'')); + $elems = @$xpath->query("//img[contains(@src, '$string')]", $this->readability->dom); + // check for matches + if ($elems && $elems->length > 0) { + $this->debug('Stripping '.$elems->length.' image elements'); + for ($i=$elems->length-1; $i >= 0; $i--) { + $elems->item($i)->parentNode->removeChild($elems->item($i)); + } + } + } + // strip elements using Readability.com and Instapaper.com ignore class names + // .entry-unrelated and .instapaper_ignore + // See https://www.readability.com/publishers/guidelines/#view-plainGuidelines + // and http://blog.instapaper.com/post/730281947 + $elems = @$xpath->query("//*[contains(concat(' ',normalize-space(@class),' '),' entry-unrelated ') or contains(concat(' ',normalize-space(@class),' '),' instapaper_ignore ')]", $this->readability->dom); + // check for matches + if ($elems && $elems->length > 0) { + $this->debug('Stripping '.$elems->length.' .entry-unrelated,.instapaper_ignore elements'); + for ($i=$elems->length-1; $i >= 0; $i--) { + $elems->item($i)->parentNode->removeChild($elems->item($i)); + } + } + + // strip elements that contain style="display: none;" + $elems = @$xpath->query("//*[contains(@style,'display:none')]", $this->readability->dom); + // check for matches + if ($elems && $elems->length > 0) { + $this->debug('Stripping '.$elems->length.' elements with inline display:none style'); + for ($i=$elems->length-1; $i >= 0; $i--) { + $elems->item($i)->parentNode->removeChild($elems->item($i)); + } + } + + // try to get body + foreach ($this->config->body as $pattern) { + $elems = @$xpath->query($pattern, $this->readability->dom); + // check for matches + if ($elems && $elems->length > 0) { + $this->debug('Body matched'); + if ($elems->length == 1) { + $this->body = $elems->item(0); + // prune (clean up elements that may not be content) + if ($this->config->prune) { + $this->debug('Pruning content'); + $this->readability->prepArticle($this->body); + } + break; + } else { + $this->body = $this->readability->dom->createElement('div'); + $this->debug($elems->length.' body elems found'); + foreach ($elems as $elem) { + if (!isset($elem->parentNode)) continue; + $isDescendant = false; + foreach ($this->body->childNodes as $parent) { + if ($this->isDescendant($parent, $elem)) { + $isDescendant = true; + break; + } + } + if ($isDescendant) { + $this->debug('Element is child of another body element, skipping.'); + } else { + // prune (clean up elements that may not be content) + if ($this->config->prune) { + $this->debug('Pruning content'); + $this->readability->prepArticle($elem); + } + $this->debug('Element added to body'); + $this->body->appendChild($elem); + } + } + } + } + } + + // auto detect? + $detect_title = $detect_body = $detect_author = $detect_date = false; + // detect title? + if (!isset($this->title)) { + if (empty($this->config->title) || $this->config->autodetect_on_failure) { + $detect_title = true; + } + } + // detect body? + if (!isset($this->body)) { + if (empty($this->config->body) || $this->config->autodetect_on_failure) { + $detect_body = true; + } + } + // detect author? + if (empty($this->author)) { + if (empty($this->config->author) || $this->config->autodetect_on_failure) { + $detect_author = true; + } + } + // detect date? + if (!isset($this->date)) { + if (empty($this->config->date) || $this->config->autodetect_on_failure) { + $detect_date = true; + } + } + + // check for hNews + if ($detect_title || $detect_body) { + // check for hentry + $elems = @$xpath->query("//*[contains(concat(' ',normalize-space(@class),' '),' hentry ')]", $this->readability->dom); + if ($elems && $elems->length > 0) { + $this->debug('hNews: found hentry'); + $hentry = $elems->item(0); + + if ($detect_title) { + // check for entry-title + $elems = @$xpath->query(".//*[contains(concat(' ',normalize-space(@class),' '),' entry-title ')]", $hentry); + if ($elems && $elems->length > 0) { + $this->debug('hNews: found entry-title'); + $this->title = $elems->item(0)->textContent; + // remove title from document + $elems->item(0)->parentNode->removeChild($elems->item(0)); + $detect_title = false; + } + } + + if ($detect_date) { + // check for time element with pubdate attribute + $elems = @$xpath->query(".//time[@pubdate] | .//abbr[contains(concat(' ',normalize-space(@class),' '),' published ')]", $hentry); + if ($elems && $elems->length > 0) { + $this->debug('hNews: found publication date'); + $this->date = strtotime(trim($elems->item(0)->textContent)); + // remove date from document + //$elems->item(0)->parentNode->removeChild($elems->item(0)); + if ($this->date) { + $detect_date = false; + } else { + $this->date = null; + } + } + } + + if ($detect_author) { + // check for time element with pubdate attribute + $elems = @$xpath->query(".//*[contains(concat(' ',normalize-space(@class),' '),' vcard ') and (contains(concat(' ',normalize-space(@class),' '),' author ') or contains(concat(' ',normalize-space(@class),' '),' byline '))]", $hentry); + if ($elems && $elems->length > 0) { + $this->debug('hNews: found author'); + $author = $elems->item(0); + $fn = @$xpath->query(".//*[contains(concat(' ',normalize-space(@class),' '),' fn ')]", $author); + if ($fn && $fn->length > 0) { + foreach ($fn as $_fn) { + if (trim($_fn->textContent) != '') { + $this->author[] = trim($_fn->textContent); + } + } + } else { + if (trim($author->textContent) != '') { + $this->author[] = trim($author->textContent); + } + } + $detect_author = empty($this->author); + } + } + + // check for entry-content. + // according to hAtom spec, if there are multiple elements marked entry-content, + // we include all of these in the order they appear - see http://microformats.org/wiki/hatom#Entry_Content + if ($detect_body) { + $elems = @$xpath->query(".//*[contains(concat(' ',normalize-space(@class),' '),' entry-content ')]", $hentry); + if ($elems && $elems->length > 0) { + $this->debug('hNews: found entry-content'); + if ($elems->length == 1) { + // what if it's empty? (some sites misuse hNews - place their content outside an empty entry-content element) + $e = $elems->item(0); + if (($e->tagName == 'img') || (trim($e->textContent) != '')) { + $this->body = $elems->item(0); + // prune (clean up elements that may not be content) + if ($this->config->prune) { + $this->debug('Pruning content'); + $this->readability->prepArticle($this->body); + } + $detect_body = false; + } else { + $this->debug('hNews: skipping entry-content - appears not to contain content'); + } + unset($e); + } else { + $this->body = $this->readability->dom->createElement('div'); + $this->debug($elems->length.' entry-content elems found'); + foreach ($elems as $elem) { + if (!isset($elem->parentNode)) continue; + $isDescendant = false; + foreach ($this->body->childNodes as $parent) { + if ($this->isDescendant($parent, $elem)) { + $isDescendant = true; + break; + } + } + if ($isDescendant) { + $this->debug('Element is child of another body element, skipping.'); + } else { + // prune (clean up elements that may not be content) + if ($this->config->prune) { + $this->debug('Pruning content'); + $this->readability->prepArticle($elem); + } + $this->debug('Element added to body'); + $this->body->appendChild($elem); + } + } + $detect_body = false; + } + } + } + } + } + + // check for elements marked with instapaper_title + if ($detect_title) { + // check for instapaper_title + $elems = @$xpath->query("//*[contains(concat(' ',normalize-space(@class),' '),' instapaper_title ')]", $this->readability->dom); + if ($elems && $elems->length > 0) { + $this->debug('title found (.instapaper_title)'); + $this->title = $elems->item(0)->textContent; + // remove title from document + $elems->item(0)->parentNode->removeChild($elems->item(0)); + $detect_title = false; + } + } + // check for elements marked with instapaper_body + if ($detect_body) { + $elems = @$xpath->query("//*[contains(concat(' ',normalize-space(@class),' '),' instapaper_body ')]", $this->readability->dom); + if ($elems && $elems->length > 0) { + $this->debug('body found (.instapaper_body)'); + $this->body = $elems->item(0); + // prune (clean up elements that may not be content) + if ($this->config->prune) { + $this->debug('Pruning content'); + $this->readability->prepArticle($this->body); + } + $detect_body = false; + } + } + + // Find author in rel="author" marked element + // We only use this if there's exactly one. + // If there's more than one, it could indicate more than + // one author, but it could also indicate that we're processing + // a page listing different articles with different authors. + if ($detect_author) { + $elems = @$xpath->query("//a[contains(concat(' ',normalize-space(@rel),' '),' author ')]", $this->readability->dom); + if ($elems && $elems->length == 1) { + $this->debug('Author found (rel="author")'); + $author = trim($elems->item(0)->textContent); + if ($author != '') { + $this->author[] = $author; + $detect_author = false; + } + } + } + + // Find date in pubdate marked time element + // For the same reason given above, we only use this + // if there's exactly one element. + if ($detect_date) { + $elems = @$xpath->query("//time[@pubdate]", $this->readability->dom); + if ($elems && $elems->length == 1) { + $this->debug('Date found (pubdate marked time element)'); + $this->date = strtotime(trim($elems->item(0)->textContent)); + // remove date from document + //$elems->item(0)->parentNode->removeChild($elems->item(0)); + if ($this->date) { + $detect_date = false; + } else { + $this->date = null; + } + } + } + + // still missing title or body, so we detect using Readability + if ($detect_title || $detect_body) { + $this->debug('Using Readability'); + // clone body if we're only using Readability for title (otherwise it may interfere with body element) + if (isset($this->body)) $this->body = $this->body->cloneNode(true); + $success = $this->readability->init(); + } + if ($detect_title) { + $this->debug('Detecting title'); + $this->title = $this->readability->getTitle()->textContent; + } + if ($detect_body && $success) { + $this->debug('Detecting body'); + $this->body = $this->readability->getContent(); + if ($this->body->childNodes->length == 1 && $this->body->firstChild->nodeType === XML_ELEMENT_NODE) { + $this->body = $this->body->firstChild; + } + // prune (clean up elements that may not be content) + if ($this->config->prune) { + $this->debug('Pruning content'); + $this->readability->prepArticle($this->body); + } + } + if (isset($this->body)) { + // remove scripts + $this->readability->removeScripts($this->body); + // remove any h1-h6 elements that appear as first thing in the body + // and which match our title + if (isset($this->title) && ($this->title != '')) { + $firstChild = $this->body->firstChild; + while ($firstChild->nodeType && ($firstChild->nodeType !== XML_ELEMENT_NODE)) { + $firstChild = $firstChild->nextSibling; + } + if (($firstChild->nodeType === XML_ELEMENT_NODE) + && in_array(strtolower($firstChild->tagName), array('h1', 'h2', 'h3', 'h4', 'h5', 'h6')) + && (strtolower(trim($firstChild->textContent)) == strtolower(trim($this->title)))) { + $this->body->removeChild($firstChild); + } + } + $this->success = true; + } + + // if we've had no success and we've used tidy, there's a chance + // that tidy has messed up. So let's try again without tidy... + if (!$this->success && $tidied && $smart_tidy) { + $this->debug('Trying again without tidy'); + $this->process($original_html, $url, false); + } + + return $this->success; + } + + private function isDescendant(DOMElement $parent, DOMElement $child) { + $node = $child->parentNode; + while ($node != null) { + if ($node->isSameNode($parent)) return true; + $node = $node->parentNode; + } + return false; + } + + public function getContent() { + return $this->body; + } + + public function getTitle() { + return $this->title; + } + + public function getAuthors() { + return $this->author; + } + + public function getLanguage() { + return $this->language; + } + + public function getDate() { + return $this->date; + } + + public function getSiteConfig() { + return $this->config; + } +} +?> \ No newline at end of file diff --git a/inc/3rdparty/content-extractor/SiteConfig.php b/inc/3rdparty/content-extractor/SiteConfig.php new file mode 100644 index 000000000..089e10c62 --- /dev/null +++ b/inc/3rdparty/content-extractor/SiteConfig.php @@ -0,0 +1,184 @@ + 200) || !preg_match(self::HOSTNAME_REGEX, $host)) return false; + // check for site configuration + $try = array($host); + $split = explode('.', $host); + if (count($split) > 1) { + array_shift($split); + $try[] = '.'.implode('.', $split); + } + foreach ($try as $h) { + if (array_key_exists($h, self::$config_cache)) { + self::debug("... cached ($h)"); + return self::$config_cache[$h]; + } elseif (file_exists(self::$config_path."/$h.txt")) { + self::debug("... from file ($h)"); + $file = self::$config_path."/$h.txt"; + break; + } + } + if (!isset($file)) { + if (isset(self::$config_path_fallback)) { + self::debug("... trying fallback ($host)"); + foreach ($try as $h) { + if (file_exists(self::$config_path_fallback."/$h.txt")) { + self::debug("... from fallback file ($h)"); + $file = self::$config_path_fallback."/$h.txt"; + break; + } + } + if (!isset($file)) { + self::debug("... no match in fallback directory"); + return false; + } + } else { + self::debug("... no match ($host)"); + return false; + } + } + $config_file = file($file, FILE_IGNORE_NEW_LINES | FILE_SKIP_EMPTY_LINES); + if (!$config_file || !is_array($config_file)) return false; + $config = new SiteConfig(); + foreach ($config_file as $line) { + $line = trim($line); + + // skip comments, empty lines + if ($line == '' || $line[0] == '#') continue; + + // get command + $command = explode(':', $line, 2); + // if there's no colon ':', skip this line + if (count($command) != 2) continue; + $val = trim($command[1]); + $command = trim($command[0]); + if ($command == '' || $val == '') continue; + + // check for commands where we accept multiple statements + if (in_array($command, array('title', 'body', 'author', 'date', 'strip', 'strip_id_or_class', 'strip_image_src', 'single_page_link', 'single_page_link_in_feed', 'http_header'))) { + array_push($config->$command, $val); + // check for single statement commands that evaluate to true or false + } elseif (in_array($command, array('tidy', 'prune', 'autodetect_on_failure'))) { + $config->$command = ($val == 'yes'); + // check for single statement commands stored as strings + } elseif (in_array($command, array('test_url', 'parser'))) { + $config->$command = $val; + } elseif ((substr($command, -1) == ')') && preg_match('!^([a-z0-9_]+)\((.*?)\)$!i', $command, $match)) { + if (in_array($match[1], array('replace_string'))) { + $command = $match[1]; + array_push($config->$command, array($match[2], $val)); + } + } + } + return $config; + } +} +?> \ No newline at end of file diff --git a/inc/3rdparty/feedwriter/DummySingleItemFeed.php b/inc/3rdparty/feedwriter/DummySingleItemFeed.php new file mode 100644 index 000000000..5f2f871e4 --- /dev/null +++ b/inc/3rdparty/feedwriter/DummySingleItemFeed.php @@ -0,0 +1,24 @@ +item = new DummySingleItem($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 $this->url; } + public function get_title() { return ''; } + public function get_date($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; } +} \ No newline at end of file diff --git a/inc/3rdparty/feedwriter/FeedItem.php b/inc/3rdparty/feedwriter/FeedItem.php new file mode 100644 index 000000000..71e6e98ce --- /dev/null +++ b/inc/3rdparty/feedwriter/FeedItem.php @@ -0,0 +1,167 @@ + + * @link http://www.ajaxray.com/projects/rss + */ + class FeedItem + { + private $elements = array(); //Collection of feed elements + private $version; + + /** + * Constructor + * + * @param contant (RSS1/RSS2/ATOM) RSS2 is default. + */ + function __construct($version = RSS2) + { + $this->version = $version; + } + + /** + * Add an element to elements array + * + * @access public + * @param srting The tag name of an element + * @param srting The content of tag + * @param array Attributes(if any) in 'attrName' => 'attrValue' format + * @return void + */ + public function addElement($elementName, $content, $attributes = null) + { + $this->elements[$elementName]['name'] = $elementName; + $this->elements[$elementName]['content'] = $content; + $this->elements[$elementName]['attributes'] = $attributes; + } + + /** + * Set multiple feed elements from an array. + * Elements which have attributes cannot be added by this method + * + * @access public + * @param array array of elements in 'tagName' => 'tagContent' format. + * @return void + */ + public function addElementArray($elementArray) + { + if(! is_array($elementArray)) return; + foreach ($elementArray as $elementName => $content) + { + $this->addElement($elementName, $content); + } + } + + /** + * Return the collection of elements in this feed item + * + * @access public + * @return array + */ + public function getElements() + { + return $this->elements; + } + + // Wrapper functions ------------------------------------------------------ + + /** + * Set the 'dscription' element of feed item + * + * @access public + * @param string The content of 'description' element + * @return void + */ + public function setDescription($description) + { + $tag = ($this->version == ATOM)? 'summary' : 'description'; + $this->addElement($tag, $description); + } + + /** + * @desc Set the 'title' element of feed item + * @access public + * @param string The content of 'title' element + * @return void + */ + public function setTitle($title) + { + $this->addElement('title', $title); + } + + /** + * Set the 'date' element of feed item + * + * @access public + * @param string The content of 'date' element + * @return void + */ + public function setDate($date) + { + if(! is_numeric($date)) + { + $date = strtotime($date); + } + + if($this->version == ATOM) + { + $tag = 'updated'; + $value = date(DATE_ATOM, $date); + } + elseif($this->version == RSS2) + { + $tag = 'pubDate'; + $value = date(DATE_RSS, $date); + } + else + { + $tag = 'dc:date'; + $value = date("Y-m-d", $date); + } + + $this->addElement($tag, $value); + } + + /** + * Set the 'link' element of feed item + * + * @access public + * @param string The content of 'link' element + * @return void + */ + public function setLink($link) + { + if($this->version == RSS2 || $this->version == RSS1) + { + $this->addElement('link', $link); + } + else + { + $this->addElement('link','',array('href'=>$link)); + $this->addElement('id', FeedWriter::uuid($link,'urn:uuid:')); + } + + } + + /** + * Set the 'encloser' element of feed item + * For RSS 2.0 only + * + * @access public + * @param string The url attribute of encloser tag + * @param string The length attribute of encloser tag + * @param string The type attribute of encloser tag + * @return void + */ + public function setEncloser($url, $length, $type) + { + $attributes = array('url'=>$url, 'length'=>$length, 'type'=>$type); + $this->addElement('enclosure','',$attributes); + } + + } // end of class FeedItem +?> diff --git a/inc/3rdparty/feedwriter/FeedWriter.php b/inc/3rdparty/feedwriter/FeedWriter.php new file mode 100644 index 000000000..d5d6648a9 --- /dev/null +++ b/inc/3rdparty/feedwriter/FeedWriter.php @@ -0,0 +1,434 @@ + + * @link http://www.ajaxray.com/projects/rss + */ + class FeedWriter + { + private $self = null; // self URL - http://feed2.w3.org/docs/warning/MissingAtomSelfLink.html + private $hubs = array(); // PubSubHubbub hubs + private $channels = array(); // Collection of channel elements + private $items = array(); // Collection of items as object of FeedItem class. + private $data = array(); // Store some other version wise data + private $CDATAEncoding = array(); // The tag names which have to encoded as CDATA + private $xsl = null; // stylesheet to render RSS (used by Chrome) + private $json = null; // JSON object + + private $version = null; + + /** + * Constructor + * + * @param constant the version constant (RSS2 or JSON). + */ + function __construct($version = RSS2) + { + $this->version = $version; + + // Setting default value for assential channel elements + $this->channels['title'] = $version . ' Feed'; + $this->channels['link'] = 'http://www.ajaxray.com/blog'; + + //Tag names to encode in CDATA + $this->CDATAEncoding = array('description', 'content:encoded', 'content', 'subtitle', 'summary'); + } + + public function setFormat($format) { + $this->version = $format; + } + + // Start # public functions --------------------------------------------- + + /** + * Set a channel element + * @access public + * @param srting name of the channel tag + * @param string content of the channel tag + * @return void + */ + public function setChannelElement($elementName, $content) + { + $this->channels[$elementName] = $content ; + } + + /** + * Set multiple channel elements from an array. Array elements + * should be 'channelName' => 'channelContent' format. + * + * @access public + * @param array array of channels + * @return void + */ + public function setChannelElementsFromArray($elementArray) + { + if(! is_array($elementArray)) return; + foreach ($elementArray as $elementName => $content) + { + $this->setChannelElement($elementName, $content); + } + } + + /** + * Genarate the actual RSS/JSON file + * + * @access public + * @return void + */ + public function genarateFeed() + { + if ($this->version == RSS2) { + header('Content-type: text/xml; charset=UTF-8'); + } elseif ($this->version == JSON) { + header('Content-type: application/json; charset=UTF-8'); + $this->json = new stdClass(); + } + $this->printHead(); + $this->printChannels(); + $this->printItems(); + $this->printTale(); + if ($this->version == JSON) { + echo json_encode($this->json); + } + } + + /** + * Create a new FeedItem. + * + * @access public + * @return object instance of FeedItem class + */ + public function createNewItem() + { + $Item = new FeedItem($this->version); + return $Item; + } + + /** + * Add a FeedItem to the main class + * + * @access public + * @param object instance of FeedItem class + * @return void + */ + public function addItem($feedItem) + { + $this->items[] = $feedItem; + } + + // Wrapper functions ------------------------------------------------------------------- + + /** + * Set the 'title' channel element + * + * @access public + * @param srting value of 'title' channel tag + * @return void + */ + public function setTitle($title) + { + $this->setChannelElement('title', $title); + } + + /** + * Add a hub to the channel element + * + * @access public + * @param string URL + * @return void + */ + public function addHub($hub) + { + $this->hubs[] = $hub; + } + + /** + * Set XSL URL + * + * @access public + * @param string URL + * @return void + */ + public function setXsl($xsl) + { + $this->xsl = $xsl; + } + + /** + * Set self URL + * + * @access public + * @param string URL + * @return void + */ + public function setSelf($self) + { + $this->self = $self; + } + + /** + * Set the 'description' channel element + * + * @access public + * @param srting value of 'description' channel tag + * @return void + */ + public function setDescription($desciption) + { + $tag = ($this->version == ATOM)? 'subtitle' : 'description'; + $this->setChannelElement($tag, $desciption); + } + + /** + * Set the 'link' channel element + * + * @access public + * @param srting value of 'link' channel tag + * @return void + */ + public function setLink($link) + { + $this->setChannelElement('link', $link); + } + + /** + * Set the 'image' channel element + * + * @access public + * @param srting title of image + * @param srting link url of the imahe + * @param srting path url of the image + * @return void + */ + public function setImage($title, $link, $url) + { + $this->setChannelElement('image', array('title'=>$title, 'link'=>$link, 'url'=>$url)); + } + + // End # public functions ---------------------------------------------- + + // Start # private functions ---------------------------------------------- + + /** + * Prints the xml and rss namespace + * + * @access private + * @return void + */ + private function printHead() + { + if ($this->version == RSS2) + { + $out = ''."\n"; + if ($this->xsl) $out .= 'xsl).'"?>' . PHP_EOL; + $out .= '+ * Cookies are stored like this: + * [domain][path][name] = array + * where array is: + * 0 => value, 1 => secure, 2 => expires + *+ * @var array + * @access private + */ + public $cookies = array(); + public $debug = false; + + /** + * Constructor + */ + function __construct() { + } + + protected function debug($msg, $file=null, $line=null) { + if ($this->debug) { + $mem = round(memory_get_usage()/1024, 2); + $memPeak = round(memory_get_peak_usage()/1024, 2); + echo '* ',$msg; + if (isset($file, $line)) echo " ($file line $line)"; + echo ' - mem used: ',$mem," (peak: $memPeak)\n"; + ob_flush(); + flush(); + } + } + + /** + * Get matching cookies + * + * Only use this method if you cannot use add_cookie_header(), for example, if you want to use + * this cookie jar class without using the request class. + * + * @param array $param associative array containing 'domain', 'path', 'secure' keys + * @return string + * @see add_cookie_header() + */ + public function getMatchingCookies($url) + { + if (($parts = @parse_url($url)) && isset($parts['scheme'], $parts['host'], $parts['path'])) { + $param['domain'] = $parts['host']; + $param['path'] = $parts['path']; + $param['secure'] = (strtolower($parts['scheme']) == 'https'); + unset($parts); + } else { + return false; + } + // RFC 2965 notes: + // If multiple cookies satisfy the criteria above, they are ordered in + // the Cookie header such that those with more specific Path attributes + // precede those with less specific. Ordering with respect to other + // attributes (e.g., Domain) is unspecified. + $domain = $param['domain']; + if (strpos($domain, '.') === false) $domain .= '.local'; + $request_path = $param['path']; + if ($request_path == '') $request_path = '/'; + $request_secure = $param['secure']; + $now = time(); + $matched_cookies = array(); + // domain - find matching domains + $this->debug('Finding matching domains for '.$domain, __FILE__, __LINE__); + while (strpos($domain, '.') !== false) { + if (isset($this->cookies[$domain])) { + $this->debug(' domain match found: '.$domain); + $cookies =& $this->cookies[$domain]; + } else { + $domain = $this->_reduce_domain($domain); + continue; + } + // paths - find matching paths starting from most specific + $this->debug(' - Finding matching paths for '.$request_path); + $paths = array_keys($cookies); + usort($paths, array($this, '_cmp_length')); + foreach ($paths as $path) { + // continue to next cookie if request path does not path-match cookie path + if (!$this->_path_match($request_path, $path)) continue; + // loop through cookie names + $this->debug(' path match found: '.$path); + foreach ($cookies[$path] as $name => $values) { + // if this cookie is secure but request isn't, continue to next cookie + if ($values[1] && !$request_secure) continue; + // if cookie is not a session cookie and has expired, continue to next cookie + if (is_int($values[2]) && ($values[2] < $now)) continue; + // cookie matches request + $this->debug(' cookie match: '.$name.'='.$values[0]); + $matched_cookies[] = $name.'='.$values[0]; + } + } + $domain = $this->_reduce_domain($domain); + } + // return cookies + return implode('; ', $matched_cookies); + } + + /** + * Parse Set-Cookie values. + * + * Only use this method if you cannot use extract_cookies(), for example, if you want to use + * this cookie jar class without using the response class. + * + * @param array $set_cookies array holding 1 or more "Set-Cookie" header values + * @param array $param associative array containing 'host', 'path' keys + * @return void + * @see extract_cookies() + */ + public function storeCookies($url, $set_cookies) + { + if (count($set_cookies) == 0) return; + $param = @parse_url($url); + if (!is_array($param) || !isset($param['host'])) return; + $request_host = $param['host']; + if (strpos($request_host, '.') === false) $request_host .= '.local'; + $request_path = @$param['path']; + if ($request_path == '') $request_path = '/'; + // + // loop through set-cookie headers + // + foreach ($set_cookies as $set_cookie) { + $this->debug('Parsing: '.$set_cookie); + // temporary cookie store (before adding to jar) + $tmp_cookie = array(); + $param = explode(';', $set_cookie); + // loop through params + for ($x=0; $x
[\s\h\v]*
!u', '', $html); + } + } + } + + $title = ($extractor->getTitle() != '' ? $extractor->getTitle() : _('Untitled')); + $content = array ('title' => $title, 'body' => $html); + + return $content; } - public function fetchContent() + private function convert_to_utf8($html, $header=null) { - if ($this->isCorrect()) { - $this->clean(); - $html = Encoding::toUTF8(Tools::getFile($this->getUrl())); - - # if Tools::getFile() if not able to retrieve HTTPS content, try the same URL with HTTP protocol - if (!preg_match('!^https?://!i', $this->getUrl()) && (!isset($html) || strlen($html) <= 0)) { - $this->setUrl('http://' . $this->getUrl()); - $html = Encoding::toUTF8(Tools::getFile($this->getUrl())); + $encoding = null; + if ($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 + } 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"); } - - if (function_exists('tidy_parse_string')) { - $tidy = tidy_parse_string($html, array(), 'UTF8'); - $tidy->cleanRepair(); - - //Warning: tidy might fail so, ensure there is still a content - $body = $tidy->body(); - - //hasChildren does not seem to work, just check the string - //returned (and do not forget to clean the white spaces) - if (preg_replace('/\s+/', '', $body->value) !== "") { - $html = $tidy->value; + // TODO: check to see if encoding is supported (can we convert it?) + // If it's not, result will be empty string. + // For now we'll check for invalid encoding types returned by some sites, e.g. 'none' + // Problem URL: http://facta.co.jp/blog/archives/20111026001026.html + if (!$encoding || $encoding == 'none') { + // search for encoding in HTML - only look at the first 35000 characters + $html_head = substr($html, 0, 40000); + 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; + } + } } } - - $parameters = array(); - if (isset($html) and strlen($html) > 0) - { - $readability = new Readability($html, $this->getUrl()); - $readability->convertLinksToFootnotes = CONVERT_LINKS_FOOTNOTES; - $readability->revertForcedParagraphElements = REVERT_FORCED_PARAGRAPH_ELEMENTS; - - if($readability->init()) - { - $content = $readability->articleContent->innerHTML; - $parameters['title'] = ($readability->articleTitle->innerHTML != '' ? $readability->articleTitle->innerHTML : _('Untitled')); - $parameters['content'] = $content; - - return $parameters; + if (isset($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) { + $encoding = 'utf-8'; + } else { + if (strtolower($encoding) != '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', $encoding); + } + */ } } } - else { - #$msg->add('e', _('error during url preparation : the link is not valid')); - Tools::logm($this->getUrl() . ' is not a valid url'); - } + return $html; + } - return FALSE; + private function makeAbsolute($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); + $this->makeAbsoluteAttr($base, $e, $attr); + } + if (strtolower($elem->tagName) == $tag) $this->makeAbsoluteAttr($base, $elem, $attr); + } + } + + private function makeAbsoluteAttr($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.