From 6405fdea2a0790807430b7201794cece33fcce11 Mon Sep 17 00:00:00 2001 From: Fabien Meurillon Date: Mon, 4 Aug 2014 01:56:12 +0200 Subject: [PATCH] Begin to properly import Readability classes in Symfony --- app/AppKernel.php | 1 + app/config/routing.yml | 5 + composer.json | 3 +- composer.lock | 56 ++- .../Controller/DefaultController.php | 19 + .../DependencyInjection/Configuration.php | 29 ++ .../WallabagReadabilityExtension.php | 28 ++ .../Readability/WallabagReadability.php | 48 +++ .../Resources/config/services.xml | 14 + .../Resources/views/Default/index.html.twig | 1 + .../Controller/DefaultControllerTest.php | 17 + .../ReadabilityBundle/Url/DummySingleItem.php | 16 + .../Url/DummySingleItemFeed.php | 13 + .../ReadabilityBundle/Url/UrlExtractor.php | 392 ++++++++++++++++++ .../WallabagReadabilityBundle.php | 9 + 15 files changed, 649 insertions(+), 2 deletions(-) create mode 100644 src/Wallabag/Bundle/ReadabilityBundle/Controller/DefaultController.php create mode 100644 src/Wallabag/Bundle/ReadabilityBundle/DependencyInjection/Configuration.php create mode 100644 src/Wallabag/Bundle/ReadabilityBundle/DependencyInjection/WallabagReadabilityExtension.php create mode 100644 src/Wallabag/Bundle/ReadabilityBundle/Readability/WallabagReadability.php create mode 100644 src/Wallabag/Bundle/ReadabilityBundle/Resources/config/services.xml create mode 100644 src/Wallabag/Bundle/ReadabilityBundle/Resources/views/Default/index.html.twig create mode 100644 src/Wallabag/Bundle/ReadabilityBundle/Tests/Controller/DefaultControllerTest.php create mode 100644 src/Wallabag/Bundle/ReadabilityBundle/Url/DummySingleItem.php create mode 100644 src/Wallabag/Bundle/ReadabilityBundle/Url/DummySingleItemFeed.php create mode 100644 src/Wallabag/Bundle/ReadabilityBundle/Url/UrlExtractor.php create mode 100644 src/Wallabag/Bundle/ReadabilityBundle/WallabagReadabilityBundle.php diff --git a/app/AppKernel.php b/app/AppKernel.php index a737a7f..a9403c9 100644 --- a/app/AppKernel.php +++ b/app/AppKernel.php @@ -20,6 +20,7 @@ class AppKernel extends Kernel new Wallabag\Bundle\CoreBundle\WallabagCoreBundle(), new Wallabag\Bundle\ApiBundle\WallabagApiBundle(), new Wallabag\Bundle\CliBundle\WallabagCliBundle(), + new Wallabag\Bundle\ReadabilityBundle\WallabagReadabilityBundle(), ); if (in_array($this->getEnvironment(), array('dev', 'test'))) { diff --git a/app/config/routing.yml b/app/config/routing.yml index c228057..b14cf91 100644 --- a/app/config/routing.yml +++ b/app/config/routing.yml @@ -1,3 +1,8 @@ +wallabag_readability: + resource: "@WallabagReadabilityBundle/Controller/" + type: annotation + prefix: / + wallabag_cli: resource: "@WallabagCliBundle/Controller/" type: annotation diff --git a/composer.json b/composer.json index 2024992..2b16b42 100644 --- a/composer.json +++ b/composer.json @@ -19,7 +19,8 @@ "sensio/framework-extra-bundle": "~3.0", "incenteev/composer-parameter-handler": "~2.0", "fivefilters/php-readability": "v1.0", - "friendsofsymfony/user-bundle": "2.0.x-dev" + "friendsofsymfony/user-bundle": "2.0.x-dev", + "simplepie/simplepie": "1.3.1" }, "require-dev": { "sensio/generator-bundle": "~2.3" diff --git a/composer.lock b/composer.lock index e1e32b3..135367b 100644 --- a/composer.lock +++ b/composer.lock @@ -4,7 +4,7 @@ "Read more about it at http://getcomposer.org/doc/01-basic-usage.md#composer-lock-the-lock-file", "This file is @generated automatically" ], - "hash": "4f4cbc95623d5596ec8cf7cb1e3e4b11", + "hash": "aaca2ba5bd2f244555f16fb4cfa50fd4", "packages": [ { "name": "doctrine/annotations", @@ -1138,6 +1138,60 @@ ], "time": "2014-05-22 23:27:44" }, + { + "name": "simplepie/simplepie", + "version": "1.3.1", + "source": { + "type": "git", + "url": "https://github.com/simplepie/simplepie.git", + "reference": "ce53709778bc1e2e4deda1651b66e5081398d5cc" + }, + "dist": { + "type": "zip", + "url": "https://api.github.com/repos/simplepie/simplepie/zipball/ce53709778bc1e2e4deda1651b66e5081398d5cc", + "reference": "ce53709778bc1e2e4deda1651b66e5081398d5cc", + "shasum": "" + }, + "require": { + "php": ">=5.2.0" + }, + "type": "library", + "autoload": { + "psr-0": { + "SimplePie": "library" + } + }, + "notification-url": "https://packagist.org/downloads/", + "license": [ + "BSD-3-Clause" + ], + "authors": [ + { + "name": "Ryan Parman", + "homepage": "http://ryanparman.com/", + "role": "Creator, alumnus developer" + }, + { + "name": "Geoffrey Sneddon", + "homepage": "http://gsnedders.com/", + "role": "Alumnus developer" + }, + { + "name": "Ryan McCue", + "email": "me@ryanmccue.info", + "homepage": "http://ryanmccue.info/", + "role": "Developer" + } + ], + "description": "A simple Atom/RSS parsing library for PHP", + "homepage": "http://simplepie.org/", + "keywords": [ + "atom", + "feeds", + "rss" + ], + "time": "2012-10-30 17:54:03" + }, { "name": "swiftmailer/swiftmailer", "version": "v5.2.1", diff --git a/src/Wallabag/Bundle/ReadabilityBundle/Controller/DefaultController.php b/src/Wallabag/Bundle/ReadabilityBundle/Controller/DefaultController.php new file mode 100644 index 0000000..0bffd01 --- /dev/null +++ b/src/Wallabag/Bundle/ReadabilityBundle/Controller/DefaultController.php @@ -0,0 +1,19 @@ + $name); + } +} diff --git a/src/Wallabag/Bundle/ReadabilityBundle/DependencyInjection/Configuration.php b/src/Wallabag/Bundle/ReadabilityBundle/DependencyInjection/Configuration.php new file mode 100644 index 0000000..5079252 --- /dev/null +++ b/src/Wallabag/Bundle/ReadabilityBundle/DependencyInjection/Configuration.php @@ -0,0 +1,29 @@ +root('wallabag_readability'); + + // Here you should define the parameters that are allowed to + // configure your bundle. See the documentation linked above for + // more information on that topic. + + return $treeBuilder; + } +} diff --git a/src/Wallabag/Bundle/ReadabilityBundle/DependencyInjection/WallabagReadabilityExtension.php b/src/Wallabag/Bundle/ReadabilityBundle/DependencyInjection/WallabagReadabilityExtension.php new file mode 100644 index 0000000..b2c06dd --- /dev/null +++ b/src/Wallabag/Bundle/ReadabilityBundle/DependencyInjection/WallabagReadabilityExtension.php @@ -0,0 +1,28 @@ +processConfiguration($configuration, $configs); + + $loader = new Loader\XmlFileLoader($container, new FileLocator(__DIR__.'/../Resources/config')); + $loader->load('services.xml'); + } +} diff --git a/src/Wallabag/Bundle/ReadabilityBundle/Readability/WallabagReadability.php b/src/Wallabag/Bundle/ReadabilityBundle/Readability/WallabagReadability.php new file mode 100644 index 0000000..5b6c4d5 --- /dev/null +++ b/src/Wallabag/Bundle/ReadabilityBundle/Readability/WallabagReadability.php @@ -0,0 +1,48 @@ +getInnerText($this->dom->getElementsByTagName('title')->item(0)); + } catch(Exception $e) {} + + if (preg_match('/ [\|\-] /', $curTitle)) + { + $curTitle = preg_replace('/(.*)[\|\-] .*/i', '$1', $origTitle); + + if (count(explode(' ', $curTitle)) < 3) { + $curTitle = preg_replace('/[^\|\-]*[\|\-](.*)/i', '$1', $origTitle); + } + } + else if(strlen($curTitle) > 150 || strlen($curTitle) < 15) + { + $hOnes = $this->dom->getElementsByTagName('h1'); + if($hOnes->length == 1) + { + $curTitle = $this->getInnerText($hOnes->item(0)); + } + } + + $curTitle = trim($curTitle); + + if (count(explode(' ', $curTitle)) <= 4) { + $curTitle = $origTitle; + } + + $articleTitle = $this->dom->createElement('h1'); + $articleTitle->innerHTML = $curTitle; + + return $articleTitle; + } +} diff --git a/src/Wallabag/Bundle/ReadabilityBundle/Resources/config/services.xml b/src/Wallabag/Bundle/ReadabilityBundle/Resources/config/services.xml new file mode 100644 index 0000000..18cea25 --- /dev/null +++ b/src/Wallabag/Bundle/ReadabilityBundle/Resources/config/services.xml @@ -0,0 +1,14 @@ + + + + + + Wallabag\Bundle\ReadabilityBundle\Url\UrlExtractor + + + + + + diff --git a/src/Wallabag/Bundle/ReadabilityBundle/Resources/views/Default/index.html.twig b/src/Wallabag/Bundle/ReadabilityBundle/Resources/views/Default/index.html.twig new file mode 100644 index 0000000..4ce626e --- /dev/null +++ b/src/Wallabag/Bundle/ReadabilityBundle/Resources/views/Default/index.html.twig @@ -0,0 +1 @@ +Hello {{ name }}! diff --git a/src/Wallabag/Bundle/ReadabilityBundle/Tests/Controller/DefaultControllerTest.php b/src/Wallabag/Bundle/ReadabilityBundle/Tests/Controller/DefaultControllerTest.php new file mode 100644 index 0000000..f01bcc4 --- /dev/null +++ b/src/Wallabag/Bundle/ReadabilityBundle/Tests/Controller/DefaultControllerTest.php @@ -0,0 +1,17 @@ +request('GET', '/hello/Fabien'); + + $this->assertTrue($crawler->filter('html:contains("Hello Fabien")')->count() > 0); + } +} diff --git a/src/Wallabag/Bundle/ReadabilityBundle/Url/DummySingleItem.php b/src/Wallabag/Bundle/ReadabilityBundle/Url/DummySingleItem.php new file mode 100644 index 0000000..d97bf4e --- /dev/null +++ b/src/Wallabag/Bundle/ReadabilityBundle/Url/DummySingleItem.php @@ -0,0 +1,16 @@ +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, $prefer=null) { return null; } + public function get_enclosures() { return null; } + public function get_categories() { return null; } +} diff --git a/src/Wallabag/Bundle/ReadabilityBundle/Url/DummySingleItemFeed.php b/src/Wallabag/Bundle/ReadabilityBundle/Url/DummySingleItemFeed.php new file mode 100644 index 0000000..e4977f9 --- /dev/null +++ b/src/Wallabag/Bundle/ReadabilityBundle/Url/DummySingleItemFeed.php @@ -0,0 +1,13 @@ +item = new DummySingleItem($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); } +} diff --git a/src/Wallabag/Bundle/ReadabilityBundle/Url/UrlExtractor.php b/src/Wallabag/Bundle/ReadabilityBundle/Url/UrlExtractor.php new file mode 100644 index 0000000..329701d --- /dev/null +++ b/src/Wallabag/Bundle/ReadabilityBundle/Url/UrlExtractor.php @@ -0,0 +1,392 @@ + + * @copyright 2013 + * @license http://www.wtfpl.net/ see COPYING file + */ + +namespace Wallabag\Bundle\ReadabilityBundle\Url; + +use Wallabag\Bundle\ReadabilityBundle\Readability\WallabagReadability; + +class UrlExtractor +{ + public $url; + + private $fingerprints = array( + // Posterous + ' array('hostname'=>'fingerprint.posterous.com', 'head'=>true), + // Blogger + ' array('hostname'=>'fingerprint.blogspot.com', 'head'=>true), + ' array('hostname'=>'fingerprint.blogspot.com', 'head'=>true), + // WordPress (self-hosted and hosted) + '\"$_name\""; + } else { + $html = "Download $_name"; + } + $title = $_name; + $do_content_extraction = false; + break; + } + } + } + unset($_mime, $_act, $_name, $match); + } + if ($do_content_extraction) { + $html = $response['body']; + // remove strange things + $html = str_replace('', '', $html); + $html = $this->convert_to_utf8($html, $response['headers']); + + // check site config for single page URL - fetch it if found + if ($single_page_response = $this->getSinglePage($item, $html, $effective_url)) { + $html = $single_page_response['body']; + // remove strange things + $html = str_replace('', '', $html); + $html = $this->convert_to_utf8($html, $single_page_response['headers']); + $effective_url = $single_page_response['effective_url']; + unset($single_page_response); + } + $extract_result = $extractor->process($html, $effective_url); + $readability = $extractor->readability; + $content_block = ($extract_result) ? $extractor->getContent() : null; + } + } + if ($do_content_extraction) { + // if we failed to extract content... + if (!$extract_result) { + $html = $this->error_message; + // keep the original item description + $html .= $item->get_description(); + } else { + $readability->clean($content_block, 'select'); + if ($this->rewrite_relative_urls) $this->makeAbsolute($effective_url, $content_block); + if ($content_block->childNodes->length == 1 && $content_block->firstChild->nodeType === XML_ELEMENT_NODE) { + $html = $content_block->firstChild->innerHTML; + } else { + $html = $content_block->innerHTML; + } + // post-processing cleanup + $html = preg_replace('!

[\s\h\v]*

!u', '', $html); + } + } + } + + $title = ($extractor->getTitle() != '' ? $extractor->getTitle() : _('Untitled')); + $content = array ('title' => $title, 'body' => $html); + + return $content; + } + + private function convert_to_utf8($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 + } 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 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; + } + } + } + } + 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); + } + */ + } + } + } + return $html; + } + + private function makeAbsolute($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. + $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, $absolute); + } + } + } + } + + private function makeAbsoluteStr($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 $absolute; + } + return false; + } + } + + // returns single page response, or false if not found + private function getSinglePage($item, $html, $url) { + global $http, $extractor; + $host = @parse_url($url, PHP_URL_HOST); + $site_config = \SiteConfig::build($host); + if ($site_config === false) { + // check for fingerprints + if (!empty($extractor->fingerprints) && ($_fphost = $extractor->findHostUsingFingerprints($html))) { + $site_config = \SiteConfig::build($_fphost); + } + if ($site_config === false) $site_config = new \SiteConfig(); + \SiteConfig::add_to_cache($host, $site_config); + return false; + } else { + \SiteConfig::add_to_cache($host, $site_config); + } + $splink = null; + if (!empty($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 + $readability = new WallabagReadability($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 = $item->getAttribute('href'); + break; + } elseif ($item instanceof DOMAttr && $item->value) { + $single_page_url = $item->value; + break; + } + } + } + } + // If we've got URL, resolve against $url + if (isset($single_page_url) && ($single_page_url = $this->makeAbsoluteStr($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; + } +} diff --git a/src/Wallabag/Bundle/ReadabilityBundle/WallabagReadabilityBundle.php b/src/Wallabag/Bundle/ReadabilityBundle/WallabagReadabilityBundle.php new file mode 100644 index 0000000..4042441 --- /dev/null +++ b/src/Wallabag/Bundle/ReadabilityBundle/WallabagReadabilityBundle.php @@ -0,0 +1,9 @@ +