From 7abda3ba524219b8ce304ef2774e6352747c829e Mon Sep 17 00:00:00 2001 From: Jeremy Benoist Date: Wed, 29 May 2019 17:05:12 +0200 Subject: Drop SimplePie It was only used to make an absolute url when downloading images. The deps is still there (in the `composer.lock`) because Graby use it (not for absolute but for encoding). --- composer.json | 1 - composer.lock | 2 +- src/Wallabag/CoreBundle/Helper/DownloadImages.php | 31 +++++++++++------------ 3 files changed, 16 insertions(+), 18 deletions(-) diff --git a/composer.json b/composer.json index 55e7f765..b0c17385 100644 --- a/composer.json +++ b/composer.json @@ -63,7 +63,6 @@ "nelmio/api-doc-bundle": "^2.13.2", "mgargano/simplehtmldom": "~1.5", "wallabag/tcpdf": "^6.2.26", - "simplepie/simplepie": "~1.5", "willdurand/hateoas-bundle": "~1.3", "liip/theme-bundle": "^1.4.6", "lexik/form-filter-bundle": "^5.0.4", diff --git a/composer.lock b/composer.lock index cbb9265d..a2a48c1e 100644 --- a/composer.lock +++ b/composer.lock @@ -4,7 +4,7 @@ "Read more about it at https://getcomposer.org/doc/01-basic-usage.md#installing-dependencies", "This file is @generated automatically" ], - "content-hash": "883f44eda34a48c8ddabc3294498d996", + "content-hash": "c42e1b50f4a2b8a59ca06c5ccb24e6a3", "packages": [ { "name": "bdunogier/guzzle-site-authenticator", diff --git a/src/Wallabag/CoreBundle/Helper/DownloadImages.php b/src/Wallabag/CoreBundle/Helper/DownloadImages.php index 7a39a2e4..1d361d6d 100644 --- a/src/Wallabag/CoreBundle/Helper/DownloadImages.php +++ b/src/Wallabag/CoreBundle/Helper/DownloadImages.php @@ -2,6 +2,8 @@ namespace Wallabag\CoreBundle\Helper; +use GuzzleHttp\Psr7\Uri; +use GuzzleHttp\Psr7\UriResolver; use Http\Client\Common\HttpMethodsClient; use Http\Client\Common\Plugin\ErrorPlugin; use Http\Client\Common\PluginClient; @@ -45,10 +47,8 @@ class DownloadImages public static function extractImagesUrlsFromHtml($html) { $crawler = new Crawler($html); - $imagesCrawler = $crawler - ->filterXpath('//img'); - $imagesUrls = $imagesCrawler - ->extract(['src']); + $imagesCrawler = $crawler->filterXpath('//img'); + $imagesUrls = $imagesCrawler->extract(['src']); $imagesSrcsetUrls = self::getSrcsetUrls($imagesCrawler); return array_unique(array_merge($imagesUrls, $imagesSrcsetUrls)); @@ -220,22 +220,25 @@ class DownloadImages private static function getSrcsetUrls(Crawler $imagesCrawler) { $urls = []; - $iterator = $imagesCrawler - ->getIterator(); + $iterator = $imagesCrawler->getIterator(); + while ($iterator->valid()) { $srcsetAttribute = $iterator->current()->getAttribute('srcset'); + if ('' !== $srcsetAttribute) { // Couldn't start with " OR ' OR a white space // Could be one or more white space // Must be one or more digits followed by w OR x $pattern = "/(?:[^\"'\s]+\s*(?:\d+[wx])+)/"; preg_match_all($pattern, $srcsetAttribute, $matches); + $srcset = \call_user_func_array('array_merge', $matches); $srcsetUrls = array_map(function ($src) { return trim(explode(' ', $src, 2)[0]); }, $srcset); $urls = array_merge($srcsetUrls, $urls); } + $iterator->next(); } @@ -292,20 +295,16 @@ class DownloadImages return $url; } - $base = new \SimplePie_IRI($base); + $base = new Uri($base); - // remove '//' in URL path (causes URLs not to resolve properly) - if (isset($base->ipath)) { - $base->ipath = preg_replace('!//+!', '/', $base->ipath); - } + // in case the url has no scheme & host + if ('' === $base->getAuthority() || '' === $base->getScheme()) { + $this->logger->error('DownloadImages: Can not make an absolute link', ['base' => $base, 'url' => $url]); - if ($absolute = \SimplePie_IRI::absolutize($base, $url)) { - return $absolute->get_uri(); + return false; } - $this->logger->error('DownloadImages: Can not make an absolute link', ['base' => $base, 'url' => $url]); - - return false; + return (string) UriResolver::resolve($base, new Uri($url)); } /** -- cgit v1.2.3