X-Git-Url: https://git.immae.eu/?a=blobdiff_plain;f=src%2FWallabag%2FCoreBundle%2FHelper%2FContentProxy.php;h=c7ccfe4a0e2f871c9daa1528b6394663e7ddfc55;hb=refs%2Fheads%2Furl-3529;hp=1ac7ad83c8ab753231ea72ec5c05db1205ae9aaf;hpb=e56f82142ac890c916b74700087424017bb2acf9;p=github%2Fwallabag%2Fwallabag.git diff --git a/src/Wallabag/CoreBundle/Helper/ContentProxy.php b/src/Wallabag/CoreBundle/Helper/ContentProxy.php index 1ac7ad83..c7ccfe4a 100644 --- a/src/Wallabag/CoreBundle/Helper/ContentProxy.php +++ b/src/Wallabag/CoreBundle/Helper/ContentProxy.php @@ -24,8 +24,9 @@ class ContentProxy protected $mimeGuesser; protected $fetchingErrorMessage; protected $eventDispatcher; + protected $storeArticleHeaders; - public function __construct(Graby $graby, RuleBasedTagger $tagger, ValidatorInterface $validator, LoggerInterface $logger, $fetchingErrorMessage) + public function __construct(Graby $graby, RuleBasedTagger $tagger, ValidatorInterface $validator, LoggerInterface $logger, $fetchingErrorMessage, $storeArticleHeaders = false) { $this->graby = $graby; $this->tagger = $tagger; @@ -33,6 +34,7 @@ class ContentProxy $this->logger = $logger; $this->mimeGuesser = new MimeTypeExtensionGuesser(); $this->fetchingErrorMessage = $fetchingErrorMessage; + $this->storeArticleHeaders = $storeArticleHeaders; } /** @@ -63,6 +65,13 @@ class ContentProxy // so we'll be able to refetch it in the future $content['url'] = !empty($content['url']) ? $content['url'] : $url; + // In one case (at least in tests), url is empty here + // so we set it using $url provided in the updateEntry call. + // Not sure what are the other possible cases where this property is empty + if (empty($entry->getUrl()) && !empty($url)) { + $entry->setUrl($url); + } + $this->stockEntry($entry, $content); } @@ -125,7 +134,7 @@ class ContentProxy $date = $value; // is it a timestamp? - if (filter_var($date, FILTER_VALIDATE_INT) !== false) { + if (false !== filter_var($date, FILTER_VALIDATE_INT)) { $date = '@' . $date; } @@ -142,6 +151,38 @@ class ContentProxy } } + /** + * Helper to extract and save host from entry url. + * + * @param Entry $entry + */ + public function setEntryDomainName(Entry $entry) + { + $domainName = parse_url($entry->getUrl(), PHP_URL_HOST); + if (false !== $domainName) { + $entry->setDomainName($domainName); + } + } + + /** + * Helper to set a default title using: + * - url basename, if applicable + * - hostname. + * + * @param Entry $entry + */ + public function setDefaultEntryTitle(Entry $entry) + { + $url = parse_url($entry->getUrl()); + $path = pathinfo($url['path'], PATHINFO_BASENAME); + + if (empty($path)) { + $path = $url['host']; + } + + $entry->setTitle($path); + } + /** * Stock entry with fetched or imported content. * Will fall back to OpenGraph data if available. @@ -151,13 +192,18 @@ class ContentProxy */ private function stockEntry(Entry $entry, array $content) { - $entry->setUrl($content['url']); - - $domainName = parse_url($entry->getUrl(), PHP_URL_HOST); - if (false !== $domainName) { - $entry->setDomainName($domainName); + // When a redirection occurs while fetching an entry + // we move the original url in origin_url property if empty + // and set the entry url with the final value + if (!empty($content['url']) && $entry->getUrl() !== $content['url']) { + if (empty($entry->getOriginUrl())) { + $entry->setOriginUrl($entry->getUrl()); + } + $entry->setUrl($content['url']); } + $this->setEntryDomainName($entry); + if (!empty($content['title'])) { $entry->setTitle($content['title']); } elseif (!empty($content['open_graph']['og_title'])) { @@ -185,7 +231,7 @@ class ContentProxy $entry->setPublishedBy($content['authors']); } - if (!empty($content['all_headers'])) { + if (!empty($content['all_headers']) && $this->storeArticleHeaders) { $entry->setHeaders($content['all_headers']); }