X-Git-Url: https://git.immae.eu/?a=blobdiff_plain;f=src%2FWallabag%2FCoreBundle%2FHelper%2FContentProxy.php;h=ddecd6f4839de3c5e53c9f2638646f1269a0f5d8;hb=f808b01692a835673f328d7221ba8c212caa9b61;hp=cd18c668f7ea3b1b342260bca27591a6022c4103;hpb=d5c2cc54b5490b0bec46f39d7706d5d46869e872;p=github%2Fwallabag%2Fwallabag.git diff --git a/src/Wallabag/CoreBundle/Helper/ContentProxy.php b/src/Wallabag/CoreBundle/Helper/ContentProxy.php index cd18c668..ddecd6f4 100644 --- a/src/Wallabag/CoreBundle/Helper/ContentProxy.php +++ b/src/Wallabag/CoreBundle/Helper/ContentProxy.php @@ -4,10 +4,12 @@ namespace Wallabag\CoreBundle\Helper; use Graby\Graby; use Psr\Log\LoggerInterface; +use Symfony\Component\HttpFoundation\File\MimeType\MimeTypeExtensionGuesser; +use Symfony\Component\Validator\Constraints\Locale as LocaleConstraint; +use Symfony\Component\Validator\Constraints\Url as UrlConstraint; +use Symfony\Component\Validator\Validator\ValidatorInterface; use Wallabag\CoreBundle\Entity\Entry; use Wallabag\CoreBundle\Tools\Utils; -use Symfony\Component\HttpFoundation\File\MimeType\MimeTypeExtensionGuesser; -use Symfony\Component\Config\Definition\Exception\Exception; /** * This kind of proxy class take care of getting the content from an url @@ -17,72 +19,50 @@ class ContentProxy { protected $graby; protected $tagger; + protected $validator; protected $logger; protected $mimeGuesser; protected $fetchingErrorMessage; protected $eventDispatcher; - public function __construct(Graby $graby, RuleBasedTagger $tagger, LoggerInterface $logger, $fetchingErrorMessage) + public function __construct(Graby $graby, RuleBasedTagger $tagger, ValidatorInterface $validator, LoggerInterface $logger, $fetchingErrorMessage) { $this->graby = $graby; $this->tagger = $tagger; + $this->validator = $validator; $this->logger = $logger; $this->mimeGuesser = new MimeTypeExtensionGuesser(); $this->fetchingErrorMessage = $fetchingErrorMessage; } /** - * Update existing entry by fetching from URL using Graby. - * - * @param Entry $entry Entry to update - * @param string $url Url to grab content for - */ - public function updateEntry(Entry $entry, $url) - { - $content = $this->graby->fetchContent($url); - - // be sure to keep the url in case of error - // so we'll be able to refetch it in the future - $content['url'] = $content['url'] ?: $url; - - $this->stockEntry($entry, $content); - } - - /** - * Import entry using either fetched or provided content. + * Update entry using either fetched or provided content. * * @param Entry $entry Entry to update + * @param string $url Url of the content * @param array $content Array with content provided for import with AT LEAST keys title, html, url to skip the fetchContent from the url * @param bool $disableContentUpdate Whether to skip trying to fetch content using Graby */ - public function importEntry(Entry $entry, array $content, $disableContentUpdate = false) + public function updateEntry(Entry $entry, $url, array $content = [], $disableContentUpdate = false) { - try { - $this->validateContent($content); - } catch (\Exception $e) { - // validation failed but do we want to disable updating content? - if (true === $disableContentUpdate) { - throw $e; - } + if (!empty($content['html'])) { + $content['html'] = $this->graby->cleanupHtml($content['html'], $url); } - if (false === $disableContentUpdate) { - try { - $fetchedContent = $this->graby->fetchContent($content['url']); - } catch (\Exception $e) { - $this->logger->error('Error while trying to fetch content from URL.', [ - 'entry_url' => $content['url'], - 'error_msg' => $e->getMessage(), - ]); - } + if ((empty($content) || false === $this->validateContent($content)) && false === $disableContentUpdate) { + $fetchedContent = $this->graby->fetchContent($url); // when content is imported, we have information in $content // in case fetching content goes bad, we'll keep the imported information instead of overriding them - if ($fetchedContent['html'] !== $this->fetchingErrorMessage) { + if (empty($content) || $fetchedContent['html'] !== $this->fetchingErrorMessage) { $content = $fetchedContent; } } + // be sure to keep the url in case of error + // so we'll be able to refetch it in the future + $content['url'] = !empty($content['url']) ? $content['url'] : $url; + $this->stockEntry($entry, $content); } @@ -91,7 +71,7 @@ class ContentProxy * Will fall back to OpenGraph data if available. * * @param Entry $entry Entry to stock - * @param array $content Array with at least title and URL + * @param array $content Array with at least title, url & html */ private function stockEntry(Entry $entry, array $content) { @@ -120,17 +100,17 @@ class ContentProxy // is it a timestamp? if (filter_var($date, FILTER_VALIDATE_INT) !== false) { - $date = '@'.$content['date']; + $date = '@' . $content['date']; } try { $entry->setPublishedAt(new \DateTime($date)); } catch (\Exception $e) { - $this->logger->warning('Error while defining date', ['e' => $e, 'url' => $url, 'date' => $content['date']]); + $this->logger->warning('Error while defining date', ['e' => $e, 'url' => $content['url'], 'date' => $content['date']]); } } - if (!empty($content['authors'])) { + if (!empty($content['authors']) && is_array($content['authors'])) { $entry->setPublishedBy($content['authors']); } @@ -138,7 +118,24 @@ class ContentProxy $entry->setHeaders($content['all_headers']); } - $entry->setLanguage(isset($content['language']) ? $content['language'] : ''); + $this->validateAndSetLanguage( + $entry, + isset($content['language']) ? $content['language'] : null + ); + + $this->validateAndSetPreviewPicture( + $entry, + isset($content['open_graph']['og_image']) ? $content['open_graph']['og_image'] : null + ); + + // if content is an image, define it as a preview too + if (!empty($content['content_type']) && in_array($this->mimeGuesser->guess($content['content_type']), ['jpeg', 'jpg', 'gif', 'png'], true)) { + $this->validateAndSetPreviewPicture( + $entry, + $content['url'] + ); + } + $entry->setMimetype(isset($content['content_type']) ? $content['content_type'] : ''); $entry->setReadingTime(Utils::getReadingTime($html)); @@ -147,15 +144,6 @@ class ContentProxy $entry->setDomainName($domainName); } - if (!empty($content['open_graph']['og_image'])) { - $entry->setPreviewPicture($content['open_graph']['og_image']); - } - - // if content is an image define as a preview too - if (!empty($content['content_type']) && in_array($this->mimeGuesser->guess($content['content_type']), ['jpeg', 'jpg', 'gif', 'png'], true)) { - $entry->setPreviewPicture($content['url']); - } - try { $this->tagger->tag($entry); } catch (\Exception $e) { @@ -170,19 +158,59 @@ class ContentProxy * Validate that the given content has at least a title, an html and a url. * * @param array $content + * + * @return bool true if valid otherwise false */ private function validateContent(array $content) { - if (empty($content['title'])) { - throw new Exception('Missing title from imported entry!'); - } + return !empty($content['title']) && !empty($content['html']) && !empty($content['url']); + } - if (empty($content['url'])) { - throw new Exception('Missing URL from imported entry!'); + /** + * Use a Symfony validator to ensure the language is well formatted. + * + * @param Entry $entry + * @param string $value Language to validate + */ + private function validateAndSetLanguage($entry, $value) + { + // some lang are defined as fr-FR, es-ES. + // replacing - by _ might increase language support + $value = str_replace('-', '_', $value); + + $errors = $this->validator->validate( + $value, + (new LocaleConstraint()) + ); + + if (0 === count($errors)) { + $entry->setLanguage($value); + + return; } - if (empty($content['html'])) { - throw new Exception('Missing html from imported entry!'); + $this->logger->warning('Language validation failed. ' . (string) $errors); + } + + /** + * Use a Symfony validator to ensure the preview picture is a real url. + * + * @param Entry $entry + * @param string $value URL to validate + */ + private function validateAndSetPreviewPicture($entry, $value) + { + $errors = $this->validator->validate( + $value, + (new UrlConstraint()) + ); + + if (0 === count($errors)) { + $entry->setPreviewPicture($value); + + return; } + + $this->logger->warning('PreviewPicture validation failed. ' . (string) $errors); } }