use Wallabag\CoreBundle\Tools\Utils;
/**
- * This kind of proxy class take care of getting the content from an url
- * and update the entry with what it found.
+ * This kind of proxy class takes care of getting the content from an url
+ * and updates the entry with what it found.
*/
class ContentProxy
{
$this->updateLanguage($entry, $content['language']);
}
+ $previewPictureUrl = '';
if (!empty($content['open_graph']['og_image'])) {
- $this->updatePreviewPicture($entry, $content['open_graph']['og_image']);
+ $previewPictureUrl = $content['open_graph']['og_image'];
}
// if content is an image, define it as a preview too
if (!empty($content['content_type']) && \in_array($this->mimeGuesser->guess($content['content_type']), ['jpeg', 'jpg', 'gif', 'png'], true)) {
- $this->updatePreviewPicture($entry, $content['url']);
+ $previewPictureUrl = $content['url'];
+ } elseif (empty($previewPictureUrl)) {
+ $this->logger->debug('Extracting images from content to provide a default preview picture');
+ $imagesUrls = DownloadImages::extractImagesUrlsFromHtml($content['html']);
+ $this->logger->debug(\count($imagesUrls) . ' pictures found');
+ if (!empty($imagesUrls)) {
+ $previewPictureUrl = $imagesUrls[0];
+ }
+ }
+
+ if (!empty($previewPictureUrl)) {
+ $this->updatePreviewPicture($entry, $previewPictureUrl);
}
if (!empty($content['content_type'])) {
}
/**
- * Process the html and extract image from it, save them to local and return the updated html.
+ * Process the html and extract images URLs from it.
*
- * @param int $entryId ID of the entry
* @param string $html
- * @param string $url Used as a base path for relative image and folder
*
- * @return string
+ * @return string[]
*/
- public function processHtml($entryId, $html, $url)
+ public static function extractImagesUrlsFromHtml($html)
{
$crawler = new Crawler($html);
$imagesCrawler = $crawler
->filterXpath('//img');
$imagesUrls = $imagesCrawler
->extract(['src']);
- $imagesSrcsetUrls = $this->getSrcsetUrls($imagesCrawler);
- $imagesUrls = array_unique(array_merge($imagesUrls, $imagesSrcsetUrls));
+ $imagesSrcsetUrls = self::getSrcsetUrls($imagesCrawler);
+
+ return array_unique(array_merge($imagesUrls, $imagesSrcsetUrls));
+ }
+
+ /**
+ * Process the html and extract image from it, save them to local and return the updated html.
+ *
+ * @param int $entryId ID of the entry
+ * @param string $html
+ * @param string $url Used as a base path for relative image and folder
+ *
+ * @return string
+ */
+ public function processHtml($entryId, $html, $url)
+ {
+ $imagesUrls = self::extractImagesUrlsFromHtml($html);
$relativePath = $this->getRelativePath($entryId);
*
* @return array An array of urls
*/
- private function getSrcsetUrls(Crawler $imagesCrawler)
+ private static function getSrcsetUrls(Crawler $imagesCrawler)
{
$urls = [];
$iterator = $imagesCrawler