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
$this->assertSame('1.1.1.1', $entry->getDomainName());
}
+ public function testWithContentAndContentImage()
+ {
+ $tagger = $this->getTaggerMock();
+ $tagger->expects($this->once())
+ ->method('tag');
+
+ $graby = $this->getMockBuilder('Graby\Graby')
+ ->setMethods(['fetchContent'])
+ ->disableOriginalConstructor()
+ ->getMock();
+
+ $graby->expects($this->any())
+ ->method('fetchContent')
+ ->willReturn([
+ 'html' => "<h1>Test</h1><p><img src='http://3.3.3.3/cover.jpg'/></p>",
+ 'title' => 'this is my title',
+ 'url' => 'http://1.1.1.1',
+ 'content_type' => 'text/html',
+ 'language' => 'fr',
+ 'status' => '200',
+ 'open_graph' => [
+ 'og_title' => 'my OG title',
+ 'og_description' => 'OG desc',
+ 'og_image' => null,
+ ],
+ ]);
+
+ $proxy = new ContentProxy($graby, $tagger, $this->getValidator(), $this->getLogger(), $this->fetchingErrorMessage);
+ $entry = new Entry(new User());
+ $proxy->updateEntry($entry, 'http://0.0.0.0');
+
+ $this->assertSame('http://1.1.1.1', $entry->getUrl());
+ $this->assertSame('this is my title', $entry->getTitle());
+ $this->assertSame("<h1>Test</h1><p><img src='http://3.3.3.3/cover.jpg'/></p>", $entry->getContent());
+ $this->assertSame('http://3.3.3.3/cover.jpg', $entry->getPreviewPicture());
+ $this->assertSame('text/html', $entry->getMimetype());
+ $this->assertSame('fr', $entry->getLanguage());
+ $this->assertSame('200', $entry->getHttpStatus());
+ $this->assertSame(0.0, $entry->getReadingTime());
+ $this->assertSame('1.1.1.1', $entry->getDomainName());
+ }
+
+ public function testWithContentImageAndOgImage()
+ {
+ $tagger = $this->getTaggerMock();
+ $tagger->expects($this->once())
+ ->method('tag');
+
+ $graby = $this->getMockBuilder('Graby\Graby')
+ ->setMethods(['fetchContent'])
+ ->disableOriginalConstructor()
+ ->getMock();
+
+ $graby->expects($this->any())
+ ->method('fetchContent')
+ ->willReturn([
+ 'html' => "<h1>Test</h1><p><img src='http://3.3.3.3/nevermind.jpg'/></p>",
+ 'title' => 'this is my title',
+ 'url' => 'http://1.1.1.1',
+ 'content_type' => 'text/html',
+ 'language' => 'fr',
+ 'status' => '200',
+ 'open_graph' => [
+ 'og_title' => 'my OG title',
+ 'og_description' => 'OG desc',
+ 'og_image' => 'http://3.3.3.3/cover.jpg',
+ ],
+ ]);
+
+ $proxy = new ContentProxy($graby, $tagger, $this->getValidator(), $this->getLogger(), $this->fetchingErrorMessage);
+ $entry = new Entry(new User());
+ $proxy->updateEntry($entry, 'http://0.0.0.0');
+
+ $this->assertSame('http://1.1.1.1', $entry->getUrl());
+ $this->assertSame('this is my title', $entry->getTitle());
+ $this->assertSame("<h1>Test</h1><p><img src='http://3.3.3.3/nevermind.jpg'/></p>", $entry->getContent());
+ $this->assertSame('http://3.3.3.3/cover.jpg', $entry->getPreviewPicture());
+ $this->assertSame('text/html', $entry->getMimetype());
+ $this->assertSame('fr', $entry->getLanguage());
+ $this->assertSame('200', $entry->getHttpStatus());
+ $this->assertSame(0.0, $entry->getReadingTime());
+ $this->assertSame('1.1.1.1', $entry->getDomainName());
+ }
+
public function testWithContentAndBadLanguage()
{
$tagger = $this->getTaggerMock();
$records = $handler->getRecords();
- $this->assertCount(1, $records);
+ $this->assertCount(3, $records);
$this->assertContains('Error while defining date', $records[0]['message']);
}