Link without domain.

[warmspringwinds]

Sometimes when I fetch an image from the article's page, I get the a link without the domain from article.top_image.src like '/IMG/someimage.jpg'

Should it work this way?

[jice-lavocat]

I came for the same story.
I guess an absolute URL would be better suited for more flexible behavior.

[Edit] I had a look to the code in the repo, it seems fine. I am using an old Goose version that I need to update first to confirm the behavior.

[jice-lavocat]

I confirm the bug for version 1.0.25.
I'm trying to find a way to fix it. At the moment, it seems that, when ImageExtractor is called, the object

is empty. Hence, there is no self.target_url that can be used to rebuild the absolute URL.

[Edit] The URL I am using to test this is : http://www.suwa.fr/news/terre/trail-randonnee/trail-blanc-de-la-restonica-2014-549