Kelson
I have fixed the recipe and started a new scrape with `dev` https://farm.openzim.org/pipeline/a555932f-52be-4f79-a9b0-9449af0df2da
Unfortunately not yet available for FOSS, see https://github.com/actions/runner-images/issues/10820
@benoit74 I like the idea even if the detailed approach should be discussed. To me it sounds your proposal is to make a Zimfarm bot for GitHub.
> If you really wanna challenge the ENV variables approach, please open another issue. I can not really - right now - because I need first to understand clearly how...
Reopening as it seems we face a software bug here.
As long as we won't do a proper HTML parsing, we will suffer of these kind of problems IMHO.
@benoit74 Remarked that this was tracked nowhere, but this is a really impairing "stupid" issue. Assigning to you for assessment.
> for illustrations, we could have a system which automatically grabs the biggest favicon available on a user-provided website (most sites now have apple-size favicons which are appropriate to downscale...
https://github.com/openzim/zimfarm/issues/552 almost a duplicate
I think I agree with everything what @benoit74 has written, here is my version: To this day, we have been using an online _drive_ to save pictures, javascript, CSS files...