New request: h-node

[ToeushiMWYKaesai]

• Website URL: https://h-node.org/home/index/en
• License: **Public domain CC0
• Desired ZIM Title: h-node
• Desired ZIM Description: Database of hardware that work with a fully free software operating system.
• Desired ZIM Icon –png (URL or attach one): https://h-node.org/favicon.ico
• Language (ISO 639-3): eng
• Is this a MediaWiki?: no

[RavanJAltaie]

Recipe created
https://farm.openzim.org/recipes/h-node_en_all
Will update the library link once ready

[RavanJAltaie]

The recipe succeeded but the file is incomplete (only main page has been scrapped).
I'm thinking to separate this into 2 files: first for the issues: https://h-node.org/issues/viewall/en/1/token
Second for the downloads: https://h-node.org/download/index/en
The main page of the website is useless anyway.
What do you think? @Popolechien @rgaudin

[Popolechien]

Am looking at https://dev.library.kiwix.org/viewer#h-node_en_all_2023-09/A/h-node.org/home/index/en and the website does not look particularly sophisticated - at least the hardware tab works. Maybe we need to figure something out with the parameters?

@RavanJAltaie it would take a h-node user to know whether splitting issues from downloads from the rest is a good idea. If you (or @benoit74) can figure out a way to hide them (via CSS?) that would also be good. More importantly, I think it would be more useful to split the various languages (fr, en,de, etc..)

[rgaudin]

Am looking at https://dev.library.kiwix.org/viewer#h-node_en_all_2023-09/A/h-node.org/home/index/en and the website does not look particularly sophisticated - at least the hardware tab works.

No it doesn't and can not since it's not been scraped.

Maybe we need to figure something out with the parameters?

Yes, it's still that same mistake: you use the default scope which uses the homepage folder as prefix but the content is on a different prefix… so it's not captured.

IMO, the whole thing is probably small enough and useful together (one per language though). Hiding the elements related to Search and the Wiki (which seems empty) would help.

[RavanJAltaie]

@benoit74 could you please check this file and tell me if we can solve it or we shall tag this as an upstream?
https://dev.library.kiwix.org/viewer#h-node_en_all_2024-05
(downloads don't work properly, issue direct to external links which doesn't open)

[benoit74]

Please be more specific, which downloads are not working?

[RavanJAltaie]

@benoit74

When you click on any download icon, you'll get a page like this:

[benoit74]

This is the XML file which we get on original website, what is the problem?

[Popolechien]

@benoit74 Well I for one would expect the XML file to download as on the source website, not to display.

[rgaudin]

Interesting test case for zimit2 since this Download is not declared client-side but server-side via a Content-Disposition attachment; header.
Given we don't store nor send (duh!) WARC-recorded headers, this wont work.
Example can be added to limitations documentation IMO

[benoit74]

Oh, ok I got it.

Yes, I confirm that links which leads to items expect to be "magically" downloaded do not work in zimit1, and they won't work for zimit2 either. I will open a ticket for this in openzim/warc2zim.

Note that it is still possible to download the file with a right click on the link (but I agree this is not super convenient).