openzim

Custom main page are either:

• A other legit article in main namespace
• A special page in user space with a long/strange title

We keep that title and that strange url in the second case and we scrape the "normal" main page anyway which really does not make much sense.

We should put the custom main page in place of the main page (typically with title "main page"). At first maybe do that an

The linux builder on the CI always try to checkout a branch.
But when we push a tag, the CI try to build the tag and there is no corresponding branch with the tag name.

It would be better to detect if we are building a tag or a branch and checkout the right git ref.
Failing of the CI here is not important as nothing is done with this CI (just building), and when tag is pushed with the a branc

Currently it seems to be a - character like at http://library.kiwix.org/khan-academy-videos_en_geometric-optics-ap-physics-2-khan-academy_2021-04/M/Description

In order to update ZIM files to the new tagging system, please make sure your scraper:

• is using the lastest zimwriterfs (1.3.5+ or :latest docker image). It supports new options and fixes important metadata (tags, favicon) and LZMA compression level.
• is including a _category:xxx tag (See [kiwix-tools#317](https://github.com/kiwix/kiwix-tools/issu