Identify recipes whose product is still in /dev

[Popolechien]

It is common practice to first run a recipe to deliver in /dev, but then it probably is true that without reminder some will be forgotten then in this limbo. If a recipe is successful and moved to the appropriate directory, the initial zim file is not automatically removed from /dev either. Bottom line: /dev should either be cleaned after a recipe's target is moved elsewhere, or there should be a way to filter where zims are stored by which recipes.

[rgaudin]

Agrees with the issue but don't know how to fix it. The main problem with /dev is that there are multiple levels of validation:

• purely technical (what zimcheck says about respect of the ZIM format)
• what the scraper developer considers implemented
• what we want users to be offered with.

We have no procedure for non-technical QA and how to decide if this is good to go or not ; not even talking about who should assess this.

I think that now that we have settled on creating the CMS, this should be post-zimfarm. zimfarm should inform about some QA indicators (we have zimcheck already, we could have a draft/ready flag as well) but publishing or not is not a zimfarm decision. With the no-update within a monh limitation that we have, a testing step in CMS seems mandatory for new scrapers and maybe major scraper updates.

[kelson42]

This is a CMS problem as the publishing of new content won't be automatic anymore like today. Actually this is a non-problem to me.

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be now be reviewed manually. Thank you for your contributions.

[rgaudin]

Closing as this is not actionable and we have a feature item for removing obsolete files.