Hmm, I just found out why annotations are not implemented for imported entities, they have last_revision_id INT NOT NULL.
2024-07-17 19929, 2024
kellnerd[m]
Since an import does not have a revision, it is not possible to create a valid annotation row for it.
2024-07-17 19925, 2024
kellnerd[m]
So we either have to use a special revision ID 0 or NULL for these imported annotations (and handle the consequences in bb-site), or we just forbid importing annotations (since they are supposed to be CC0 and should not contain potentially copyrighted stuff anyway).
2024-07-17 19916, 2024
monkey[m] joined the channel
2024-07-17 19917, 2024
monkey[m]
That's unfortunate.
2024-07-17 19933, 2024
monkey[m]
I don't think we should mess with special cased revision ids
2024-07-17 19901, 2024
monkey[m]
Is there any way we could store the text content in a DB column instead?
2024-07-17 19927, 2024
monkey[m]
I mean, there are ways, I mean more "maybe we can investigate..."
2024-07-17 19959, 2024
monkey[m]
Pfff, no great ways to go about it
2024-07-17 19950, 2024
monkey[m]
I feel like annotations will be crucial to allow us at least to put the imported data that does not fit in the current schema somewhere
2024-07-17 19922, 2024
monkey[m]
The code changes to support NULL revision ID look like they would probably be minimal
2024-07-17 19955, 2024
kellnerd[m]
Yeah, currently we are hiding data which does not fit in a JSON column, exposing (parts of) it in the annotation makes sense.
2024-07-17 19926, 2024
kellnerd[m]
If we can have NULL values for annotation.last_revision_id that would be great, I already have the necessary ORM importer code ready.
2024-07-17 19935, 2024
kellnerd[m]
* not fit the schema in a
2024-07-17 19942, 2024
Jigen has quit
2024-07-17 19939, 2024
Jigen joined the channel
2024-07-17 19913, 2024
kellnerd[m]
ORM PR is up, do you want to look into the annotation schema change monkey, then I can focus on the actual importer changes which I wanted to do?
2024-07-17 19927, 2024
monkey[m]
Yes, let's do that
2024-07-17 19903, 2024
monkey[m]
Sorry for slow, there's construction happening in my apartment today, I keep having to stop
Several small changes across all three repos later importing and (editing and) approving entities with annotations is now working ⏳️
2024-07-17 19917, 2024
kellnerd[m] uploaded an image: (129KiB) < https://matrix.chatbrainz.org/_matrix/media/v3/download/matrix.org/yhFSvizmtxccbiofVcyHWPde/image.png >
2024-07-17 19923, 2024
kellnerd[m]
And I've come to the conclusion that it would be better to get rid of the "Edit & Approve" action in favor of just "Approve" and the standard "Edit" function of regular entities.
2024-07-17 19909, 2024
kellnerd[m]
The small convenience of that action does not justify the two extra code paths to convert from import data to editor form data and the separate form submission handler.
2024-07-17 19935, 2024
kellnerd[m]
Aside from the extra amount of code/maintenance, there are still disadvantages:... (full message at <https://matrix.chatbrainz.org/_matrix/media/v3/download/chatbrainz.org/TPXRvGPfWrpvFGNjCfRaKPmK>)
2024-07-17 19914, 2024
kellnerd[m]
Especially the third point is not worth solving at the expense of making the separate form submission handler even more complex rather than getting rid of it.
