Allow to add annotations to the job queue by ID

[marcospri]

For: #8727

Instead of writing a one off script or sql query to add the remaining annotations missing in annotation_slim we add a new option on the H admin pages to be able to add annotations by ID.

Testing

• First remove all annotation_slim rows

In make sql:

truncate cascades to table "annotation_metadata"

• Go to the admin pages, http://localhost:5000/admin/search

• Pick one annotation ID form the DB

select id from annotation limit 1;
                  id                  
--------------------------------------
 cdff42be-2fc0-11ef-ae06-37653ab647c1
(1 row)

• Use that annotation ID on the textarea, select "Annotation Slim" as the job type

• Process pending jobs locally, in make shell

from h.tasks.annotations import sync_annotation_slim
sync_annotation_slim.delay(limit=10)

• Check the annotations has correctly being sync-ed to annotation_slim:

select * from annotation_slim
postgres-# ;
 id  |                pubid                 |          created          |          updated          | deleted | moderated | shared | document_id | user_id | group_id 
-----+--------------------------------------+---------------------------+---------------------------+---------+-----------+--------+-------------+---------+----------
 572 | cdff42be-2fc0-11ef-ae06-37653ab647c1 | 2024-06-21 11:24:20.72035 | 2024-06-21 11:24:20.72035 | f       | f         | t      |          16 |     228 |       29
(1 row)

[marcospri]

Use the thousands separator.

New property alias for "by_ids"

[marcospri]

The first use case for this (complete the annotation_slim job) will use the full UUID format as the IDs come directly from the DB.

I think it's just nicer to support both formats anyway.

[mtomilov]

A minor thing but maybe it's worth passing the last two as keyword arguments like we do in add_annotations_by_ids.

[marcospri]

Good point, done.

[mtomilov]

Looks good