Add columbia ordering to management command

[quevon24]

This PR adds the functionality to update_opinions_order command to fill ordering_key field for columbia opinions.

It will search for clusters that include columbia in their source (Z, ZL, ZU, ZLU, etc), clusters with more than one opinion, and whose opinion have a file assigned in the local_path field.

The command to run it:

docker exec -it cl-django python /opt/courtlistener/manage.py update_opinions_order --action sort-columbia --xml-dir /opt/courtlistener/cl/assets/columbia

We need to place the xml files in a directory and then pass the path to the command, which will allow us to read the xml assigned to the opinion, this is essential because we rely on the original xml to infer the correct order because some opinions were created in the wrong order due the old columbia importer.

To test it you can clone this clusters:

docker exec -it cl-django python /opt/courtlistener/manage.py clone_from_cl --type search.OpinionCluster --id 9079419 4226317 8279828 8237637 8041148 4233235 4223647 8041161 8040857 8040920

And paste this files: test_xml.zip in your local environment in this location: /opt/courtlistener/cl/assets/columbia

And then run the command exactly as i put it above.

[quevon24]

This is the list of ids of the columbia clusters that have more than one opinion and have columbia in their source:

_SELECT_search_opinioncluster_id_FROM_search_opinioncluster_LEFT_202408260949.csv

This is the query I used to get the list:

SELECT "search_opinioncluster"."id" FROM "search_opinioncluster" LEFT OUTER JOIN "search_opinion" ON ("search_opinioncluster"."id" = "search_opinion"."cluster_id") WHERE (EXISTS(SELECT 1 AS "a" FROM "search_opinion" U0 WHERE (U0."cluster_id" = ("search_opinioncluster"."id") AND U0."local_path" > '' AND U0."local_path" IS NOT NULL) LIMIT 1) AND "search_opinioncluster"."source" IN ('Z', 'ZA', 'ZD', 'ZC', 'ZH', 'ZLC', 'ZLR', 'ZLCR', 'ZR', 'ZCR', 'ZL', 'ZM', 'ZQ', 'ZU', 'ZLU', 'ZDU', 'ZLRU', 'ZLCRU', 'ZCU', 'ZMU', 'ZRU', 'ZLCU')) GROUP BY "search_opinioncluster"."id" HAVING COUNT("search_opinion"."id") > 1 ORDER BY "search_opinioncluster"."id" asc

[flooie]

Thanks @quevon24 this looks good to me -

[sentry-io]

Suspect Issues

This pull request was deployed and Sentry observed the following issues:

• ‼️ FieldError: Cannot resolve keyword 'opinions_count' into field. Choices are: arguments, attorneys, blocked, c... cl.search.models in filter View Issue

_{Did you find this useful? React with a 👍 or 👎}