chore: Data migration script for overlay edit_type field and corresponding pghistory event

[adekoder]

Description:
Migrate overlay edit_type values and event labels, Updates existing overlay records and pghistory event tables to use new field values:

Edit Type

• add -> additional,
• remove -> ignored

PGH Label

• maintainers.add -> maintainer.add,
• maintainers.remove -> maintainer.delete,
• package.remove → package.ignore
• package.add→ package.restore

Notes

• Batched updates (1000 records)
• Temporarily disables pghistory triggers on event tables, we need it disabled for the script to run which the script handles and also re-enabled the trigger once done.

[fricklerhandwerk]

Wasn't the previous version doing just that?

[adekoder]

it started failing, for some weird reason, and this fixed it

[fricklerhandwerk]

I think it's fine to add the new checks to the existing test. Maybe update the name. The setup takes quite long, also repeating test cases is very bad for readability.

[adekoder]

This is not a repeated testcase, this testcase it testing when there is a delete the activity log reflect that it was deleted based on the new event label conventions.

[fricklerhandwerk]

But up to this line the test is identical to the previous one. Why not just continue that one?

[fricklerhandwerk]

I don't think it makes much sense moving that, since we're only using it in that one place

[adekoder]

but fixtures should be in conftests

[fricklerhandwerk]

Yeah but not for those we use exactly once for a very particular test. I'm fine moving it when there's a need :)

[fricklerhandwerk]

The tests failed due to running out of space. We may need to increase the VM disk size:

nix-security-tracker/nix/tests/default.nix

Lines 9 to 15 in 6f3b269

	defaults = {
	documentation.enable = lib.mkDefault false;
	virtualisation = {
	memorySize = 2048;
	cores = 2;
	};

[adekoder]

The tests failed due to running out of space. We may need to increase the VM disk size:

nix-security-tracker/nix/tests/default.nix

Lines 9 to 15 in 6f3b269

	defaults = {
	documentation.enable = lib.mkDefault false;
	virtualisation = {
	memorySize = 2048;
	cores = 2;
	};

Ah, Thank you , i will fix it.

[Erethon]

I haven't looked at the PR in full yet, but I think this is wrong as it doesn't take into account edit _type which adds semantics for what it means to add/remove a maintainer.

[adekoder]

sorry, not sure i understand your comment, please clarify more ?

[fricklerhandwerk]

I guess we can close this as duplicate of #945