Mark Keysets as Deleted.

[KvngMikey]

Fixes #731

Summary

This PR fixes the wallet behavior when a keyset disappears from a mint's /keysets response.

Changes

• Updated load_mint_keysets to mark missing keysets as deleted.
• Added update_keyset_active in crud.py to use named SQL parameters for SQLAlchemy async compatibility.
• Verified with pytest tests/wallet/test_wallet.py -k keyset (all keyset tests now pass).

[callebtc]

LGTM but one outstanding question regarding the removed line await self.load_keysets_from_db()

what's the justification for that?

[KvngMikey]

LGTM but one outstanding question regarding the removed line await self.load_keysets_from_db()

what's the justification for that?

@callebtc I replaced "await self.load_keysets_from_db()" because we now build self.keysets inline with the already-fetched list (keysets_in_db) filtered by k.unit == self.unit and k.active.
That list is fresh (we re-fetch after marking inactive), so the old helper’s extra query + loop is redundant.
The trace message is still printed two lines below, so no observability is lost.

[callebtc]

Hi @KvngMikey, sorry for getting back to you late. We have a crud called update_keyset that supports setting the active flag (latest main also can update the fee, worth a merge of main into this branch). Why not use that directly?

[KvngMikey]

Hi @KvngMikey, sorry for getting back to you late. We have a crud called update_keyset that supports setting the active flag (latest main also can update the fee, worth a merge of main into this branch). Why not use that directly?

hi @callebtc , thanks for checking my work, I have addressed your request now.

[callebtc]

we talked about this off band: there should probably be a boolean deleted column in the keyset table of the wallet and if we flip it to True, the keyset shouldn't be loaded from the db, when we use load_keysets(). wanna try that?

[KvngMikey]

we talked about this off band: there should probably be a boolean deleted column in the keyset table of the wallet and if we flip it to True, the keyset shouldn't be loaded from the db, when we use load_keysets(). wanna try that?

yes, i remember and i'm working on it.

[KvngMikey]

@callebtc PR updated, ready for re-review, thank you.

[KvngMikey]

question:
when this PR goes in, would we need to update NUT01 to show that keysets can now be marked as deleted ?

cc @callebtc @a1denvalu3

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 75.25%. Comparing base (03f6f56) to head (57394e4).
⚠️ Report is 44 commits behind head on main.
✅ All tests successful. No failed tests found.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #797      +/-   ##
==========================================
+ Coverage   74.40%   75.25%   +0.85%     
==========================================
  Files          99       99              
  Lines       11599    11723     +124     
==========================================
+ Hits         8630     8822     +192     
+ Misses       2969     2901      -68     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[callebtc]

question: when this PR goes in, would we need to update NUT01 to show that keysets can now be marked as deleted ?

cc @callebtc @a1denvalu3

nope, this is just for the wallet and not related to the API. the mint just deletes the keyset

[KvngMikey]

question: when this PR goes in, would we need to update NUT01 to show that keysets can now be marked as deleted ?
cc @callebtc @a1denvalu3

nope, this is just for the wallet and not related to the API. the mint just deletes the keyset

great, thank you !

[KvngMikey]

@callebtc - this has also been rebased and is ready.

cc @a1denvalu3

[Copilot]

Copilot was unable to review this pull request because the user who requested the review is ineligible. To be eligible to request a review, you need a paid Copilot license, or your organization must enable Copilot code review.