chore(api): drop the unused v_challenge_disbursements view#873
Merged
Conversation
rickyrombo
added a commit
that referenced
this pull request
May 29, 2026
#872) ## Problem `v_challenge_disbursements` exposes `sol_reward_disbursements` in the legacy `challenge_disbursements` shape and resolves `user_id` by **INNER JOIN**ing `users` on the recipient wallet. Any disbursement whose recipient doesn't resolve to a current user (wallet migrations, deactivated users, case mismatches) is silently dropped from the view. Most callers only ask the identity-independent question *"has this `(challenge_id, specifier)` been disbursed?"* — and a reward is disbursed exactly once per `(challenge_id, specifier)` on-chain, regardless of recipient. For those callers, a dropped row makes a **paid** reward look **unpaid**: - **`/users/{id}/challenges`** — showed `is_complete=true, is_disbursed=false`, so the client offered an already-paid reward as claimable; the claim then failed ("No rewards to claim" / on-chain "specifier already used"). This was the original bug. - **`/challenges/undisbursed`** (HTTP handler **and** the sqlc query used by the claim path) — same: a dropped disbursement makes a paid specifier claimable, and the claim fails on-chain. This is the exact failure mode migration `0204` documents. - **`/challenges/info`** — `weekly_pool_remaining` overstated by any dropped disbursement. - **coins redeem** — the double-redeem guard counts 0 and allows a second redemption. ## Fix All specifier/challenge-keyed callers now read `sol_reward_disbursements` directly by `(challenge_id, specifier)`, aligning them with how the reward manager dedupes disbursements on-chain. `is_disbursed` on `/users/{id}/challenges` is now keyed by specifier rather than the querying user. The one caller that genuinely needs the recipient `user_id` — **`/challenges/disbursements`** (admin listing) — inlines the `users` join, using `LOWER(users.wallet) = recipient_eth_address` per migration `0204` (the stale `ddl/views` file had the pre-`0204` case-sensitive join and, because `views/` applies after `migrations/`, was silently reverting the fix — so the deployed view was the broken one; inlining fixes that too). The view is now unused but **retained in this PR** so the binary can deploy before the view is dropped. The drop is a stacked follow-up: **#873**. ## Why reads, not a data migration `user_challenges` is indexer-derived; a manual reattribution would be reverted on reindex and doesn't address the read inconsistency. Keying reads by specifier is self-healing for past and future occurrences. (Attribution in `user_challenges` is intentionally left unchanged.) ## Tests - New `TestUserChallengesDisbursedToDifferentUser`: a disbursement attributed (by wallet) to a different user now reports `is_disbursed=true` for the completer. - Existing `TestGetChallengeDisbursements` covers the inlined join across all sorts/filters incl. `challenge_user_id`. - Full `./api` + `./api/dbv1` suites pass. ## Rollout 1. Merge + deploy this (reads no longer touch the view; view still present). 2. Then merge #873 to drop the view. 🤖 Generated with [Claude Code](https://claude.com/claude-code) --------- Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
rickyrombo
force-pushed
the
mp/drop-v-challenge-disbursements
branch
from
693f6c9 to
3665a0a
Compare
All API reads have been migrated off v_challenge_disbursements (#872), so the compatibility view has no remaining consumers. Drop it via migration 0212, remove the ddl/views definition, and update the schema dump. Deploy ordering: merge/migrate this only after #872's binary is rolled out, so no running code references the dropped view. Note: the removed ddl/views file still carried the pre-0204 case-sensitive join and, because views/ applies after migrations/, was silently reverting the 0204 LOWER(users.wallet) fix. Dropping it removes that footgun. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
rickyrombo
force-pushed
the
mp/drop-v-challenge-disbursements
branch
from
3665a0a to
4491cce
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Stacked on #872. Merge and migrate only after #872's binary is deployed.
#872 migrates every API read off
v_challenge_disbursements, leaving it unused. This PR removes it:0208—DROP VIEW IF EXISTS v_challenge_disbursements;ddl/views/v_challenge_disbursements.sqldefinitionsql/01_schema.sql(clean −23-line, view-only) and the migration trackerWhy split from #872
0208drops the view while the old binary still references it. Shipping the read migrations first (#872) means that by the time this migration runs, no running code touches the view — no missing-relation errors during a rolling deploy.Rollout
0208drops the view.Note
The deleted
ddl/viewsfile still had the pre-0204case-sensitive join, and sinceviews/applies aftermigrations/it was silently reverting0204'sLOWER(users.wallet)fix — so the deployed view was the broken one. Dropping it removes that footgun (and #872 inlines the correctLOWERjoin where the resolution is still needed).🤖 Generated with Claude Code