A bulk offer edit is the catalog mutation that rewrites hundreds of rows at
once, and it was the only one writing nothing to the staff activity log: 22 of
the 43 catalog actions logged, this one did not. The entry it now writes says
what changed, not just that something did, because the log has no undo of its
own and "bulk updated 200 offers" cannot answer the question it exists for.
Deleting offers had no inverse at all. Every removed row is now kept at delete
time and the caller gets a restoreId back, so an accidental multi-select is a
click rather than a hand-edit of the table. The undo toast covers the common
case; a RecentDeletionsPanel holds the same records so a delete noticed later is
still reachable. Three refusals guard it: an id that another offer has since
taken, a category that no longer exists (which would leave an offer that sells
nowhere and shows under no page), and a delete whose restore record cannot be
written — that one rolls back rather than deleting without a way back. Reading
the audit row FOR UPDATE is also what stops two restores of one deletion from
both inserting.
sendCatalogUpdate() overwrote hotel-status.json on every write, so "which
imports reached the hotel" was answerable for the last attempt only, and a
failure two imports ago was gone by the time anyone looked. That file is now
also appended to as a bounded 50-entry tail.
The tree route carried four copies of the same page-select-plus-counts
shaping, of which the BC branches had already drifted: one counted offers
through the VARCHAR-tolerant helper, the other inline and swallowing errors.
All of it is one readPages() now, and readFullTree sends both catalogs through
one depth computation instead of delegating normal to getTreeFlat while
computing BC here — a split that left two implementations behind one function
name. getTreeFlat is gone. The BC ancestor walk also went from 20 levels to 50,
matching getAncestors, so a deeply nested catalog no longer loses its
breadcrumb.
Bulk editing reaches the BC catalog, which previously had no way to edit or
duplicate offers in bulk. The catalog is part of the operation identity now, so
replaying one request key against the other catalog is not mistaken for the
same work.
Integration tests failed to import: the next/cache mock supplied only
revalidatePath, and catalog-totals calls unstable_cache at module scope.
Replace raw db.execute tuple casts with queryRows/rowsFrom/execResult/
affectedRows helpers from lib/db, drop redundant mysql2 casts on typed
query builders, and centralize per-test fakeForm into test/fake-form.
Update db mocks in tests so helpers resolve against mocked execute.
Database:
- Add missing indexes (users.credits, users_currency(type,amount),
users_settings.respects_received, camera_web.timestamp,
messenger_offline.user_id) via migrations 0020/0021
- Use partial .select() everywhere instead of SELECT * (tickets, users,
rooms, audit logs, catalog tree, polls, radio, password reset)
- Add queryPrepared/queryPreparedOne (server-side prepared statements)
and switch the login check to a prepared statement; drop dead
cache options from the pool config
- Raise total_users/total_rooms COUNT(*) cache TTL to 5m
Caching:
- Consolidate the three cache helpers (cached, redisCache, cachedQuery)
into a single memory-first implementation backed by Redis
- invalidateKey now clears the in-process cache as well as Redis
- Cache homepage sections, news list, and leaderboard tabs; share one
news_list cache key between homepage and news archive
- siteSettings: in-process cache with TTL so repeated getters no longer
pay a Redis round-trip per call
- Share a 10s poll cache across all radio SSE connections
- Normalize timestamps after cache reads (Redis JSON round-trip)
Assets:
- Enable AVIF/WebP via images.formats and remove unoptimized from news
covers and the homepage hero (149KB jpg) with proper sizes/priority
- Support ?format=webp|avif|png in the /imaging proxy via sharp
Other:
- Fix pnpm supply-chain minimumReleaseAge failures by excluding the
freshly-published packages (next 16.3.1, hookform resolvers 5.8.0,
resend 6.20.0)
- Remove unused before/after fields from housekeeping AuditEntry
- Replace Prisma client runtime with Drizzle ORM (zero Prisma engine/query engine in production)
- Add Prisma-compatible facade (@/lib/prisma-facade.ts) backed by Drizzle for backwards compatibility
- Runtime queries route through Drizzle ORM; @prisma/client is now devDependency (types only)
- Remove @prisma/adapter-mariadb dependency; delete prisma-pool.ts and types/prisma.ts
- New Drizzle schema layer: src/db/schema.ts (176 tables) and src/lib/db.ts (connection)
- Update README documenting the dual-layer ORM architecture
- Restore src/generated/ gitignore (build artifact for local type generation)
- 0 TypeScript errors, 583 tests passing
The facade intentionally uses `any` types to match the Prisma Client API surface,
allowing existing code to run unmodified while routing queries through Drizzle at runtime.
Align onlyBuiltDependencies with the workspace, fail fast without AUTH_SECRET in production, and delete catalog_items via VARCHAR-safe SQL so page deletes do not leave orphans.
Co-authored-by: Cursor <[email protected]>
Use raw SQL for counts/loads/creates/moves so Habbo DBs with VARCHAR page_id and no AUTO_INCREMENT still show and persist furni.
Co-authored-by: Cursor <[email protected]>
Seed root tabs from SSR, load the full tree without CLEAR_TREE races, coerce parent ids, and tolerate catalog_items page_id type mismatches so category pages actually appear.
Co-authored-by: Cursor <[email protected]>
openhands
Simo
remco