kacerr/fuj-management
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
120 Commits 43 Branches 39 Tags
723152cdadd2ea2ec2fe3fc766672c57723a3f6f
Commit Graph

3 Commits

Author SHA1 Message Date
Jan Novak
8b3064ffab fix(go): default CacheDir to tmp/go to avoid Python collision
All checks were successful
Deploy to K8s / deploy (push) Successful in 9s
Details 
Previously both backends defaulted to `CacheDir=tmp` and used the
same cache keys (`attendance_regular`, `attendance_juniors`,
`payments_transactions`, `exceptions_dict`) but stored different
shapes: Python caches post-processed view-model tuples
(e.g. `(members, sorted_months)`), Go caches raw sheet rows.
Whichever backend wrote last poisoned the cache for the other,
producing `ValueError: too many values to unpack (expected 2,
got 68)` on Python's /adults after the Go side populated the
file with 68 raw CSV rows.

This breaks the M5.4 `make parity` workflow that requires both
backends running side-by-side.

Fix: change Go's default to `tmp/go` so the two cache trees
never overlap.  `CACHE_DIR` env var override still works.
`os.MkdirAll` already handles creating the new subdirectory on
first write.

Recovery for users with poisoned `tmp/`: hit /flush-cache on
the Python side once after pulling, then restart the Go server.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
 2026-05-07 23:06:34 +02:00
Jan Novak
6465e2a221 feat(go): IO layer behind interfaces (M4)
All checks were successful
Deploy to K8s / deploy (push) Successful in 11s
Details 
- io/attendance: CSV-over-public-URL client + Fake for adult/junior tabs
- io/drive: Drive v3 modifiedTime client + Fake
- io/sheets: Sheets v4 client (GetValues/AppendValues/BatchUpdateValues/
  WriteHeader/SortByDateColumn) + Fake with call-capture
- io/cache: Drive-modifiedTime-gated FileCache; two TTL knobs; atomic
  writes; generic Get[T]; Python-compatible JSON format; Flush()
- io/fio: Client interface backed by Fio REST API (apiClient) and HTML
  scraper (transparentClient); Fake; testdata fixtures
- membership/sources: NewSources wires attendance CSV + Sheets + cache
  into LoadAdults/LoadJuniors/LoadTransactions/LoadExceptions; Czech
  month parsing + merged-month maps
- banksync: SyncToSheets (SHA-256 dedup, optional sort) and
  InferPayments ([?] review prefix, dry-run) — tested with fakes
- cmd/fuj: sync and infer subcommands wired; fees and reconcile use
  real NewSources; go.mod gains google.golang.org/api + x/net
- gofumpt extra-rules applied across all packages; lint clean

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
 2026-05-07 01:05:59 +02:00
Jan Novak
cf0f176d3f feat: Go rewrite M1 — skeleton, tooling, and hello server 
Stand up the Go project alongside the Python backend so both run
independently during migration. `make web-go` builds and serves on :8080;
`make web-py` (alias: `make web`) keeps the Python side on :5001.

- go/: new module `fuj-management/go` (Go 1.26)
  - cmd/fuj: stdlib-flag dispatcher; `server` + `version` work,
    fees/reconcile/sync/infer stubbed for M2/M4
  - internal/config: env loader mirroring scripts/config.py
  - internal/logging: slog setup, level taken from config
  - internal/web: net/http ServeMux + request-timer middleware
  - build/Dockerfile: golang:1.26 → alpine:3 multi-stage image
  - .golangci.yml: govet, staticcheck, errcheck, gofumpt, unused
- Makefile: web→web-py alias; go-build/go-test/go-run/go-lint/web-go
- CI: parallel build-go job in .gitea/workflows/build.yaml (<tag>-go image)
- docs/plans/: M1 kickoff plan + progress tracker (M1 complete)
- .claude/settings.json: gofumpt + golangci-lint permissions

Gate: make go-build ✓  make go-lint ✓  make go-test ✓  curl :8080 ✓

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
 2026-05-04 12:05:46 +02:00