Root cause of the Grafana "readdirent /etc/grafana/dashboards: no such file or
directory": the CI runner checks out into an ephemeral act workspace that is
removed after the job, so binding the compose config files straight from it
dangles the mounts in the long-lived containers (verified the act source dir is
emptied after the job). caddy/otelcol/prometheus/tempo read their config once at
startup so they survive, but would break on a restart — same latent bug.
Fix (mirrors ../galaxy-game's $HOME/.galaxy-dev/monitoring): the deploy job seeds
the config dirs to a stable $HOME/.scrabble-deploy and the compose binds them via
${SCRABBLE_CONFIG_DIR:-.} (local runs keep "."). Documented in the compose header,
deploy/README.md and the ci.yaml step.
- deploy/docker-compose.yml: mount the provisioned dashboards at
/etc/grafana/dashboards, not /var/lib/grafana/dashboards — the grafana-data
volume mounts over the latter and shadows the nested bind, so the provider
logged "readdirent /var/lib/grafana/dashboards: no such file or directory".
dashboards.yaml provider path updated to match.
- Connector telemetry stays OTLP. The VPN sidecar's netns reaches the collector's
internal IP fine (connected route, off-tunnel), but the sidecar's DNS hijacks
name resolution: AWG_CONF must NOT carry a DNS= directive, else otelcol won't
resolve ("produced zero addresses"). Without DNS= the netns uses Docker's
resolver (resolves both otelcol and api.telegram.org). Documented in
deploy/README.md (AWG_CONF row + wiring note), ARCHITECTURE §13, compose comment.
- PLAN.md: new Stage 17 "Test-contour verification & defect fixes" (exercise the
deployed contour end-to-end and fix what it surfaces — connector liveness check,
path-conditional CI); the former prod-deploy stage becomes Stage 18.
- Renumber every "Stage 17" prod-deploy reference to "Stage 18" across docs,
compose, Caddyfile, ci.yaml and CLAUDE.md; the post-Stage-14 split range is now
"Stages 15–18".
- bot.New now selects Telegram's test environment with the library's native
tgbot.UseTestEnvironment() instead of a token += "/test" hack (functionally
identical URL /bot<token>/test/METHOD, but idiomatic) + a bot test asserting
the getMe path for both test and prod.
- ci.yaml pins TELEGRAM_TEST_ENV=true for the test contour (it IS the test
environment) instead of a TEST_TELEGRAM_TEST_ENV variable: removes the
confusing double-TEST, telegram-specific, prefixed operator knob and the
secret-vs-variable footgun. Prod (Stage 17) leaves it false.
- deploy/README.md + PLAN.md updated.
- deploy/README.md documents the services, how to run it locally and in CI, and
every variable: required (the four :? ones + ≥1 bot token) and optional with
defaults, marked secret-vs-variable and with the TEST_/PROD_ Gitea mapping;
plus the fixed internal wiring and the host-side setup.
- ci.yaml maps the remaining POSTGRES_DB/USER, DICT_VERSION and LOG_LEVEL (unset
renders empty -> the compose ":-" defaults apply), so every documented var is
per-contour overridable.
- .env.example points at the README for the full reference.
- backend + gateway multi-stage distroless Dockerfiles; the gateway embeds and
serves the SPA at / and /telegram/ via go:embed (committed dist placeholder,
real build baked in by the image's node stage)
- deploy/docker-compose.yml: backend + gateway + Postgres + Telegram connector
(VPN sidecar) + OTel Collector + Prometheus (15d) + Tempo (72h) + Grafana,
fronted by a caddy owning a single /_gm Basic-Auth (admin console + Grafana
subpath); inter-service on a private network, only caddy on the edge network
- new metrics: backend accounts_created_total{kind} (robots excluded) and an
in-memory gateway active_users{window=24h,7d} gauge
- CI: single .gitea/workflows/ci.yaml (unit/integration/ui + a gated test-contour
deploy) on the new feature/* -> development -> master branch model; the old
go-unit/integration/ui-test workflows are folded in; the connector-scoped
compose is retired (superseded by deploy/)
- docs: ARCHITECTURE §11/§12/§13, root + gateway READMEs, CLAUDE.md branching,
PLAN.md (stage 16 done + refinements + Stage 17 forward-notes)
Service-agnostic refinement of the owner's idea: the sign-in service returns a
set of supported game languages with the user identity, and the lobby gates the
New Game variant choice by it (en -> English; ru -> Russian + Эрудит).
- Connector hosts two bots in one container (one per service language, each its
own token + game channel; the same telegram_id spans both). ValidateInitData
tries each token and returns the validating bot's service_language +
supported_languages. Per-language config (TELEGRAM_BOT_TOKEN_EN/_RU, channels).
- supported_languages rides the Session (fbs, session-scoped, not persisted); the
UI offers only the matching variants on New Game — gating only the START of a
new game (auto-match + friend invite), not accept/open/play; backend does not
enforce.
- service_language persisted (accounts.service_language, migration 00010, written
every login, last-login-wins) and routes the user-facing Notify push back
through the right bot (push-target coalesces with preferred_language).
- Admin SendToUser/SendToGameChannel gain an operator-chosen language selector in
the console (unrelated to ValidateInitData).
- Non-Telegram logins carry the gateway default set
(GATEWAY_DEFAULT_SUPPORTED_LANGUAGES, all variants).
Wire (committed regen): ValidateInitDataResponse +service_language
+supported_languages; Session +supported_languages; SendToUser/SendToGameChannel
+language. Docs (ARCHITECTURE/FUNCTIONAL/_ru/READMEs) + PLAN updated; stage marked done.
- backend/go.mod pins gitea.iliadenisov.ru/developer/scrabble-solver v1.0.0; the engine's
imports use the published module path; go.work drops the solver replace (GOPRIVATE fetches
it directly from Gitea). The solver's wordlist/dictdawg are now public packages.
- CI (go-unit, integration): drop the solver sibling-clone, set GOPRIVATE, and download the
dictionary DAWG release artifact (scrabble-dawg-<DICT_VERSION>.tar.gz from the new
scrabble-dictionary repo) for BACKEND_DICT_DIR.
- Docs: ARCHITECTURE §5/§11/§13/§14 + backend/README updated to the published-module +
release-artifact model. PLAN.md re-scoped Stage 14 to the split and added Stages 15 (deploy
infra & test contour), 16 (prod contour), 17 (dual Telegram bots); TODO-1/TODO-2 marked done.
Live play now exchanges per-variant alphabet indices instead of concrete
letters (rack out; submit-play, evaluate, exchange, word-check in). The client
caches each variant's (index, letter, value) table behind
StateRequest.include_alphabet and renders the rack and blank chooser from it,
dropping the hardcoded value/alphabet tables. History, the durable journal and
GCG stay decoded concrete characters (ARCHITECTURE §9.1, unchanged).
- pkg/fbs: new AlphabetEntry + PlayTile; StateView.rack -> [ubyte] + alphabet;
StateRequest.include_alphabet; SubmitPlay/Eval tiles -> [PlayTile];
Exchange tiles + CheckWord word -> [ubyte] (committed Go + TS regenerated).
- engine: AlphabetTable + a cached per-variant codec (LetterForIndex/EncodeRack/
DecodeTiles/DecodeWord) + BlankIndex sentinel; Go parity test.
- backend server edge maps index<->letter (new thin game.Service.GameVariant);
game.Service domain methods, engine.Game and the robot keep one letter-based
play path. The gateway forwards indices verbatim (no alphabet table).
- ui: lib/alphabet.ts in-memory cache; codec encodes/decodes indices; premiums.ts
is geometry-only; the mock seeds a fixture table; the UI normalises display to
upper case (codec + cache), leaving placement/board/checkword unchanged.
Parity moved to the Go engine.AlphabetTable test; premiums.ts loses its value
tables. Discharges TODO-4.
New platform/telegram connector (own container, bot token only there):
- go-telegram/bot long-poll loop: /start deep-links + Mini App launch button.
- gRPC API pkg/proto/telegram/v1 (Telegram service): ValidateInitData, Notify
(renders a localized message + deep-link button), SendToUser/SendToGameChannel
(admin, wired in Stage 10). Generic methods are platform-agnostic (external_id).
- Bot API base override for Telegram's test environment; Dockerfile + compose
(VPN sidecar, no public ingress); README.
Gateway:
- initData validation relocated from the gateway into the connector; the gateway
calls ValidateInitData over gRPC (GATEWAY_CONNECTOR_ADDR), drops the bot token,
and deletes internal/auth.
- Out-of-app push: runPushPump routes events whose recipient has no live in-app
stream to connector.Notify, gated by /internal/push-target + the in-app-only
flag (race-free de-dup); HasSubscribers added to the push hub.
Backend:
- Migration 00007 accounts.notifications_in_app_only (default true) + jetgen.
- ProvisionTelegram seeds a new account's language/display name from the launch
fields; IdentityExternalID reverse lookup; /internal/push-target handler.
UI:
- Telegram Mini App launch: detect initData, apply themeParams, authTelegram,
route the deep-link start_param (g/i/f); /telegram/ guard redirects outside
Telegram. Vite relative base + telegram-web-app.js. In-app-only profile toggle;
share-to-Telegram link for a friend code. Vitest + Playwright coverage.
Wire/docs/CI: fbs Profile/UpdateProfileRequest gain notifications_in_app_only
(Go + TS); go.work uses ./platform/telegram; go-unit.yaml covers it; PLAN,
ARCHITECTURE, FUNCTIONAL (+ru), UI_DESIGN, READMEs updated.
Lock the polish behaviours so a future edit surfaces as a failing test:
- backend: UpdateProfile now rejects a bad name layout, an away window over 12h, and
a malformed offset timezone (confirming it wires the Stage 8 validators); a new
integration test accepts and resolves a "+03:00" offset timezone.
- e2e (mock): the lobby notification badge count, the play-with-friends required
game type + invitation send, the in-game add-to-friends flipping to a disabled
"request sent", the profile-edit invalid-name Save guard, and the chat send/nudge
icon buttons.
Third owner-review pass (iPhone):
- Modals (and the chat) size their backdrop to window.visualViewport, so they stay
fully above the software keyboard (dvh alone left the sheet partly behind it).
- On the owner's call, every profile / new-game picker is a native <select> for
consistent cross-platform behaviour: the away window returns to hour + 10-minute
selects (which also avoids the iOS time-wheel "clear" button), alongside the offset
timezone and the game-type / move-time / hints selects. Native time/wheel inputs
render differently per OS and cannot be forced to match.
- New-game "play with friends" has no preselected game type — an explicit, required
pick (empty placeholder); Send invitation stays disabled until both a type and a
friend are chosen. A smart default (from play history / language) is TODO-6.
Second owner-review pass (iPhone simulator):
- Chat (and the modal) are sized in dvh so they shrink above the software keyboard,
keeping the start of the conversation on screen instead of pushed off the top.
- The profile away window returns to a native <input type="time" step="600"> (the iOS
wheel with 10-minute steps) instead of separate dropdowns; the timezone stays a
native offset <select>.
- A finished game reserves the rack's height (min-height) so the footer no longer
collapses when the final rack is empty — no layout jump versus an active game.
- New-game "play with friends" is made compact: a searchable, bounded-scroll friend
list, the game-type / move-time / hints controls as native selects in one row
(labels above), and Send invitation pinned at the bottom — it scales to many friends.
Owner-review follow-up on the Stage 8 branch:
- Friend code is copyable (📋 + toast). The lobby notification badge is fixed —
it had inherited the hamburger-bar style — into a proper round count dot.
- Safari: min-width:0 on flex text inputs (friend code, profile, chat) so they
shrink instead of pushing the adjacent button off-screen.
- Profile editing is validated on both the UI and the backend: display-name format
(letters joined by single space/./_ separators, no leading/trailing/adjacent
separators, <=32 runes), a UTC-offset timezone picker (account.ResolveZone parses
±HH:MM or a legacy IANA name), a 10-minute away grid capped at 12h (wrap-aware),
and email format; Save is disabled and invalid fields red-bordered until valid.
Language stays in Settings.
- In a game, an "add to friends" menu item flips to a disabled "request sent"; chat
send/nudge became ⬆️/🛎️ icon buttons.
- A finished game drops its last-word highlight, hides Check word / Drop game,
disables zoom, and draws an inert (greyed) footer instead of hiding it.
Tests: account validators (name/away/zone), UI profileValidation, e2e for the
finished-game footer/menu and the copy control. Docs (PLAN, ARCHITECTURE,
FUNCTIONAL +ru, UI_DESIGN) updated for the display-name rule, UTC-offset timezone
and the 12h away window.
Wire the deferred Stage 7 surfaces end-to-end (UI -> gateway transcode ->
backend REST -> existing domain services): friends (incl. one-time friend
codes), per-user blocks, friend-game invitations, profile editing + email
binding, the statistics screen, and the in-game history + GCG export.
Friends gain two add paths (interview decision, a deliberate plan change):
one-time 6-digit codes (friend_codes table, 12h TTL, single-use, rate-limited
redeem); and play-gated requests (shared game required) where an explicit
decline is permanent, an ignored request lapses after 30 days, and a code
bypasses a decline. Migration 00006 widens friendships_status_chk and adds
friend_codes.
Lobby notification badge is poll + push: a new generic `notify` event drives
it live; the client polls on open/focus. Language stays a single Settings
control that writes through to the durable account's preferred_language. GCG
export is finished-only (game.ErrGameActive) and shares/downloads the .gcg file.
Tests: backend unit + inttest (friend gate/decline/code, ListInvitations,
GetStats, GCG gate), gateway transcode round-trips + notify constructor, UI
vitest (codecs, win-rate, share choice) + Playwright social specs. Docs: PLAN
(Stage 8 done + refinements + TODO-5), ARCHITECTURE, FUNCTIONAL(+ru), UI_DESIGN,
TESTING, module READMEs.
Lock the polish branch's behaviour so a future UI edit surfaces as a failing
assertion to re-agree or fix.
Unit (vitest, node env):
- placement: recallIndex, cellOccupied/isBlankSlot, non-linear direction, the
single-tile submit default, and placementFromHint blank-fallback / rack-exhausted.
- banner: the marquee scroll-cycle repeat-then-advance, stop(), root-relative and
multiple links.
- client.GatewayError. Extract the check-word constraints out of Game.svelte into a
pure lib/checkword.ts (sanitize + canCheck) and cover them.
E2E (playwright mock, Chromium + WebKit):
- commit via the 🏁 control, history slide-down + close, the exchange dialog,
check-word input sanitising + verdict, resign-to-finished, and the Settings
board-label mode changing the on-board labels.
Add a webkit project to the Playwright config so the hermetic mock-mode specs run
in both Chromium and WebKit, and install both browsers in CI. WebKit's Debian build
runs headless without extra host system libraries (verified locally: smoke + zoom
pass in webkit); the workflow comment records the one-time host install-deps fallback
if a runner ever lacks a library. Desktop WebKit does not reproduce iOS Safari's text
auto-inflation, so the app.css text-size-adjust guard stays outside e2e coverage.
The Playwright MCP writes page snapshots/screenshots into a .playwright-mcp/
directory while inspecting the UI; keep that scratch output out of the tree.
Drive the focus-centring with requestAnimationFrame across the ~0.25s width
transition instead of a single scrollTo after transitionend. The board now stays
locked on the placed cell as it grows, removing the visible 'centre top-left, then
correct' double motion.
- AdBanner: move the side inset onto the scrolling track so the long message
scrolls to its very end; pin body text-size-adjust:100% so iOS/Safari stops
inflating the long marquee text.
- Game: do not zoom on drag start (the player may change their mind) — zoom and
centre happen on drop, in attemptPlace; a stray tap on an occupied cell no
longer cancels the rack selection (wait for an empty cell).
- Board: centre the focus cell after the zoom width transition finishes (was
clamping to top-left mid-transition); compute the cell from the rendered
scrollWidth.
- item 5: move container-type to the zoom-scaled .scaler so cqw labels grow WITH the board (magnifying-glass zoom); new e2e measures the font grows ~1.85x
- item 8: confirm popovers anchor to the trigger's right edge (no longer run off-screen)
- item 9: last-word flash runs 2 cycles then settles to normal (was infinite)
- nav bar grows ONLY in game (other screens: minimal nav, content fills); tab bar always bottom
- tab bar: tighter icon/label spacing, bigger icons, hint badge on the icon corner
- board zoom reworked to width-based (real native scroll, fixes Safari/Chrome) + constant cqw labels; pinch & swipe-to-history dropped (conflict), double-tap kept, history via menu
- beginner bonus labels shrunk to fit cells
- Draw opens exchange directly (no confirm); confirm popovers restyled like the hamburger dropdown (vertical); removed the floating direction toggle
- pending tiles darker bg (no outline); last-word dark-tile highlight (static / 1s flash)
- check button disabled for <2/>15 chars, already-checked, or 5s cooldown
- global user-select:none (inputs exempt); docs updated; TODO-4 alphabet-on-wire
System libs are now provisioned once on the runner host (install-deps); the job
downloads the browser into the runner cache and runs the smoke strictly (no
--with-deps, no sudo, no continue-on-error; timeouts kept as a hang guard).
The Gitea runner is an act host executor without apt privileges, so
'playwright install --with-deps chromium' (apt-get) hung ~51m then failed. Drop
--with-deps, bound the browser-install + e2e steps with timeouts and mark them
continue-on-error so a runner lacking GUI libs can't block the gate. The strict
gate stays check/unit/build/size; the e2e smoke remains a hard local pre-push check.
Ilia Denisov
developer