Reported from #general: "@Thermograph Vilnius, Lithuania" came back "I don't
track a city called 'Vilnius, Lithuania' yet." Vilnius has been in cities.json
the whole time, with country "Lithuania" sitting right there in the record. The
resolver matched the query against `name` alone, exact-or-prefix, so the comma
form could never match anything and the country field was never consulted.
The same message also showed the two neighbouring failures: "Vilinus" (one
transposition) fell through to the same refusal, which reads as "we've never
heard of Vilnius" rather than "you typed it wrong"; and "Tell me about weather
in lithuania" was quoted back verbatim as a city we don't track, which is what
made it look broken rather than merely unmatched.
So the resolver now goes, in descending order of confidence: exact name
honouring a "City, Country/Region" qualifier; the same ignoring an unrecognised
qualifier (better to grade Paris and name it in the reply than refuse over
"Paris, Wherever"); prefix; close-enough name for typos; and finally a city or
country named somewhere inside a sentence, so "weather in lithuania" lands on
Vilnius. Accents fold both ways, since people type Zurich and Sao Paulo far more
often than Zürich and São Paulo.
Free-text matching is whole-word and ignores names under four characters, so
ordinary chatter doesn't turn into a weather report -- "how are you today"
still resolves to nothing. And the unmatched reply no longer reads a whole
sentence back as a place name.
This fixes /grade identically; it shared the resolver and the same bug.
Claude-Session: https://claude.ai/code/session_015Z1ebLbhUxeZ9ozpNrVTCP
After a cell's archive advances, the first content request recomputes a 45yr payload cold. Add warm_cities.warm_content — idempotent, cache-only, bounded — that pre-populates the content derived-store under the same (kind,key,content_token) the routes read, wired into the leader-gated notifier loop (50 cities/30-min tick) so the ~1000-city set stays warm within hours of each daily advance.
The gateway bot and APScheduler were long-lived stateful I/O loops running
inside the async web app under a leader election. They move into a single Go
binary that owns ONLY that I/O -- websocket, RESUME, heartbeat, backoff, timers.
It owns no grading logic. Anything needing data calls back over a new
internal-only surface (/internal/discord/grade, /internal/jobs/*). Grading
depends on polars and the parquet cache; reimplementing it in Go would let the
bot's grades drift from the API's. The grade route returns gateway-ready JSON
and Go relays the bytes verbatim.
The binary ships in the backend image and runs as a second compose service off
the same tag, so the two ends of the /internal/* contract can never skew.
deploy.sh rolls daemon alongside backend -- without that the service would never
be created, since a single-service deploy uses --no-deps. It also probes the
image first and skips the daemon when rolling a tag that predates the binary:
infra tracks main while image tags are env-staged, so a host can legitimately be
asked to roll an older backend image, and creating the service anyway would
leave a container crash-looping on a missing binary.
replicas: 1 with order: stop-first replaces the leader election -- Discord
permits one gateway connection per bot token.
THERMOGRAPH_INTERNAL_TOKEN is optional: both ends derive it from
THERMOGRAPH_AUTH_SECRET via HMAC under a domain-separation label, so this needs
no new vault entry. The derivation is pinned to a shared cross-language test
vector asserted on both sides, so drift fails CI instead of 401ing every call.
Fail closed when neither secret is set.
Improvements over the Python: a close intended for RESUME uses 4000 rather than
1000 (Discord invalidates a session closed 1000, so the old default defeated its
own resume); MESSAGE_CREATE runs on a bounded worker pool; and a malformed HELLO
returns an error rather than a clean reconnect, which would otherwise reset
backoff and hot-loop against the gateway.
365 Python tests pass; Go build/vet/test -race clean; shellcheck 0 findings.
