thermograph/frontend
Emi Griffith 9eecfc8eef
All checks were successful
PR build (required check) / changes (pull_request) Successful in 6s
secrets-guard / encrypted (pull_request) Successful in 7s
PR build (required check) / build-backend (pull_request) Has been skipped
shell-lint / shellcheck (pull_request) Successful in 7s
PR build (required check) / validate-observability (pull_request) Has been skipped
PR build (required check) / build-frontend (pull_request) Successful in 1m0s
PR build (required check) / gate (pull_request) Successful in 2s
frontend: rewrite the SSR content service in Go
Ports frontend/ (Jinja2/FastAPI, ~1180 LOC) to Go with html/template.
No climate math, no DB, no auth here -- every route fetches from the
backend's /content/* API, so this is I/O-bound glue with no hard-porting
wall; the risk was always in reproducing the rendering exactly, not the
language.

Verified with a golden-HTML diff, not just unit tests: both the Python
original and the Go rewrite were run against the same committed fixtures
(frontend/tests/fixtures/) and every one of the 11 routes compared
byte-for-byte. The only surviving differences after that process are
insignificant inter-tag whitespace and one attribute where Go's stricter
escaper HTML-encodes an apostrophe Jinja left literal (functionally
identical in every browser) -- confirmed programmatically by normalizing
whitespace and unescaping before diffing, not by eyeballing.

That process caught defects unit tests alone would have missed, because
map[string]any has no compile-time field check:

- Render-context keys were snake_case throughout (content.py's Jinja
  convention, ported verbatim) while the templates -- written
  independently -- read PascalCase fields. A missing map key doesn't
  error in html/template, it silently renders empty, so this was invisible
  in every status code and every "it built" signal: title, meta
  description, canonical URL, OpenGraph tags, the homepage's entire ranked
  list, and the brand-tag/nav-active state were all blank across every
  page. Fixed by renaming every key to match each template's own header
  comment (the authoritative per-page field contract) and, where an
  API struct's exported fields already matched what a template needed
  (contentapi.CityInfo, Crumb, HomeRanked, HubCountry, ...), passing the
  struct straight through instead of hand-rewrapping it in a map --
  removes a whole layer of future drift risk, not just this instance of it.
- Three pages 500'd outright: `.ToolHref` needed a fully-composed href
  string, not the bare "lat,lon" fragment the handlers were building; the
  all-time-records table needed the raw contentapi.AllTimeRecords struct,
  not a re-wrapped map.
- JSON-LD was being double-encoded: `<script type="application/ld+json">`
  is JAVASCRIPT context to html/template's contextual escaper regardless
  of the script's `type` attribute, so a template.HTML-typed value placed
  there gets re-escaped as a quoted JS string instead of emitted raw --
  the entire structured-data payload shipped as a JSON string containing
  JSON, which no crawler would parse as the intended object. Needed
  template.JS instead, the type that actually means "trusted JS source."
  The glossary term page's JSON-LD was simply never built at all (the
  Jinja original assembled it inline in the template rather than through
  content.py's context dict, and that got lost in translation) -- added.
- html/template silently strips literal HTML comments AND JavaScript
  comments from the parsed output (verified in isolation, zero template
  actions involved) -- confirmed as real engine behavior, not a bug in
  either port, so both need a FuncMap function returning template.HTML /
  template.JS respectively to survive parsing rather than a literal
  `<!-- -->` or `//` in the template source.

Packaging: multi-stage Go build, final image alpine (not distroless -- the
Swarm stack's env-entrypoint.sh shim needs bash), 187MB -> 22.6MB. Two
defects caught before they reached a host:
- The Swarm stack overrides `entrypoint:` with no `command:`, which drops
  the image's own CMD entirely (Docker/Swarm semantics, not merged) --
  env-entrypoint.sh then fell through to its hardcoded `exec uvicorn
  app:app` fallback, which doesn't exist in this image. Every deploy
  would have exited 127. Fixed with an explicit `command:` on the stack's
  frontend service, and corrected the shim's stale comment claiming CMD
  passes through automatically.
- `COPY --chown=thermograph` resolves the group by NAME at copy time;
  Alpine's `adduser -S` with no `-G` doesn't create a same-named group, so
  the classic (non-BuildKit) Docker builder -- which this CI runner falls
  back to, since it installs plain `docker.io` with no buildx plugin --
  failed outright. Fixed with an explicit group and numeric --chown.

Verification: go build/vet/test -race clean across all packages; the
Docker image builds and passes its embedded go test step under both
BuildKit and the classic builder; shellcheck 0 findings on the one script
touched; rebased onto current main (the ERA5 lake stack landed on both
main and dev during this work -- confirmed additive, no overlap with
frontend/daemon).
2026-07-23 17:51:31 -07:00
..
content Subtree-merge thermograph-frontend (origin/main) into frontend/ 2026-07-22 22:01:11 -07:00
scripts shell: add shellcheck CI guard and drive the tree to zero findings (#19) 2026-07-23 22:26:05 +00:00
server frontend: rewrite the SSR content service in Go 2026-07-23 17:51:31 -07:00
static Subtree-merge thermograph-frontend (origin/main) into frontend/ 2026-07-22 22:01:11 -07:00
templates Subtree-merge thermograph-frontend (origin/main) into frontend/ 2026-07-22 22:01:11 -07:00
tests Subtree-sync frontend to split main 510d94d (dev->main promotion: reconciled hardening, two-tier suite); subtree workflow copy stays deleted (CI lives at root) 2026-07-22 22:37:29 -07:00
tools Subtree-merge thermograph-frontend (origin/main) into frontend/ 2026-07-22 22:01:11 -07:00
.dockerignore frontend: rewrite the SSR content service in Go 2026-07-23 17:51:31 -07:00
.gitignore Subtree-sync frontend to split main 510d94d (dev->main promotion: reconciled hardening, two-tier suite); subtree workflow copy stays deleted (CI lives at root) 2026-07-22 22:37:29 -07:00
api_client.py Subtree-merge thermograph-frontend (origin/main) into frontend/ 2026-07-22 22:01:11 -07:00
app.py Subtree-merge thermograph-frontend (origin/main) into frontend/ 2026-07-22 22:01:11 -07:00
CLAUDE.md Subtree-merge thermograph-frontend (origin/main) into frontend/ 2026-07-22 22:01:11 -07:00
content.py Subtree-merge thermograph-frontend (origin/main) into frontend/ 2026-07-22 22:01:11 -07:00
content_loader.py Subtree-merge thermograph-frontend (origin/main) into frontend/ 2026-07-22 22:01:11 -07:00
DESIGN.md Subtree-merge thermograph-frontend (origin/main) into frontend/ 2026-07-22 22:01:11 -07:00
docker-compose.test.yml Subtree-sync frontend to split main 510d94d (dev->main promotion: reconciled hardening, two-tier suite); subtree workflow copy stays deleted (CI lives at root) 2026-07-22 22:37:29 -07:00
Dockerfile frontend: rewrite the SSR content service in Go 2026-07-23 17:51:31 -07:00
format.py Subtree-merge thermograph-frontend (origin/main) into frontend/ 2026-07-22 22:01:11 -07:00
Makefile Subtree-sync frontend to split main 510d94d (dev->main promotion: reconciled hardening, two-tier suite); subtree workflow copy stays deleted (CI lives at root) 2026-07-22 22:37:29 -07:00
paths.py Subtree-merge thermograph-frontend (origin/main) into frontend/ 2026-07-22 22:01:11 -07:00
README.md Subtree-merge thermograph-frontend (origin/main) into frontend/ 2026-07-22 22:01:11 -07:00
requirements-dev.txt Subtree-sync frontend to split main 510d94d (dev->main promotion: reconciled hardening, two-tier suite); subtree workflow copy stays deleted (CI lives at root) 2026-07-22 22:37:29 -07:00
requirements.txt Subtree-merge thermograph-frontend (origin/main) into frontend/ 2026-07-22 22:01:11 -07:00

Thermograph frontend

The server-rendered content pages, the interactive-tool SPA shells, and every static asset for Thermograph — grades recent local weather against ~45 years of climate history. This repo holds no climate data and does no compute; it renders pages and serves the JS/CSS/image assets that call the backend's API from the browser. It was split out of the emi/thermograph monorepo (frontend_ssr/ + frontend/ there) so it can build its own image and deploy independently of the backend. See CLAUDE.md for the full agent-facing detail on the split topology and deploy contract, and thermograph-docs (sibling repo) for the architecture decision record behind the split.

Layout

api_client.py       HTTP client for the backend's content API (api/content_routes.py
                     on the backend). Fails loud at import if
                     THERMOGRAPH_API_BASE_INTERNAL is unset. TTL-cached (10 min
                     default) in front of every call.
app.py               FastAPI app: /healthz, the interactive tool's SPA-shell
                     routes (calendar/day/score/compare/legend/alerts), and the
                     StaticFiles mount for everything else. Calls
                     content.register(app) for the SSR routes below.
content.py           Server-rendered pages: climate hub, per-city, month,
                     records, glossary, about, privacy, robots.txt, sitemap.xml.
content_loader.py    Loads content/glossary.yaml + content/pages.yaml (SSR copy),
                     validated fail-loud at load time.
format.py            Unit-aware (°C/°F) formatting for the SSR templates,
                     ContextVar-scoped active unit.
paths.py             Canonical filesystem locations (STATIC_DIR, TEMPLATES_DIR,
                     CONTENT_DIR), resolved from the repo root.
templates/           Jinja templates content.py renders (base, home, city, month,
                     hub, records, glossary, glossary_term, about, privacy).
static/              Every static asset: the interactive tool's JS
                     (app.js, calendar.js, day.js, score.js, compare.js,
                     account.js, cache.js, shared.js, units.js, mappicker.js, …),
                     the SPA-shell HTML pages, style.css (all design tokens —
                     see DESIGN.md), icons/favicons, manifest.webmanifest, sw.js.
content/             Structured SSR copy: glossary.yaml, pages.yaml.
tests/               pytest suite — see the KNOWN GAP note below before relying on it.
tools/               shoot.py — the make-shots visual-verify screenshot tool
                     (see DESIGN.md).
Dockerfile           Builds this repo's own image (python:3.12-slim,
                     uvicorn app:app), independent of the backend's.

How it fits the split

Four sibling repos replace the old monorepo:

  • thermograph-backend — the FastAPI API, grading/scoring/grid compute, and DB. This repo's only runtime dependency.
  • thermograph-frontend (this repo).
  • thermograph-infra — Terraform, compose files, deploy/deploy.sh (the script both repos' deploy workflows SSH into), the secrets vault.
  • thermograph-docs — architecture/decision docs and runbooks.

Each app repo now builds and publishes its own image (git.thermograph.org/emi/thermograph-frontend/app, tagged by git SHA and semver) instead of the old shared emi/thermograph/app image, and deploys via its own .forgejo/workflows/deploy.yml (main → beta) and deploy-prod.yml (release → prod), each rolling only the frontend compose service on the target VPS.

Talking to the backend

All data comes from the backend's content API over HTTP (api_client.py) — this process holds nothing in-process. Two env vars control the topology:

  • THERMOGRAPH_API_BASE_INTERNAL (required) — where this process reaches the backend server-side, e.g. http://backend:8137 in compose, or http://127.0.0.1:8137 for a local backend checkout. api_client.py raises at import if this is unset — a missing backend URL should break boot, not silently 500 the first request.
  • THERMOGRAPH_API_BASE_PUBLIC (optional) — the backend's browser-facing origin, used only when frontend and backend are served cross-origin (e.g. a genuinely separate LAN-dev topology). Left empty (the default), asset/API references stay relative and same-origin, which is today's real deployed topology.

Same-origin cookie-auth caveat: the interactive tool's auth (static/account.js) is an HttpOnly session cookie. It uses credentials: "include" (not the fetch default) specifically so the cookie still rides along if this script is ever loaded cross-origin, but the backend must actually be configured to accept credentialed cross-origin requests (CORS + SameSite) for that case to work — in the default same-origin deployment this is a non-issue. Don't assume a fully decoupled, independently- hosted frontend "just works" for logged-in features without checking that cookie/CORS configuration first.

THERMOGRAPH_BASE (default /thermograph) must be set identically on both frontend and backend in every real deployment — it's how both processes agree on the shared URL prefix (or the deployed clean-root /, which the Dockerfile sets by default).

Build & run

python3 -m venv .venv && .venv/bin/pip install -r requirements.txt
THERMOGRAPH_API_BASE_INTERNAL=http://127.0.0.1:8137 \
THERMOGRAPH_BASE=/thermograph \
  .venv/bin/uvicorn app:app --host 0.0.0.0 --port 8080

Or build the image directly:

docker build -t thermograph-frontend .
docker run -p 8080:8080 -e THERMOGRAPH_API_BASE_INTERNAL=http://backend:8137 thermograph-frontend

The image healthchecks GET /healthz (liveness only — it does no I/O, so it stays cheap; it does not prove the backend is reachable).

More detail

  • CLAUDE.md — agent-facing instructions: deploy contract, API-version pinning, the cross-repo /cell/ETag and pctOrd() contracts, and the current known test-suite gap.
  • DESIGN.md — the visual system (tokens, components, breakpoints) and make shots visual verification.
  • thermograph-docs — the repo-topology decision record this split implements.