Skip to content

Add CI - #1

Merged
nicodes merged 1 commit into
mainfrom
chore/add-ci
Aug 2, 2026
Merged

Add CI#1
nicodes merged 1 commit into
mainfrom
chore/add-ci

Conversation

@nicodes

@nicodes nicodes commented Aug 2, 2026

Copy link
Copy Markdown
Member

This repository had no .github at all, while its three sibling -web repos — gdam-web, gdlint-web, termcade-web — all had CI. Nothing has ever checked it.

  • .github/actions/build — install from the committed lockfile, build
  • .github/actions/test — the page was actually written
  • .github/workflows/ci.yml — one job, pull_request + push: [main] (no CD here, so nothing else covers a merge), SHA-pinned, timeout-minutes: 15

pnpm, not bun

The other four -web repos in this org are bun with the version pinned in .mise.toml. This one has a pnpm-lock.yaml at lockfileVersion 9.0 and no .mise.toml.

I install what the committed lockfile describes rather than converting it. Adding a check and migrating the package manager in one PR would make any build failure ambiguous — and the migration deserves its own decision. Flagging it as the one repo out of step with its family.

No no-JavaScript assertion

Unlike the nicodes static sites, this one ships Qwik — a build emits 13 .js files on purpose, so that check would fail on its first run. The test action says so, to stop it being added later for consistency.

Verified locally

pnpm install --frozen-lockfile and pnpm run build both succeed; dist/index.html written, 33 files out.

Phase 3 of the standardization plan (F10).

This repository had no .github at all, while its three sibling -web
repos -- gdam-web, gdlint-web, termcade-web -- all had one. Nothing has
ever checked it.

Build installs from the committed lockfile and builds; Test asserts the
page was actually written, since a build that emits nothing still exits
0.

pnpm, not bun. The other four -web repos in this organisation are bun
with the version pinned in .mise.toml; this one has a pnpm-lock.yaml at
lockfileVersion 9.0 and no .mise.toml. Installing what the committed
lockfile describes is the right thing for a change that is only meant to
add a check -- converting the lockfile and adopting mise is a separate
change with its own risk, and doing both at once would make a build
failure ambiguous.

No no-JavaScript assertion, unlike the static sites in the nicodes
organisation. This one ships Qwik: a build emits thirteen .js files on
purpose, so that check would fail on its first run.

Verified locally: install --frozen-lockfile and build both succeed,
dist/index.html written, 33 files out.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
@vercel

vercel Bot commented Aug 2, 2026

Copy link
Copy Markdown

The latest updates on your projects. Learn more about Vercel for GitHub.

Project Deployment Actions Updated (UTC)
fieldsofrevik-web Ready Ready Preview Aug 2, 2026 12:59am

@nicodes
nicodes merged commit 48e09cb into main Aug 2, 2026
3 checks passed
@nicodes
nicodes deleted the chore/add-ci branch August 2, 2026 01:00
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant