Two jobs meant checking out, setting up Node and running npm ci (and in the
frontend the build) twice, about two minutes each, because every job starts
in a fresh container. Now one job tests and, for pushes to master only,
deploys. The push-to-master condition is spelled out on every deploy step
instead of an env flag compared to 'true', which would skip the deploy
silently and stay green if the boolean were rendered differently. A failed
test still stops the deploy (an if: without a status function implies
success()).
Docs: the one-job layout, and that Jenkins deployed as root - files it left
in an app folder must be chowned to the subscription user before the first
Gitea deploy (hit on the API).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Tests every push and pull request (tsc, unit tests); pushes to master build,
prune dev dependencies and rsync dist/, src/, public/, node_modules/ and
package.json over SSH as the Plesk subscription user, then restart Passenger
via tmp/restart.txt. Only the app's own folders are mirrored with --delete;
.env, logs/ and tmp/ in the app root are never touched. Optional smoke test
via DEPLOY_SMOKE_URL. Same pattern as the calendar frontend; setup and what
the deploy may touch are in docs/deployment.md.
Also removes leftovers: bcrypt (unused since better-auth, and the only
production dependency with a native build) and the SonarQube setup
(sonar-project.properties, vitest-sonar-reporter).
Checked locally: actionlint; in a clean node:26 container npm ci, tsc,
233 unit tests, build and prune; the pruned bundle started against a
throwaway MariaDB (public feed, iCal, /docs, /admin/me); the upload step
against a throwaway SSH server with .env, logs/ and tmp/ in place.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>