c04b2ecee2638000cf6e529ce16ca19d354d1896
3
Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
  Jean-Luc Makiolaandmakiolaj
|
e10fae0710 |
Release v2.20.0 (#286)
Release 2.20.0. **Added** - Long-press an event's title, location or description to copy it ([#195]). - Settings → About lists the open-source projects Calendula ships, with copyright holder, licence and a link to each project's source ([#281]). **Changed** - Week and day view seat every hour in its own rounded cell instead of drawing a separator line across the column; an on-the-hour event fills its cell rather than overhanging the seam ([#113]). - Event titles run to the edge of their chip in all three grid views — no ellipsis. RTL keeps it, where clipping would cut the start of the title ([#164]). - Month events show their start time where the chip is wide enough, and event times across month, week and day are set in a regular weight ([#219]). - An invitation you haven't answered is drawn as an outline; declined events keep their filled, struck-through chip and now sort last in the month view ([#230]). - The top-bar title falls back to a shorter form instead of being cut off, and the bar's trailing edge lines up with the grid under it ([#165]). - The jump-to-today button carries today's date and rolls over at midnight ([#220]). - The quick-switch pill hides when there is nothing to switch between, and views can now be turned off down to a single one ([#150]). **Fixed** - A long location wraps and its card grows, instead of scrolling sideways out of sight ([#273]). - The event form capitalizes sentences again; location stays uncapitalized, being as often a URL as an address ([#146]). - A dragged block's floating copy wraps its time and title the way the block does, instead of clipping to one line while held ([#267]). ### Release preparation `versionName`/`versionCode` → 2.20.0/22000, `CHANGELOG.md`'s Unreleased section moved under a dated 2.20.0 heading, and a hand-written 497-character `fastlane/metadata/android/en-US/changelogs/22000.txt`. Six of the merged PRs had no CHANGELOG entry (#146, #150, #195, #230, #267, #281); they are written up here. The existing #164 entry only described the month view and did not mention the RTL exception — both corrected. ### Also here `scripts/check_changelog_lengths.sh`, run in CI next to the reproducible-release invariant. Play rejects a "What's New" over 500 characters, which would fail the upload *after* the release had already shipped to F-Droid and Codeberg — 2.19.4 went out at 697. The script walks every locale, fails on the current `versionCode` and only warns on the already-published files, of which 13 are over the limit from when the pipeline auto-generated them from `CHANGELOG.md`. Deviation worth naming: this does not fix those 13. They are published and Play never re-reads them, so failing on them would paint every PR red for no gain; `--strict` covers them if they are ever backfilled. Follow-up filed as #285 — the store listing exists in only 4 of 12 locales and its graphics still don't qualify for Play. All twelve issues above are closed by their own PRs; this branch ships them. Co-authored-by: Jean-Luc Makiola <business@jeanlucmakiola.de> Reviewed-on: https://codeberg.org/jlmakiola/calendula/pulls/286 |
||
|
Jean-Luc Makiolaandmakiolaj
|
3fca28810b |
Don't run the Android build for forge-housekeeping changes (#98)
CI already skips the Android build for docs-only pull requests, and that part works — a README-only PR reports green in ~11s. The skip-list was just too narrow to catch a realistic docs PR. PR #96 is the example: a contributing guide, docs corrections and issue templates. It ran the full lint + test + assemble four times, over exactly two files Gradle never reads — `.forgejo/ISSUE_TEMPLATE/config.yml` and `.gitignore`. Added to the list: issue templates, `.planning/`, `licenses/`, `renovate.json5`, `.gitignore`, `.gitattributes`, `.editorconfig`. It stays a skip-list rather than a build-list, so an unfamiliar path still builds by default; the workflows themselves, `.gitmodules` and `scripts/` are deliberately not skippable. The step now also prints which files forced the build, so the next "why did my docs PR build?" is answerable from the log. One latent bug alongside it: `github.base_ref` is normalised against a `refs/heads/` prefix. Arriving in full-ref form would fail the merge-base lookup and quietly degrade the guard into "always build" — the failure mode this PR is fixing, but permanently and invisibly. Verified by replaying the new pattern over real history: PR #96 and #97 now skip; the forge migration (workflows), the search fix (app code) and a Weblate translation merge still build. No issue — reported directly. Co-authored-by: Jean-Luc Makiola <business@jeanlucmakiola.de> Reviewed-on: https://codeberg.org/jlmakiola/calendula/pulls/98 |
||
|
Jean-Luc Makiolaandmakiolaj
|
db7094c54e |
Move the canonical forge to Codeberg (#86)
Makes Codeberg canonical for git, issues, PRs, tags and releases. The self-hosted Gitea instance stays build infrastructure: signing key, F-Droid publishing, release pipeline. **This PR is its own test.** It is the first PR opened on Codeberg, so a green `CI` check proves the new runner works *and* that the submodule resolves from its new home. ### 1 · floret-kit moved Mirrored to `jlmakiola/floret-kit` (6 branches, 3 tags, every SHA verified identical) and `.gitmodules` repointed. This is what actually unblocks contributors — a clone previously could not resolve its submodule without reaching the personal Gitea instance. The Gitea copy is **kept**: every existing tag records the old submodule URL, so rebuilds of past releases (including F-Droid reproducible rebuilds) still resolve. ### 2 · Workflows split by directory Forgejo's lookup is first-match-wins across `.forgejo/` → `.gitea/` → `.github/`, and Gitea cannot see `.forgejo/` at all. So each forge sees exactly one set, with no duplicated files and no expression to keep in sync: | Directory | Runs on | Contains | Secrets | | --- | --- | --- | --- | | `.forgejo/workflows/` | Codeberg | `ci.yaml`, `translations.yaml` | **none** | | `.gitea/workflows/` | Gitea | `release.yaml`, `renovate.yml` | all of them | The line is drawn at **secrets, not CI-vs-release** — that is what makes fork PRs safe. Renovate deliberately does *not* move despite opening PRs here; it keeps running where its token already lives and merely talks to Codeberg's API. ### 3 · Two release-pipeline safety changes - `release.yaml`'s `detect` gets an explicit `repository_owner` guard. The directory split only holds while `.forgejo/` is non-empty; empty it and Codeberg would fall back to `.gitea/` and start running the release pipeline on the contributor-facing runner, without secrets. - `detect` now reads tags from **Codeberg**, not from the Gitea instance it runs on. Push mirroring is `git push --mirror`, so a tag minted on Gitea is deleted by the next sync until the Codeberg tag push propagates back — asking Gitea inside that window reports "no tag" for an already-shipped release and would cut it twice. It also now fails on any status other than 200/404 rather than reading a transient error as "no tag": a failed job is recoverable, a duplicate release is not. ### 4 · Links repointed In-app Source/License links, README badge, both F-Droid metadata files. **`Repo:` in `docs/fdroid-official/` deliberately stays on Gitea** — it keeps receiving `main` and every tag, so it remains a complete build source, and leaving it alone means no fdroiddata MR and no reproducible-build risk. ### Not in this PR Renovate + Weblate repointing, and the Codeberg → Gitea push mirror (browser-side). Supersedes Gitea PR #104. Co-authored-by: Jean-Luc Makiola <business@jeanlucmakiola.de> Reviewed-on: https://codeberg.org/jlmakiola/calendula/pulls/86 |