The check-changes job was only diffing HEAD~1..HEAD (the single most recent
commit). When CI-only commits landed after Android code changes, the deploy
was skipped every hour even though app code had changed since the last
successful Play Store publish.
Fix: fetch full history (fetch-depth: 0) and diff from LAST_DEPLOYED_SHA
when available, so all commits since the last deploy are considered. Also
simplify the Python workflow_id filter to match regardless of whether
Forgejo returns a bare filename or a full path. Fix duplicate entry in
check_coverage.dart exclusions and update the selection-mode golden that
was off by 4 pixels.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
dagger call --progress=plain -q writes its final error (e.g.
"invalid return status code") directly to the controlling terminal
rather than through stdout/stderr, so the DAGGER_OUT file that the
grep-based retry check reads ends up empty. Add RC=2 as an
additional fallback condition so the retry triggers even when the
output-capture path misses the error message.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Add "context deadline exceeded" to the retry_dagger network-error
pattern so transient Dagger engine connection timeouts (10-min
BuildKit context deadline) trigger a retry instead of immediately
failing the full check suite.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Remove the publish-website job from deploy.yml and add an hourly cron
schedule to website.yml so the website deploys independently on its own
cadence. Also fix pre-existing golden test and coverage exclusions for
files introduced in #315.
Reimplements the changes from PR #307 (branch issue-299-fix) on top of
current main, resolving all merge conflicts.
Features added:
- User preferences system (DB schema v34–v36): menuPosition,
mailViewButtonPosition, afterMailViewAction stored as a singleton row
- Preferences screen accessible from the account drawer
- Configurable menu bar position in mailbox view (bottom/top)
- Configurable back button position in single mail view (bottom/top)
- Configurable "after mail action": navigate to next message or return to
mailbox after delete/archive/spam/move/snooze
- Archive button in email detail screen; resolveMailboxByRole helper
prompts to choose or create a folder when none exists
- Improved Mark as spam: uses resolveMailboxByRole with dialog
- Show full discrepancy details in account list sync health row
- CSS fixes in SecureEmailWebView to prevent HTML email content overflow
- Preserve manually-set mailbox roles across IMAP syncs
- Tooltip on List-Unsubscribe chip showing the URL
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
## Summary
- Disables the `github-actions` Renovate manager in `renovate.json`
- Removes the previous `fileMatch` override that pointed Renovate at Forgejo workflow files
- Stops Renovate from scanning workflow YAML files for action version updates, eliminating GitHub API calls and the "GitHub token is required" warning
## Test plan
- [ ] Verify `renovate.json` is valid JSON (done locally with `python3 -m json.tool`)
- [ ] Confirm the next Renovate run no longer produces the GitHub token warning in its logs
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-authored-by: Thomas SharedInbox <sharedinbox@thomas-guettler.de>
Reviewed-on: https://codeberg.org/guettli/sharedinbox/pulls/306
The /usr/local/renovate/dist directory is owned by root.
Temporarily switch to root for the sed patch, then back to ubuntu.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Files are under dist/ not lib/, and we need to patch both
forgejo and gitea platform caches since platform=forgejo is set.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Codeberg's API times out (504) on GET /pulls?state=all&limit=100
but completes in ~9s at limit=10. Patch the compiled pr-cache.js
in the renovate:43 image before running to replace the hardcoded
20/100 page sizes with 10.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Codeberg's API times out (504) when fetching 100 closed PRs
(GET /pulls?state=all&limit=100), but succeeds with limit=20.
Renovate uses limit=100 on the first run and limit=20 on incremental
syncs. Pre-seeding the repository cache with one dummy entry tricks
Renovate into using the limit=20 incremental path from the start.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
renovate/renovate:39 did not support "forgejo" as a platform name;
v43 does. Upgrade the image and restore the correct platform name.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
renovate/renovate:39 does not recognise "forgejo" as a platform name;
the correct value is "gitea", which covers Forgejo/Gitea instances
including Codeberg.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Forgejo's API returns head_sha=null in workflow run objects; the correct
field is commit_sha. The skip-check always got None, so every hourly
schedule triggered a full redeploy of the same commit.
Forgejo's API returns head_sha=null in workflow run objects; the correct
field is commit_sha. The skip-check always got None, so every hourly
schedule triggered a full redeploy of the same commit.
Move Android Firebase instrumented tests out of deploy.yml into a new
firebase-tests.yml workflow that runs once per day (3 AM UTC) and only
when Firebase-relevant files changed in the last 24 hours. On failure,
the workflow automatically creates a Forgejo issue labelled "Ready" with
instructions to find the root cause and fix it.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Replace `task website-deploy` (which calls `hugo` directly and fails
because Hugo is not installed on the CI runner) with the Dagger-based
`task publish-website`, matching the pattern used by other jobs in
deploy.yml. Also adds Dagger remote engine setup, runner tool checks,
SSH_KNOWN_HOSTS secret, a timeout, and TLS credential cleanup.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Schema v33: add error_stack_trace and is_permanent columns to sync_logs
- SyncLogEntry gains stackTrace and isPermanent fields; SyncLogRepository.log()
gains matching optional parameters; IMAP and JMAP sync loops forward the
stack trace string and isPermanent flag when writing error entries
- New lib/ui/utils/about_markdown.dart utility shared by AboutScreen and the
sync log copy feature; builds the markdown table including device info
- AboutScreen uses the utility (refactored to remove duplicate _buildMarkdown)
- SyncLogScreen: subtitle shows "Error (permanent)" for permanent errors;
expanded view shows stack trace in red monospace; each tile has a Copy
button that copies a markdown summary of the entry plus the About section
- Migration test updated for v33; new repo test for stackTrace/isPermanent
- check_coverage.dart excludes lib/ui/utils/about_markdown.dart
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
## Summary
- The hourly `deploy.yml` schedule re-deployed the same commit repeatedly because it always diffed `HEAD~1..HEAD` — once a commit touching `lib/`/`pubspec.*` became HEAD, every hourly tick would detect "android changes" and deploy again.
- Fix: at the start of the `check-changes` job, query the Forgejo workflow runs API for the last successful `deploy.yml` run. If its `head_sha` matches current HEAD, output `android=false` / `linux=false` immediately, skipping all downstream jobs.
- `workflow_dispatch` bypasses this check (always deploys), matching the existing behaviour.
## Test plan
- [ ] Verify the `check-changes` job exits early on the next scheduled run after a successful deploy of the same commit
- [ ] Verify a new commit still triggers deployment normally
- [ ] Verify `workflow_dispatch` still deploys unconditionally
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-authored-by: Thomas SharedInbox <sharedinbox@thomas-guettler.de>
Reviewed-on: https://codeberg.org/guettli/sharedinbox/pulls/265
