zizmor flags the HealthChat release workflow now that the rebase PR shows
the whole file as new: five actions referenced by floating major tags, the
workflow_dispatch version input expanded straight into a shell script, and
a Gradle cache on a release build.
- Pin checkout, setup-java, setup-gradle, upload-artifact and
upload-google-play to commit SHAs with their version in a comment, the
way every upstream workflow does.
- Pass the version input through the environment instead of expanding it
into the run script.
- Disable the Gradle cache for the release build so a poisoned cache from
another workflow cannot reach the signed bundle.
The cached LFS checkout action checks out with persist-credentials: false
and then runs `git lfs pull` on a cache miss. Upstream is public, so that
pull works anonymously; on this private fork it fails with
`could not read Username for 'https://github.com'`, which is why the
Record-Screenshots label never produced a recording here (the only
successful runs were manual dispatches with a warm LFS cache).
Keep the job token configured for both LFS checkout steps. The push of
the recorded goldens already authenticates with the same token.
Follow-ups needed after rebasing the HealthChat series from v26.07.0 onto
v26.10.0 that span several of the replayed commits:
- Rename the five PreviewParameterProvider subclasses in features/healthchat
to the *PreviewParam suffix now enforced by Konsist.
- Rename MemberDirectoryEvents and HealthChatProfileSectionEvents to the
*Event suffix used by every upstream sealed interface.
- Import FakeStartDMAction from features/startchat/test, where it lives now.
- ListItem lost its headlineContent parameter in favour of content.
- HomeView gained onAddAccountClick and onSwitchToAccount, and onRoomClick
now carries an EventId; update HomeViewBackHandlingTest accordingly.
- Restore upstream's codecov-action pin (v7.1.1) that the CI squash carried
back to v7.0.0.
- Gate the new pr-checks.yml on the upstream repository: this fork does not
use the PR- label convention.
- Add the 26.10.0 section to HEALTHCHAT_CHANGES.md.
* feat(healthchat): add member directory tab to home navigation
Third bottom-nav tab backed by the HealthChat API: searchable member
cards with native start-chat, approval-pending banner, retry on
failure and navigation into the native profile screens.
* fix(healthchat): stop tab back handler swallowing back presses
While a screen is pushed on top of Home, the still-composed tab
BackHandler intercepted back presses during the transition and
invisibly switched to the Chats tab. Gate it on the node's target
visibility in both the parent and Home's own backstack.
* fix(healthchat): satisfy detekt, ktlint and konsist rules
Import alias order, composable parameter order in HomeView, and konsist presenter helper shape.
* ci: pass the real repository name to the screenshot recorder
recordScreenshots.yml set GITHUB_REPOSITORY from `secrets.GITHUB_REPOSITORY`.
No such secret exists on this fork, so the expression expands to an empty
string and overrides the built-in environment variable of the same name.
recordScreenshots.sh checks it up front and exits 1 with "No repo specified",
so the workflow could never record anything here.
Use the `github.repository` context instead, which is always populated.
Claude-Session: https://claude.ai/code/session_01MQbp1EPM4TSXq8P8Pyq44F
* Update screenshots
* Revert "Update screenshots"
This reverts commit b99e0d643f.
* Reapply "Update screenshots"
This reverts commit 85bceb6d67.
* test(screenshots): stop a LayoutLib teardown crash failing the shard tests
Paparazzi 2.0.0-alpha05 / LayoutLib 16.2.1 races on HandlerThread startup:
HandlerThread_Delegate.run calls Process.setThreadPriority, which throws
NoSuchMethodError for Thread.setPosixNicenessInternal and kills the
"android.bg" thread. If the test thread reaches getLooper() after that, it
gets null, BackgroundThread.ensureThreadReady() NPEs, and the static
initialiser of LatencyTracker fails. The JVM then marks that class erroneous
for the lifetime of the fork, so every later preview showing the soft
keyboard dies with NoClassDefFoundError.
LayoutLibErrorFilterStatement already swallowed the bare NoSuchMethodError,
but PaparazziLogger re-throws collected background errors from teardown()
wrapped in a MultipleFailuresException when there is more than one, and that
wrapper slipped through. A recent run reported 17 failures with zero image
comparisons failing and zero delta images written.
Unwrap the wrapper by reflection and swallow it only when every collected
cause matches the known-bug list; anything unrecognised, and anything that
cannot be decomposed, is rethrown.
Also close a hole this filter had from the start. Paparazzi runs the test as
`try { body } finally { teardown() }`, so a teardown error discards whatever
the body threw — including a genuine snapshot mismatch. renderFailureCapture
is registered inside the Paparazzi rule to record the body's exception so it
wins over any teardown error, and the filter refuses to swallow anything if
that rule is not registered. Two HorizontalFloatingToolbar previews have been
mismatching silently behind this, and may now surface as real failures.
Claude-Session: https://claude.ai/code/session_01MQbp1EPM4TSXq8P8Pyq44F
---------
Co-authored-by: ElementBot <android@element.io>
Squash of ten CI commits from 2026-08-04 (all part of HealthChat release 26.08.0):
- ci(release): move releases to the persistent self-hosted runners
- ci: grant contents:read so checkout works on this private fork
- ci: disable the Label Community PRs job on this fork
- ci: run the heavy Gradle jobs on the self-hosted runners
- ci: rely on the runner image for the Android SDK
- ci: size the Gradle heap for the self-hosted containers
- ci: report runner capacity on the self-hosted jobs
- ci: record why the Gradle heap is reduced
- ci: cap Gradle workers so test JVMs fit the container
- ci: skip the codecov upload when no token is configured
The danger.yml hunk from the contents:read commit is omitted: upstream
replaced Danger with a github-script workflow in v26.09.
Add --continue to the unit tests, screenshot tests, detekt and ktlint
Gradle invocations, so that a failure in one module does not prevent
the other modules from being checked, and the CI reports all the
failures at once.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
The enterprise repository is private, so only mention the number of new
commits in the generated PR description.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
When a PR is merged on develop with the enterprise submodule targeting a commit
which is not on the enterprise main branch, open a PR to update it to the latest
commit of main. The workflow also runs daily and manually to keep the submodule
up to date.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Each Maestro run logs in to @maestroelement:matrix.org, and the sessions
pile up on the account. Run tools/mas/delete_all_sessions.py at the end
of the Maestro job, even if the tests failed, to end them all.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Recording rewrote every screenshot unconditionally, so renderer noise far below
the verification tolerance produced large "Update screenshots" commits.
Use `app.cash.paparazzi.overwriteOnMaxPercentDifference` to keep screenshots that
are within tolerance. Existing screenshots are no longer deleted before recording;
obsolete ones are pruned afterwards, only when their preview produced no
screenshot at all.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* Remove Roborazzi, use Paparazzi for Compound
The new screenshots will be recorded with the rest of the Paparazzi ones. Note the names changed, since Paparazzi enforces a `$packagename.$testname.$name.$extension` format
* Set a max width value for color previews
* Merge the `verifyPaparazziDebug` CI steps
* Update screenshots
---------
Co-authored-by: ElementBot <android@element.io>
* Set `maxParallelForks = 1` for most tests, except for screenshot ones
It turns out the Robolectric setup penalty compounds the more parallel JVMs you spawn. Using 1 JVM vs 8 was up to 2.5x faster in local tests.
* Split the roborazzi tests tasks config and CI step
Secrets are never exposed to pull_request runs from forks, so checkEnv.js
throws on MAESTRO_PASSWORD after a full APK build and emulator boot, and the
red check is indistinguishable from a real test failure. The existing guard
checked github.event.pull_request.fork, a field that does not exist on the
pull_request payload (it lives at head.repo.fork), so it was always true.
Gate both jobs at job level on head.repo.full_name == github.repository
(workflow_dispatch still runs), and drop the dead step-level condition.
Fixes#7653
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
With the default (queue: single), a run that finishes Build APK while
another run is already pending in the maestro-test group cancels that
pending run, which then never gets a Maestro verdict for its commit.
queue: max keeps the one-at-a-time constraint (no cancel-in-progress) and
lets up to 100 runs wait, at no runner cost.
Fixes#7654
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
The PR metadata checks move to a new PR Checks workflow, and the lint,
ktlint and detekt reports are now surfaced as inline annotations from the
quality workflow. This removes the danger-js dependency, its ad-hoc npm
installs and the need for the ElementBot token, so the checks also run on
pull requests from forks.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
recordScreenshots.sh runs under `set -e`, and the `git commit -m "Update
screenshots"` at the end of it is unconditional. When a recording run
produces no diff — the common case once the snapshots are already up to
date — `git commit` exits 1 with "nothing to commit, working tree clean"
and takes the whole workflow down with it, reporting a red job for a
successful run.
Stage as before, then return early when the staged diff is empty. `set -e`
is left in place: it was added deliberately so that a failed Gradle run
cannot push a commit that deletes every snapshot. Using `git diff --cached
--quiet` as an `if` condition is `set -e`-safe, because its non-zero "there
are changes" status is consumed by the condition. The early return also
skips the push, which had nothing to send.
* Check if splitting the steps for testing and generating the reports improves performance and keeps the coverage stats
* Exclude Konsist tests and include `:app` ones
* Remove the extra dependency of `koverVerifyAll` on `:app:koverVerify{variant}`
New locales added via Localazy would change the locales line in
aaptDump.txt and break the manifest CI check if the reference file
is not kept in sync.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Sedat Qaja
joel.poyet
Benoit Marty
renovate[bot]
ElementBot
ganfra
Benoit Marty
Jorge Martin Espinosa
andreBurnt
hayaksi1