Files
sanderling/internal/driver/ioscompanion/driver.go
T
pj 9fb121e9d0 correctness fixes from the first real folio dispatch, and spec authoring skills (#82)
* docs: add the apache 2.0 license text

package.json has declared Apache-2.0 since the first release and .goreleaser.yaml
globs LICENSE* into the archives, so that glob has been matching nothing. npm
only picks up a license from the package directory, hence the copy under
pkg/spec.

* fix(sidecar): close the soft keyboard after typing on android

* test(sidecar): pin the guarded ime dismissal

* fix(sidecar): treat a failed ime probe as no keyboard open

* ci(folio): let the ios leg clear state for itself

* ci(folio): drop the stale frontboard note from the ios job

* docs(ci): record what the ios calibration assumes and where it was measured

* fix(ios): replace the session when a launch blows its bound

a launch the simulator refuses is never reported: xctest records it as a test
failure the runner cannot see, then holds the session's main thread for about
four minutes on a diagnostic chain. so the only signal is the expired bound,
and every later call queues behind the same wedge. restart the session once and
launch again, bounded so the launch path stays inside testrun's backstop.

* test(ios): cover the session replacement a wedged launch needs

* fix(ios): share one deadline across the restart and the second launch

the recovery a blown bound triggers now costs at most launchRecoveryTimeout
whatever it spends it on, so the launch path tops out at 150s and testrun's
three minute backstop stays a backstop.

* test(ios): the restart a blown launch triggers has to be bounded

* docs(ci): the ios leg does not convict on the runner, and a seed cannot fix it

seed 28 reproduced its walk on macos-15 and reached the bug at the step it
convicts at locally. it still could not be judged: the run did not return
home between step 19 and step 136, so the counting invariant saw a rise of
15 against a window of 37 submits.

* docs(sidecar): record why the stale ime flag stays out of reach

* test(folio): add commonTest source sets to core and shared

* fix(folio): reject amounts parseCents cannot represent

* fix(folio): cap a transaction at one million dollars

* test(spec): give the fake dom a real tree and a walking querySelectorAll

* style(sidecar): make ktlint clean, formatting only

ktlint -F over every kotlin file except DriverBackend.kt, then hand
fixes where the reflow read worse and for the long lines ktlint cannot
break. No behaviour changes.

DriverBackend.kt is left untouched to avoid a conflict with concurrent
work; its three over-long lines still fail fmt-kotlin.

* fix(spec): deepQueryAll returns matches in document order

* ci(folio): say what the android gate found, not why

The gate proves only that AddTransactionScreen is absent from the trace.
Claiming the run never got past login was an inference it cannot make: the
run that produced it had logged in and was stuck on the new account screen.
Name the routes the trace does record instead.

* test(runner): a relaunch must not convict the submit counting property

* test(browser): compare ax.find across both hosts on one page

* test(spec): name the shadow match in the grammar both hosts parse

* ci: move every action off the node20 runtime

checkout v4->v7, setup-go v5->v7, setup-node v4->v7, setup-java v4->v5,
upload-artifact v4->v7, cache v4->v6, upload-pages-artifact v3->v5,
deploy-pages v4->v5, setup-chrome v1->v2, setup-android v3->v4,
goreleaser-action v6->v7. setup-bun and android-emulator-runner are
already node24; buf-setup-action stays on its deliberate SHA pin.

setup-chrome v2 resolves stable from Chrome for Testing rather than the
official installer, so the ci.yml comment about the action's default no
longer held.

* feat(verifier): report a relaunch on state.lastAction

* test(verifier): pin the relaunch field on both hosts

* fix(runner): keep the action the app was relaunched after

* test: pin that a nested undefined does not survive the wire

* fix(folio): uninstall before installing in just ios

folio's signed-in session lives in the data container, which an install
over the top keeps, so a local run started right after just ios opened on
the previous run's Home screen and diverged at step 1. On CI's fresh
simulator the uninstall is a no-op, so the ios leg is unchanged.

* style(sidecar): bring the last three lines under the line limit

* fix(ios): the runner must not answer ok for a launch that failed

XCTest records a refused launch as a test issue that never throws, so the
companion returned ok for an app that never started. Check the state the
app actually reached and report the refusal instead.

* test(ios): a refusal the runner names costs no session restart

The session restart is for a launch that never answers. A launch that
reports the app's state has already said what a fresh session would.

* fix(folio): attribute a created account by its whole key, not a suffix

createdAccountHasNonZeroBalance matched the created card with endsWith, so
an older account whose name ends with the typed one ("Emergency Fund" for a
typed "Fund") was judged instead whenever the new card was clipped out of
the reading. Build both keys the card can carry, the plain name and web's
initials + name, and compare them whole.

* fix(hierarchy): object selectors resolve by the same rule as string ones

* test(verifier): both ax.find selector forms resolve the same element

* test(browser): the cross-host fixture uses the object selector form

* fix(replay-ui): wrap the tab strip so its last tabs stay clickable

* test(replay-ui): drive the fuzzer onto a violating step with a panel

* feat(folio): judge a submit against the account's own balance

The counting invariant can only close its window on Home, and the iOS run
in #78 went 117 steps between two Home readings: 37 submits against a rise
of 15 transactions is no evidence about the double tap sitting inside it.
The ledger and the add-transaction screen both show the account's own
balance, and an accepted submit pops back to the ledger, so a window
bounded by those readings holds one action.

The bound is an upper one: a balance that has not moved is a commit still
in flight, a rejected submit or a tap that never landed, and none of those
is a violation. Moving by more than the one submit in the window typed is.

* test(runner): an overlay dismissal must not convict the counting property

* docs(runner): point the guard comment at the renamed test

* docs(replay-ui): name the viewport the tab overflow was measured at

* feat(folio): close the submit window on the account's own screens

submitCommitsOneTransactionPerAction now states its rule over two windows:
the Home counts it already compared, and the account balance the ledger and
the add-transaction screen redraw on nearly every frame of the transaction
flow. Same rule, and the second window is usually one action wide.

* fix(sidecar): reach the adb server the environment names

buildDadb hardcoded localhost:5037, so a serial-addressed device always
resolved through this machine's adb server and ADB_SERVER_SOCKET was
ignored. Read the endpoint the way the adb CLI does instead.

Fixes #79

* test(sidecar): pin the adb server endpoint parsing

* fix(sidecar): close a keyboard standing in the snapshot

A tap on a text field raises the keyboard and nothing closed it, so the
tree the picker chooses from was missing every app node underneath it,
the submit control included. Close it before the read rather than after
the tap: the picker only ever sees snapshots, and the keyboard is still
on its way up when the tap returns.

Fixes #78

* test(sidecar): pin the tree-guarded keyboard dismissal

* chore(make): a target that runs folio's unit tests

* chore(folio): a just recipe for the unit tests

* ci: run folio's unit tests on every pr

* ci: switch to jdk 21 only for the folio step

* docs(sidecar): put the measured read cost in the dismissal bound

* fix(folio): decline the two demanding properties across a relaunch

The runner now keeps lastAction and marks it relaunched: true where it used
to report nothing at all, so the two properties that demand an effect judge
a step whose process may have died before the write landed.
submitChangesBalanceByTypedAmount and createdAccountHasNonZeroBalance both
decline there. The counting bound does not: a relaunch cannot manufacture a
transaction, and the submit is counted, so declining would throw away the
detection the runner fix restored.

* docs(folio): say why the merged card key cannot be made injective

Folio rejects a duplicate account name, so the twin the drop rule guards
against is two names the web key cannot tell apart, not two accounts
sharing a name. State what closing the rest would cost and what the tree
would have to carry to close it properly.

* test(folio): pin that two accounts can render the same card text

The proof behind the comment: "Travel1" holding 25 transactions and
"Travel12" holding 5 merge to the same string, so no identity key read off
a web card can tell them apart.

* fix(sidecar): bound the diagnostic adb reads

adbOutput and readLogcat read to EOF and then waited with no timeout, so
a wedged adb held the step for as long as it liked; one stall over a
remote adb server measured ~100s. The bound has to sit on the read, not
on waitFor: a wedged adb never reaches EOF, so a bounded waitFor after
the read is a line that never runs.

* test(sidecar): pin the bound on a wedged adb read

* fix(sidecar): an unreadable animation count is not idle

Defaulting the count to zero made a dumpsys that said nothing mean
nothing is animating, so a degraded link broke out of the settle early
and handed the runner a frame caught mid-animation. Unknown now waits,
inside the deadline waitForIdle already holds.

* test(sidecar): unknown animation state must not read as idle

* fix(folio): stop spending the submit budget on taps the app refused

The window is an upper bound on the transactions an interval could hold, and
a bound inflated by taps that commit nothing is a bound the app can never
exceed: #78 read a rise of 15 transactions against 37 submits. TxnSubmit is
clickable(enabled = amount.isNotBlank()) and parseCents refuses anything its
regex misses, so a tap whose landing frame shows a refused amount cannot have
committed. Over four recorded android runs that is 19, 11, 25 and 25 of 35,
26, 42 and 42 submit taps.

A relaunch is excepted: a fresh process draws an empty field whatever was
submitted.

* feat(folio): read the amount field into every submit window

Each of the three windows asks whether the tap could have committed, off the
field as the landing frame shows it.

* fix(sidecar): a foreground read that fails degrades the typing guard

An unreadable dumpsys passed a null owner to typeChunks, which switches
the mid-type focus guard off outright and lets the rest of the string
spray into whatever holds the foreground. Fall back to the launched
bundle instead: the guard stays armed, typing still happens, and the
degradation is said out loud rather than assumed away.

* test(sidecar): pin the degraded typing guard both ways

* test(runner): answer Snapshot and Hierarchy off one tree in the fakes

* feat(runner): skip a step whose tree changed between two reads

* test(runner): cover the reread's cost to the existing snapshot rules

* fix(ios): clear app state before the automation session attaches

New performs the clear-state reset, so the uninstall and reinstall no
longer land underneath a live XCTest session that is already bound to
the app. Launch refuses a clear-state request the driver was not built
for rather than reinstalling under its own session.

* fix(ios): the device path clears before its runner session too

* fix(testrun): thread clear-data into the ios drivers

* test(runner): a skipped step must not swallow the action before it

* fix(runner): hold the action back on a step nothing verified

* refactor(runner): drop the empty branch from the hold path

* docs(runner): describe both modes of the composing test driver

* fix(sidecar): erase a field by selecting it, not one delete per character

maestro's eraseText sends one delete per character through its
instrumentation, measured 29.6 ms/char on the API 34 emulator. The
4096-character string the corpus types cost ~121s to clear, a fifth of a
20 minute run spent on one step, and it recurred every time that field
was typed into again.

Select the content and delete the selection instead: two key events at
any length, measured 0.15s to 1.16s for 4096 characters across API 34,
35 and 36. The result is read back off the tree, and a field that is not
empty, or that the tree cannot report on, is finished off per character
in batches rather than assumed clear.

Fixes #80

* test(sidecar): pin the constant-cost erase and its residue check

* fix(sidecar): find the erased field by class, past the keyboard's own focus

The check that decides whether the select-all worked looked for an
"editable" attribute maestro's tree does not carry, so it answered
"cannot tell" every time and every erase paid the per-character
fallback. Worse, an open keyboard puts a second focused node in the
tree, one of the IME's own keys, carrying no text: taking the first
focused node would read a field still holding 4096 characters as empty,
which is the one answer that stops the erase early.

Match the text field by class instead. Measured against the real
backend, 4096 characters now clear in 385ms on API 34, 409ms on API 35
and 870ms on API 36, verified empty, where the fallback took ~4s.

* test(sidecar): use the tree the device really returns

* docs(ci): the android step number describes a local emulator, not ci

the leg disables animations and the number was measured with them on. the
first real dispatch carries 4 transitional steps over 200, so the cross-fade
wait does still fire in ci, just far less often.

* fix(android): say what the sdk lookup checked, not just to set ANDROID_HOME

* fix(doctor): resolve adb and emulator the way a run does

* docs(cli): the android doctor checks are not path-only

* fix(testrun): preflight resolves adb through the sdk, not just PATH

* docs(skills): add a spec review skill and the skills index

* fix(testrun): report a sidecar that dies at startup as the exit it was

* test(testrun): cover the sidecar shutdown path after an early exit

* docs(skills): add a property patterns catalogue skill

* fix(folio): bound the total-balance move instead of demanding it exactly

The write finishes before AddTransactionViewModel navigates, but nothing
establishes that Home's total has re-rendered before the frame is read, and
an equality convicts a healthy app for a total one frame behind. A delta of
zero is exactly the shape nine of the eleven measured android false
convictions had. 2x still exceeds x, so all four recorded convictions
survive, checked against the traces.

The trade is real: a balance that moves by LESS than the amount typed is no
longer judged anywhere in this spec.

* docs(folio): say what property 2 demands now that it is a bound

* docs(skills): add a spec authoring skill

covers hooks, extractors, selectors, properties, actions and the order to write them in, with a complete sample spec that typechecks against the real export surface.

* docs(skills): ground the property patterns catalogue in the merged specs

* docs(skills): name the selector keys that still substring match

* docs(skills): add a setup skill for adopting sanderling

* docs(skills): add a run triage skill

* docs(skills): point the setup skill at its siblings

* docs(manual): correct the flags the cli reference gets wrong

--launcher-activity does not exist in cmd/sanderling/main.go. --device,
--android-app-path and --arm do and were undocumented. runs.md still listed
--max-steps and --exit-on-violation as unshipped, and described --clear-data
as opt-in when the default is already true, contradicting itself ten lines on.

* test(folio): pin that a commit stays in the window until Home reads it

The interaction that keeps a stale Home card list from ever banking counts
the budget has already forgotten: a submit lands on the ledger, so the
reading that resets the window is a whole action later and the submit is
still in it. Characterization, not a regression: no code changed and it
cannot go red first.

* docs(folio): record why a banked card reading can be trusted as current

The freshness rule rests on the app popping one entry back to the ledger,
not on anything the frame carries, so the assumption and the measurements
behind it belong next to it.

* refactor(folio): name the balance property for the bound it asserts

it stopped being an equality and became |delta| <= typed, so the old name
demanded more than the property does. renamed with the ci gate's
GATED_PROPERTIES in the same commit so the gate never sees a name it does
not know.

* fix(android): a refused uninstall must not pass for clear-state

adb uninstall answers Failure [DELETE_FAILED_INTERNAL_ERROR] both when the package was never installed and when it refuses to remove one, so the failure text cannot say which happened and the old code installed over the top either way, keeping the data clear-state was asked to drop. Ask pm path instead, and fall back to pm clear when the app is still there.

* fix(ios): a failed simctl uninstall must fail the reinstall

simctl install over an installed app carries its data container across, so discarding the uninstall error reported a clear-state that never happened. Uninstalling an app that is not installed exits 0 on a booted simulator, so every failure here is a real one.

* fix(ios): a failed devicectl uninstall must fail the reinstall

same hole as the simulator path: devicectl install over an app keeps its data, and the discarded uninstall error hid it. Uninstalling a bundle id that is not installed exits 0 with 'App uninstalled.' on a paired iPhone, so a failure here is always real.

* docs(android): say why the uninstall text cannot be read

* test(android): name the uninstall failure for what it says, not why

* docs(ci): the ios leg convicts on the runner now, and why it did not before

* docs(ci): the cross-fade wait does not fire on ci, say so

* fix(runner): a bounded hold puts the swallow back one step later

the hold carries one action; letting the runner act again while the verifier
is still skipped overwrites it, so the carried action reaches no spec. hold
for as long as the verifier is skipped, and settle on a held step so the
reread pair is not tighter than the window the detector was measured over.

* test(runner): pin what the two reads are compared on

structuralShape excluding text and bounds is the decision separating this
feature from a run that verifies nothing, and only prose held it. adding
either field back now turns a case red.

* fix(ci): close shell injection into the npm publish job

A refname is attacker-controlled and git permits backtick, $, (, ; and |
in it. Three sites substituted it into a run: block, and NODE_AUTH_TOKEN
sat at job level, so a pushed tag ran arbitrary commands with the publish
credential in reach.

The tag now goes through env:, is validated against an anchored version
pattern before anything consumes it, and reaches the other jobs as a job
output. The token is scoped to the publish step. release-npm declares
contents: read instead of inheriting the repo default.

* fix(ios): recognise every shape a blown launch bound arrives in

The runner transport reports a blown budget two ways, its own comment says
so: the context's error once cancellation has landed, and the connection's
i/o timeout when the deadline armed from that context fires first. The
legacy transport reports it as a gRPC status. errors.Is against
context.DeadlineExceeded only matches the first, so the session restart
never fired for the other two and a wedged session stayed wedged.

* test(ios): drive the launch recovery with what the transports return

The wedged-session fake answered with ctx.Err() raw, which is the one
shape the guard already matched. The recovery now runs against the error
each transport really produces for the same expiry, taken from a runner
and a legacy companion that never answer.

* test(runner): pin both guard writes to what the spec reads

deleting lastAction.Relaunched or lastAction.Applied left the whole suite
green, so the only producer of the two fields every spec-side guard reads
had nothing holding it. both now assert the value out of the trace.

* fix(testrun): a run that judged nothing is not a green run

every step skipped means no property ever evaluated, so no violations is the
absence of a verdict rather than a clean one. the hold makes that reachable
now, so the run says it instead of exiting 0.

* fix(ios): stop the app before clearing its state

Launch terminated and then cleared; the clear moved to construction and
left nothing stopping the app first. The container wipe deletes files a
live app still holds open, and the CI ios leg passes no app path so the
wipe is the path it takes. simctl stops it, since the clear now runs
before any automation session exists. On a device the uninstall that is
its only clear takes the running app with it.

* test(ios): pin the stop that has to precede a clear

The ordering probe now records the stop, and a scripted xcrun holds what
reaches the tool: terminate before get_app_container, with the previous
run's files gone after. A simctl terminate that finds nothing to stop
still leaves the clear a success.

* docs(spec): an unbounded eventually is violated at run end

* docs(skills): an unreached eventually convicts at run end

* docs(skills): noUncaughtExceptions only fires on web

* fix(ios): a device clear-state that cannot happen must fail

--clear-data on a physical device with no --ios-app-path warned and then
ran anyway, so the run started on the previous run's data while the flag
said it started clean. There is no data-container wipe on a device, so
there is nothing to fall back to.

* test(ios): a device clear-state without an app path ends the run

* docs(skills): the stock properties each cover one platform

* docs(ci): the balance property demands a bound, not an equality

* fix(ios): the clear-state guard checks the bundle that was cleared

A bool only said that something was cleared, so Launch(ctx, otherBundle,
clearState=true) passed the guard and reported a reset that had reached a
different app. Record what was cleared and compare against the bundle
being launched.

* test(ios): a clear-state launch for an uncleared bundle is refused

* docs(manual): the flagship property is a bound, and say what that costs

* fix(ios): one address picker for every bring-up

bringUpRunner reads the picker from a field, and NewDevice only ever set
the device one, so a device driver that reached bringUpRunner would call
nil. The two fields held the same function; keeping one leaves no path
that can be wired without it.

* test(ios): a device driver can bring a runner up

* docs(skills): both shipped balance forms are bounds now

* docs(skills): name the balance predicate that still exists

* docs(skills): quote the doctor the binary actually prints

* docs(skills): screen= is the chrome driver's url, web only

* docs(skills): substring selector matching is native only

* docs(skills): web selectors are exact, native ones are substrings

* fix(ci): a run that wrote no trace is not evidence about folio

run_dir is empty when the run produced no output directory, and the
fallback made trace ./trace.jsonl. A stray trace in the working directory
was then read as this run's, so a run that wrote nothing reported 'found
the submit bug' and exited 0, defeating the missing-trace check below it.

* fix(ci): fail folio when a gated property is not in the spec

Nothing tied GATED_PROPERTIES to the spec it gates. Renaming a property
left the classifier matching nothing: ios and web blamed the spec for
finding a different bug, and android silently reclassified a real
conviction as 'judging health only' and stayed green.

replay-ui-summary.sh already makes this check for its own list. The spec
path becomes SPEC-overridable the same way, so the check is testable.

* test(ci): cover the folio classifier's verdicts

21 cases through a stubbed sanderling: every exit path, the drift check,
a missing trace, a zero-byte trace, an empty glob and a truncated line.
Asserts the flags that reached the binary, not just the exit code.

Invoked as bash -eo pipefail -c, which is what a run: block does. Running
folio-run.sh itself under -e would kill it at the first non-zero
sanderling test, which is the exit code it exists to read.

* docs(skills): defaultActions bundles five of the eight generators

* test(ios): name the picker test for what it covers

* docs(skills): three of the replay-ui properties are cross-panel

* docs(manual): state.exceptions is web only and reportError does not exist

* fix(folio): the bound carries no unconfirmed-submit guard

deleting confirmedApplied here broke 0 of 355 tests: under a bound a submit
that may not have landed moves the balance by 0, which the bound already
permits, so the guard could only ever drop the double commit it exists to
catch. the relaunch guard stays for a reason the bound does not cover, and
both tests now assert a verdict that changes when their guard does.

* docs(ci): three of the replay-ui properties are cross-panel

* docs(manual): the starter property only fires on web

* test(folio): judge the conjunct on the landings a real run produces

three of the 18 frames the recorded ios run drove it down, each with the
second commit the bound is there to catch. neutering the comparison reddens
it: a second commit on 357900 went unjudged.

* test(folio): the walk drives the composition the spec runs

countSubmitsInWindow never saw an amountText here, so every walk test counted
submits the app must have refused. with the field passed, a refused submit no
longer buys a later double tap an alibi: without it the window reads 3, not 1.

* docs(folio): say which double submit the conjunct can see, and which it cannot

the home landing is the counting invariant's, three of three in the recorded
ios run; this one gets the interleaving whose second pop is cancelled. it is
still the only judge on the 18 ledger landings that run produced.

* docs(folio): the narrow window is not where the detection comes from

the double taps land on home, so the counting form convicts them; what turned
0 convictions into 4 on the recorded ios run is submitCouldCommit, which drops
the windows at those three steps from 5/4/7 to 2/1/2.

* docs(skills): folio drives three platforms from one spec

* fix(folio): an amount over the app's cap spends no window budget

the corpus reaches TxnSubmit with 999999999999999999999, AMOUNT_REGEX takes it
and AddTransactionViewModel refuses it against MAX_TRANSACTION_AMOUNT_CENTS, so
counting it was budget a double submit could hide behind.

* docs(folio): say which form judged one step, not which node was read once

* docs: a bound still needs the relaunch guard, and eventually does convict

* fix(sidecar): the hierarchy rpc serves the tree the snapshot reads

the runner compares the two per step, but snapshot settles and closes a
keyboard while hierarchy was a bare contentDescriptor. measured on emulator
-5556 (api 34) with an ime open: 489 nodes against the snapshot's 134. both
now come off snapshotTree under the same lock; the reread still costs ~75ms
when no keyboard is up.

* docs(runner): say what makes the two reads comparable

the reread's comment claimed the round trip was the only interval between
them; what it left out is that the two rpcs have to read the same way, which
the repo's own android backend did not do.

* refactor(runner): name the settle predicate for what it means

* ci: add a headless-chrome composite action

The setup-chrome / apparmor sysctl / launch-check trio is copied across
three jobs. The old comment described setup-chrome v1 semantics: under v2
stable is the default and the alternative is Chrome for Testing latest,
not a dev Chromium, so it is restated for what the pin actually does.

* ci(examples): add the folio setup actions

folio-app holds the per-platform toolchain and app build, so a caller
guards one step instead of eight. folio-simulator boots the simulator,
installs folio and leaves the app stopped.

* ci(examples): add the replay-ui fixture action

Records a trace and serves it with sanderling replay. The step page URL
is a composite output rather than GITHUB_ENV, so it is scoped to the one
step that drives it.

* ci(examples): one dispatch workflow for every example

folio.yml and replay-ui.yml ran the same operation: build sanderling for
a platform, bring a target up, run a spec against it, classify the trace,
upload the run. They are now one matrix over four examples, each naming
its own runner.

The job is named for what it fuzzes. 'dogfood' named why we run it, not
what runs, the same error as a diagnostic that reports a motivation
instead of an observation.

The matrix is computed by a plan job because jobs.<id>.if cannot read the
matrix context, so a static matrix has no way to leave a leg out. Seeds,
budgets, timeouts, runners and artifact names are unchanged.

* ci: reuse the headless-chrome action in the browser job

Same three steps the examples workflow needs, and the comment explaining
the AppArmor sysctl now lives in one place.

* ci: move the folio jdk step to setup-java v5

The only setup-java left on v4; every other one moved.

* ci: pin third-party actions to commit shas

buf-setup-action was already pinned with a comment saying why; the other
five rode mutable major tags, so a tag move is an unreviewed change to
what runs. Each major currently resolves to the release named in the
comment, so this freezes today's behaviour rather than changing it.

actions/* stay on major tags: they are first-party to the runner.

* ci(replay-ui): name the run directory for what it fuzzes

runs/dogfood and the '### replay-ui dogfood' heading carried the same
naming error as the job name: dogfooding is why the run exists, not what
it fuzzes.

* docs(driver): state the log level scale on LogEntry

Claude-Session: https://claude.ai/code/session_01ShuAy8q8ZfPi8KHxwc8JpQ

* docs(sidecar): name the device the node counts came off

* fix(chrome): keep a log entry the level scale cannot rank

Claude-Session: https://claude.ai/code/session_01ShuAy8q8ZfPi8KHxwc8JpQ

* fix(chrome): record console levels on the logcat scale

Claude-Session: https://claude.ai/code/session_01ShuAy8q8ZfPi8KHxwc8JpQ

* docs(ci): say why upload-pages-artifact needs no include-hidden-files

v3 to v5 crossed v4's change to exclude dot-files. build/site has none,
so nothing was dropped, and the underscore directory is not hidden.

* test(browser): drive a console error through to the spec

Claude-Session: https://claude.ai/code/session_01ShuAy8q8ZfPi8KHxwc8JpQ

* docs(ioscompanion): name the vacuity behind the empty log slice

Claude-Session: https://claude.ai/code/session_01ShuAy8q8ZfPi8KHxwc8JpQ

* ci: run the folio classifier's test in make test-ci-scripts

* fix(chrome): keep the message of an object console argument

Claude-Session: https://claude.ai/code/session_01ShuAy8q8ZfPi8KHxwc8JpQ

* test(browser): cover console.error with an error object

Claude-Session: https://claude.ai/code/session_01ShuAy8q8ZfPi8KHxwc8JpQ

* feat(spec): let the runner install state.logs in the page

Claude-Session: https://claude.ai/code/session_01ShuAy8q8ZfPi8KHxwc8JpQ

* feat(verifier): encode state.logs for the web host

Claude-Session: https://claude.ai/code/session_01ShuAy8q8ZfPi8KHxwc8JpQ

* feat(chrome): install the step's logs in the page

Claude-Session: https://claude.ai/code/session_01ShuAy8q8ZfPi8KHxwc8JpQ

* test(ci): pin three real folio traces from run 31902501859

ios convicted on submitCommitsOneTransactionPerAction, web on both gated
properties, android ran its full 200 steps healthy. Every step is kept;
of each step only step, violations, witnesses and residuals survive.

hierarchy is replaced by the quoted "...Screen" resource ids it held, in
order. It cannot just be dropped: it is 95% of the bytes and also the
only place the android route gate's grep can match, so dropping it flips
that leg from healthy to 'never reached'. 8.4MB to 59KB.

* test(ci): drive the classifier over the real traces

Four cases on real data: each leg's real verdict, plus the android trace
cut before it reached the transaction screen, which is what proves the
route gate reads a real hierarchy dump.

Also corrects the hand-written fixtures. They set is_error to false on a
plain violation; internal/trace/writer.go tags that field omitempty, so a
real trace omits it entirely. Harmless to the classifier, but a fixture
that does not look like reality is the thing that hides drift.

* test(runner): teach the web fakes to take the step's logs

Claude-Session: https://claude.ai/code/session_01ShuAy8q8ZfPi8KHxwc8JpQ

* fix(runner): install the step's logs before the page extracts

On web every extractor reading is replaced by the one the page computed,
and the page answered logs: [], so noLogcatErrors counted an empty array
however full of errors the console was.

Claude-Session: https://claude.ai/code/session_01ShuAy8q8ZfPi8KHxwc8JpQ

* test(runner): cover the logs reaching the page and failing to

Claude-Session: https://claude.ai/code/session_01ShuAy8q8ZfPi8KHxwc8JpQ

* test(spec): cover the host pushing state.logs into the page

Claude-Session: https://claude.ai/code/session_01ShuAy8q8ZfPi8KHxwc8JpQ

* test(browser): drive console.error through to a fired property

Claude-Session: https://claude.ai/code/session_01ShuAy8q8ZfPi8KHxwc8JpQ

* fix(runner): report a log fetch the driver could not make

The comment claimed the failure was warned about; nothing warned, so a
device whose log fetch failed every step held noLogcatErrors on evidence
nobody collected.

Claude-Session: https://claude.ai/code/session_01ShuAy8q8ZfPi8KHxwc8JpQ

* test(runner): cover the silently dropped log fetch

Claude-Session: https://claude.ai/code/session_01ShuAy8q8ZfPi8KHxwc8JpQ

* fix(ci): read the route the spec reports, not the hierarchy dump

The android health gate grepped the trace for "AddTransactionScreen",
which occurs in exactly one place: the hierarchy dump, as a resource-id.
That is a debug artifact standing in for a fact the spec already reports,
and it was wrong in both directions. Against the real 8.4MB trace with
hierarchy stripped, the old gate failed a healthy 200-step run; against a
trace carrying the marker on a transition frame without the route ever
being reported, it passed and called it healthy.

It reads extractor_changes.route now, whose values come from SCREENS in
the spec, so the gate and the app agree on what being on a screen means.
routeOf answers null on a frame showing two screens, which is exactly the
frame the marker was matching.

The drift check grows to cover both new names: extract("route") and the
SCREENS key. The fixtures are re-derived keeping the route entry and no
hierarchy at all; the full artifacts and the fixtures give byte-identical
verdicts, which is what proves the coupling is gone.

Script and fixtures move together: either alone leaves the suite red.

* ci: check that the workflow references resolve

actionlint reads a local action's inputs but never checks its path
exists: uses: ./.github/actions/typo lints clean and fails only when the
job runs. Covers composite action paths, make targets including the ones
the examples matrix builds from $SANDERLING, and the scripts a run: step
invokes plus their executable bit.

Fails when it parses fewer references out of a file than that file
mentions, because a checker that matches nothing reports a safety it
never looked for.

* ci: lint the workflows on every pr

The workflows that fuzz the examples are dispatch-only, and GitHub will
not dispatch a workflow that is not on the default branch, so their first
real run is after merge. actionlint and the reference checker are the
only things that can fail before that.

actionlint is pinned by commit, and its tool version is pinned too so a
new release cannot change what CI enforces.

* ci: collapse the four workflows into one

Nine jobs written out one by one, each with its own steps and its own
calibrated seed and budget as literals. Triggers are pull requests, master
and v* tags, and a dispatch with no inputs.

Claude-Session: https://claude.ai/code/session_01ShuAy8q8ZfPi8KHxwc8JpQ

* ci: inline the two composite actions with one caller each

Both existed to give the matrix a per-target hook. folio-app and
headless-chrome stay: three and three callers.

Claude-Session: https://claude.ai/code/session_01ShuAy8q8ZfPi8KHxwc8JpQ

* ci: check that no run: block interpolates an expression

A ${{ }} lands in the script text before bash reads the line, and
actionlint only flags the contexts it already knows are attacker
controlled. Nothing enforced the rule the workflow follows. Also drops the
matrix table lookup, which has no table to read now.

Claude-Session: https://claude.ai/code/session_01ShuAy8q8ZfPi8KHxwc8JpQ

* ci(folio): name the run, not the fuzzer, in the clean-run message

Claude-Session: https://claude.ai/code/session_01ShuAy8q8ZfPi8KHxwc8JpQ

* docs: point at the workflow that holds the release secrets now

Claude-Session: https://claude.ai/code/session_01ShuAy8q8ZfPi8KHxwc8JpQ

* ci: name every job Category (variant), and gate the lot on one check

Follows the convention in antithesishq/bombadil: the display name is what
groups a run in the Actions UI, so Check (tests), Check (browser),
Check (workflows), Folio (android), Folio (ios), Folio (web), Replay UI,
Release and Docs. Every job carries a name, so none of them falls back to
its kebab-case id.

All checks passed needs all nine and runs with if: always(), so branch
protection has one check to point at and a skipped job cannot read as a
pass. Release and docs now gate on startsWith(github.ref, 'refs/tags/v')
alongside master, which is the form the trigger filter already uses.

Claude-Session: https://claude.ai/code/session_01ShuAy8q8ZfPi8KHxwc8JpQ

* ci: split the release job back in two

Collapsing them left the npm publish steps in a job holding contents:
write, because GoReleaser needs it, so npm ci ran its dependency lifecycle
scripts with a write-capable GITHUB_TOKEN in reach of the same job as a
live NPM_TOKEN. Release (npm) is back on contents: read and Release (cli)
keeps contents: write, which is what they each had before.

Each validates the tag from its own copy of the pattern rather than
waiting on a job that exists only to pass a string. Release (cli) is tags
only: there is no CLI to cut on a merge.

Claude-Session: https://claude.ai/code/session_01ShuAy8q8ZfPi8KHxwc8JpQ

* ci: run folio on pull requests

folio was skipped on pull requests, so ios, android and web only ever ran
after a merge. The three legs are 3 to 19 minutes and run in parallel, and
a superseded pull request run already cancels itself.

Claude-Session: https://claude.ai/code/session_01ShuAy8q8ZfPi8KHxwc8JpQ

* ci: draw each group as its own box in the run graph

The run graph boxes jobs together when they share the same dependencies and
the same dependents. All ten jobs fed only all-checks-passed, so all ten drew
as one pile. A gate per group gives each group a dependent that is exactly
that group.

Release and docs now need the checks, which they should have all along: npm
publish and the pages deploy ran on a merge without waiting for the test job.
Folio stays unblocked so a 20 minute leg does not wait on a 3 minute one.

Claude-Session: https://claude.ai/code/session_01ShuAy8q8ZfPi8KHxwc8JpQ
2026-08-16 13:23:53 +05:30

1354 lines
48 KiB
Go

// Package ioscompanion drives an iOS simulator through the native simulator
// companion. This file implements the DeviceDriver surface on top of the
// brand-free transport, supervises the companion child process, and recovers
// from a dropped connection with one in-place restart.
package ioscompanion
import (
"bytes"
"context"
"errors"
"fmt"
"image"
_ "image/png"
"io"
"net"
"os"
"os/exec"
"path/filepath"
"strings"
"sync"
"syscall"
"time"
"google.golang.org/grpc/codes"
"google.golang.org/grpc/status"
"github.com/priyanshujain/sanderling/internal/driver"
"github.com/priyanshujain/sanderling/internal/driver/ioscompanion/companionassets"
"github.com/priyanshujain/sanderling/internal/driver/ioscompanion/transport"
"github.com/priyanshujain/sanderling/internal/hierarchy"
)
// startupTimeout bounds how long New waits for the spawned companion to accept
// a connection and answer a health probe.
const startupTimeout = 30 * time.Second
// runnerStartupTimeout bounds the in-simulator runner's startup. The runner is
// hosted by a test session whose cold start is far slower than the companion's.
const runnerStartupTimeout = 120 * time.Second
// shutdownGrace bounds how long the companion child gets to exit after SIGTERM
// before it is killed. A variable so the kill-escalation test can shrink it.
var shutdownGrace = 15 * time.Second
// launchTimeout bounds a single app lifecycle RPC. The runner serves lifecycle
// inside its XCTest session, and a launch the simulator rejects sends that
// session down a recovery chain (a 120s accessibility wait, a spindump, then an
// idle wait) that answers minutes late or never. Callers reach Launch with an
// undeadlined context, since it runs before the run's duration clock starts, so
// the bound has to come from here or a wedged session hangs the run with no
// trace, no error, and no end. Kept under runnerStartupTimeout: launching an
// app inside a live session must cost less than cold-starting that session.
// A variable so the timeout test can shrink it.
var launchTimeout = 90 * time.Second
// launchRecoveryTimeout bounds the whole recovery a blown launch bound
// triggers, the session restart and the second attempt together. It keeps the
// launch path inside the three minutes testrun allows it, so what a user sees
// when the app really cannot be launched stays the driver's error rather than
// that backstop firing over the top of it. A variable so the bound test can
// shrink it.
var launchRecoveryTimeout = 60 * time.Second
// longPressHoldMilliseconds is how long LongPress holds the finger down.
const longPressHoldMilliseconds = 600
// Options configures a Driver.
type Options struct {
// UniqueDeviceIdentifier selects the booted simulator the companion drives.
UniqueDeviceIdentifier string
// BundleID is the app under test. Launch and Terminate act on it.
BundleID string
// AppPath is the .app bundle directory. Required for clear-state reinstall;
// when empty, clear state falls back to resetting the data container.
AppPath string
// ClearState resets the app to first-launch state while New runs, before
// any automation session attaches. Clear state is a property of the driver
// rather than of a launch: see Launch.
ClearState bool
// Output receives companion stdout and stderr plus driver warnings.
Output io.Writer
// DoubleTapGapMilliseconds overrides the synthesized double-tap gap.
DoubleTapGapMilliseconds float64
// These are test seams. Production leaves them nil and New wires the real
// extraction, spawn, dial and simctl calls.
spawnChild func(ctx context.Context, address string) (*exec.Cmd, error)
dialCompanion func(address string) (transport.Companion, error)
pickAddress func() (string, error)
spawnRunner func(ctx context.Context, address string) (*exec.Cmd, error)
dialRunner func(address string) (transport.Companion, error)
reinstallApp func(ctx context.Context) error
resetContainer func(ctx context.Context) error
terminateApp func(ctx context.Context) error
}
// Driver implements driver.DeviceDriver against an iOS simulator companion.
type Driver struct {
companion transport.Companion
udid string
bundleID string
appPath string
output io.Writer
// clearedBundleID names the app New (or NewDevice) reset to first-launch
// state before attaching, which is the only point in a run where clearing
// is safe. Launch refuses a clear-state request for anything else rather
// than reinstalling under a live session or reporting a reset that only
// ever reached another bundle.
clearedBundleID string
screenWidth int
screenHeight int
doubleTapGapMilliseconds float64
// mu guards Snapshot's hierarchy+screenshot pairing and the lastTap record.
mu sync.Mutex
lastTap struct {
x, y float64
set bool
}
clearStateWarned bool
// restart rebuilds the transport in place after a connection-level failure.
// It is a seam so tests exercise the supervision logic without spawning a
// real companion. restarting guards against re-entrant restarts.
restart func(ctx context.Context) error
restarting bool
address string
// resetContainer wipes the app data container for the clear-state fallback.
// A seam so tests skip the xcrun shell-out.
resetContainer func(ctx context.Context) error
// reinstallApp uninstalls and reinstalls the app bundle for clear-state.
// A seam so tests skip the simctl shell-outs.
reinstallApp func(ctx context.Context) error
// terminateApp stops the app before its state is cleared. A seam so tests
// skip the simctl shell-out.
terminateApp func(ctx context.Context) error
// grantPaste pre-authorizes the app's pasteboard access. A seam so tests
// skip the sqlite shell-out.
grantPaste func(ctx context.Context) error
// idleClock drives WaitForIdle's settle poll. A seam so tests substitute a
// fake clock and avoid the real settle cap.
idleClock Clock
spawnChild func(ctx context.Context, address string) (*exec.Cmd, error)
dial func(address string) (transport.Companion, error)
child *exec.Cmd
// The hybrid simulator companion pairs the legacy child (HID gestures,
// lifecycle, screenshot) with an in-simulator runner that serves
// collapse-free accessibility snapshots and native unicode typing.
// runnerClient is nil on the legacy-only path.
runnerClient transport.Companion
runnerChild *exec.Cmd
runnerAddress string
spawnRunner func(ctx context.Context, address string) (*exec.Cmd, error)
dialRunner func(address string) (transport.Companion, error)
hybrid bool
// pickRunnerAddress hands every bring-up a free loopback port, on the
// simulator and the device alike. One field, so no path can be wired
// without it.
pickRunnerAddress func() (string, error)
// Device-mode fields. On the physical-device path d.companion is the runner
// dialed over a usbmux tunnel, hybrid is false, and runnerClient is nil.
// coreDeviceID feeds devicectl; tunnel is the in-process usbmux forwarder
// bridging the host loopback port to the runner's device-side port.
deviceMode bool
coreDeviceID string
tunnel io.Closer
startTunnel func(ctx context.Context, hardwareUDID, localAddress, devicePort string) (io.Closer, error)
// processContext owns the companion child's lifetime: it is derived from
// New's context (so a canceled run still reaps the child) and canceled by
// Close. Spawning under a startup-scoped context would SIGTERM the child
// the moment startup finishes.
processContext context.Context
processCancel context.CancelFunc
// deviceLock is the exclusive claim on the target, held for the driver's
// whole life and released by Close.
deviceLock io.Closer
}
// acquireDeviceLock takes an exclusive advisory lock on the target so only one
// run drives it at a time. Two runs on one device interleave app lifecycle: the
// second run's uninstall and reinstall land under the first's live automation
// session, leaving its app proxies bound to a bundle the simulator no longer
// knows, and every later snapshot and launch on that session stalls. Failing
// fast beats recovering silently, since the other run owns the device and would
// be corrupted either way. The lock lives on the file descriptor, so a crashed
// run's claim is released by the kernel and never strands the device.
func acquireDeviceLock(udid string) (io.Closer, error) {
path := filepath.Join(os.TempDir(), "sanderling-ios-"+udid+".lock")
file, err := os.OpenFile(path, os.O_CREATE|os.O_RDWR, 0o644)
if err != nil {
return nil, fmt.Errorf("open device lock %s: %w", path, err)
}
if err := syscall.Flock(int(file.Fd()), syscall.LOCK_EX|syscall.LOCK_NB); err != nil {
file.Close()
return nil, fmt.Errorf(
"ios target %s is already driven by another sanderling run (lock %s); "+
"wait for that run to finish or point this one at a different device with --ios-device",
udid, path)
}
return file, nil
}
// New extracts the embedded companion, spawns it against the configured
// simulator, dials the transport, health-probes it, and caches the screen
// point dimensions. Call Close when done to stop the child.
func New(ctx context.Context, options Options) (*Driver, error) {
if options.UniqueDeviceIdentifier == "" {
return nil, errors.New("ios companion: UniqueDeviceIdentifier is required")
}
if options.ClearState && options.BundleID == "" {
return nil, errors.New("ios companion: clear-state needs BundleID: there is nothing to uninstall or wipe without it")
}
output := options.Output
if output == nil {
output = io.Discard
}
gap := options.DoubleTapGapMilliseconds
if gap <= 0 {
gap = DefaultDoubleTapGapMilliseconds
}
driverInstance := &Driver{
udid: options.UniqueDeviceIdentifier,
bundleID: options.BundleID,
appPath: options.AppPath,
output: output,
doubleTapGapMilliseconds: gap,
spawnChild: options.spawnChild,
dial: options.dialCompanion,
spawnRunner: options.spawnRunner,
dialRunner: options.dialRunner,
reinstallApp: options.reinstallApp,
resetContainer: options.resetContainer,
terminateApp: options.terminateApp,
hybrid: hybridCompanionEnabled(),
}
if driverInstance.spawnChild == nil {
driverInstance.spawnChild = driverInstance.realSpawnChild
}
if driverInstance.dial == nil {
driverInstance.dial = transport.Dial
}
if driverInstance.spawnRunner == nil {
driverInstance.spawnRunner = driverInstance.realSpawnRunner
}
if driverInstance.dialRunner == nil {
driverInstance.dialRunner = func(address string) (transport.Companion, error) {
return transport.DialRunner(address, driverInstance.udid, driverInstance.bundleID)
}
}
pickAddress := options.pickAddress
if pickAddress == nil {
pickAddress = pickLoopbackAddress
}
address, err := pickAddress()
if err != nil {
return nil, err
}
driverInstance.address = address
driverInstance.pickRunnerAddress = pickAddress
driverInstance.restart = driverInstance.respawnAndRedial
if driverInstance.resetContainer == nil {
driverInstance.resetContainer = driverInstance.resetDataContainer
}
if driverInstance.reinstallApp == nil {
driverInstance.reinstallApp = driverInstance.simctlReinstall
}
if driverInstance.terminateApp == nil {
driverInstance.terminateApp = driverInstance.simctlTerminate
}
driverInstance.grantPaste = driverInstance.grantPasteboardAccess
driverInstance.processContext, driverInstance.processCancel = context.WithCancel(ctx)
lock, err := acquireDeviceLock(driverInstance.udid)
if err != nil {
driverInstance.processCancel()
return nil, err
}
driverInstance.deviceLock = lock
if options.ClearState {
if err := driverInstance.clearAppState(ctx); err != nil {
driverInstance.Close()
return nil, err
}
driverInstance.clearedBundleID = options.BundleID
}
if err := driverInstance.bringUp(ctx); err != nil {
driverInstance.Close()
return nil, err
}
if driverInstance.hybrid {
if err := driverInstance.bringUpRunner(ctx); err != nil {
driverInstance.Close()
return nil, fmt.Errorf("simulator runner: %w (set SANDERLING_SIMULATOR_COMPANION=legacy to bypass)", err)
}
}
description, err := driverInstance.companion.Describe(ctx)
if err != nil {
driverInstance.Close()
return nil, fmt.Errorf("describe target: %w", err)
}
driverInstance.screenWidth = description.WidthPoints
driverInstance.screenHeight = description.HeightPoints
return driverInstance, nil
}
// bringUp spawns the companion child, waits for the listener, dials, and
// confirms health. It is used by New and by the in-place restart.
func (d *Driver) bringUp(ctx context.Context) error {
startupCtx, cancel := context.WithTimeout(ctx, startupTimeout)
defer cancel()
child, err := d.spawnChild(d.processContext, d.address)
if err != nil {
return fmt.Errorf("spawn companion: %w", err)
}
d.child = child
if err := waitForListener(startupCtx, d.address); err != nil {
d.stopChild()
return fmt.Errorf("companion listener: %w", err)
}
companion, err := d.dial(d.address)
if err != nil {
d.stopChild()
return fmt.Errorf("dial companion: %w", err)
}
d.companion = companion
if err := d.waitForHealth(startupCtx); err != nil {
_ = companion.Close()
d.stopChild()
return fmt.Errorf("companion health: %w", err)
}
return nil
}
// waitForHealth probes AccessibilityInfo until it succeeds or the context
// expires. A successful describe-all means the companion is attached to the
// simulator and ready to serve.
func (d *Driver) waitForHealth(ctx context.Context) error {
ticker := time.NewTicker(250 * time.Millisecond)
defer ticker.Stop()
for {
if _, err := d.companion.AccessibilityInfo(ctx); err == nil {
return nil
}
select {
case <-ctx.Done():
return ctx.Err()
case <-ticker.C:
}
}
}
// respawnAndRedial tears down the current transports and children, then brings
// fresh ones up at the same addresses. Used as the supervision restart. On the
// hybrid path both halves restart together: their failure modes overlap (a
// rebooted simulator drops both) and one orchestration keeps recovery simple.
func (d *Driver) respawnAndRedial(ctx context.Context) error {
// The dead transports are closed but kept in place until their fresh
// replacements land: if the restart fails, later calls error gracefully on
// the closed transport instead of dereferencing nil, and a later incident
// earns another restart attempt.
if d.companion != nil {
_ = d.companion.Close()
}
d.stopChild()
if d.runnerClient != nil {
_ = d.runnerClient.Close()
}
d.stopRunnerChild()
if err := d.bringUp(ctx); err != nil {
return err
}
if d.hybrid {
return d.bringUpRunner(ctx)
}
return nil
}
// hybridCompanionEnabled reports whether the simulator driver should pair the
// legacy companion with the in-simulator runner. The hybrid is the default;
// SANDERLING_SIMULATOR_COMPANION=legacy forces the legacy companion alone.
func hybridCompanionEnabled() bool {
return os.Getenv("SANDERLING_SIMULATOR_COMPANION") != "legacy"
}
// bringUpRunner spawns the in-simulator runner, waits for its listener, dials,
// and confirms it serves snapshots. Used by New and the in-place restart.
func (d *Driver) bringUpRunner(ctx context.Context) error {
startupCtx, cancel := context.WithTimeout(ctx, runnerStartupTimeout)
defer cancel()
// A fresh port every bring-up: after a restart the dying session's
// listener may still answer on the old port and would satisfy the wait
// below with a dead server.
address, err := d.pickRunnerAddress()
if err != nil {
return err
}
d.runnerAddress = address
child, err := d.spawnRunner(d.processContext, d.runnerAddress)
if err != nil {
return fmt.Errorf("spawn runner: %w", err)
}
d.runnerChild = child
if err := waitForListener(startupCtx, d.runnerAddress); err != nil {
d.stopRunnerChild()
return fmt.Errorf("runner listener: %w", err)
}
client, err := d.dialRunner(d.runnerAddress)
if err != nil {
d.stopRunnerChild()
return fmt.Errorf("dial runner: %w", err)
}
ticker := time.NewTicker(250 * time.Millisecond)
defer ticker.Stop()
for {
if _, healthErr := client.AccessibilityInfo(startupCtx); healthErr == nil {
break
}
select {
case <-startupCtx.Done():
_ = client.Close()
d.stopRunnerChild()
return fmt.Errorf("runner health: %w", startupCtx.Err())
case <-ticker.C:
}
}
d.runnerClient = client
return nil
}
// withRecovery runs call, and on a connection-level failure performs one
// in-place restart before retrying the call once. A non-connection error, or a
// second failure of any kind, surfaces to the caller. The restart budget is per
// failure incident: each healthy call resets restarting to false, so a later
// drop earns its own single restart.
func (d *Driver) withRecovery(ctx context.Context, call func() error) error {
err := call()
if err == nil || !isConnectionError(err) || d.restarting || d.restart == nil {
return err
}
d.restarting = true
defer func() { d.restarting = false }()
fmt.Fprintf(d.output, "companion connection lost (%v); restarting once\n", err)
// The restart runs under the driver's own lifetime context, not the
// failed call's: an action whose deadline already expired must not doom
// the recovery that later actions depend on.
restartCtx := d.processContext
if restartCtx == nil {
restartCtx = ctx
}
if restartErr := d.restart(restartCtx); restartErr != nil {
return fmt.Errorf("companion restart failed: %w (original: %v)", restartErr, err)
}
return call()
}
// isConnectionError reports whether err is a dropped-connection signal that a
// restart can recover from: the transport's unavailable sentinel, a gRPC
// Unavailable status, or an EOF.
func isConnectionError(err error) bool {
if err == nil {
return false
}
if errors.Is(err, io.EOF) || errors.Is(err, transport.ErrCompanionUnavailable) {
return true
}
if statusValue, ok := status.FromError(err); ok {
return statusValue.Code() == codes.Unavailable
}
return false
}
// isBudgetExpiry reports whether err is a call that outlived its bound rather
// than a failure the transport can name. Each transport says so differently:
// the runner wraps the context's error when cancellation has landed and the
// connection's i/o timeout when the deadline it armed from that context fires
// first, and the legacy companion returns a gRPC status. Only an expiry earns
// a session restart; an error the runner reports has already said what a fresh
// session would say.
func isBudgetExpiry(err error) bool {
if err == nil {
return false
}
if errors.Is(err, context.DeadlineExceeded) || errors.Is(err, os.ErrDeadlineExceeded) {
return true
}
if statusValue, ok := status.FromError(err); ok {
return statusValue.Code() == codes.DeadlineExceeded
}
return false
}
func (d *Driver) Launch(ctx context.Context, bundleID string, clearState bool, env map[string]string) error {
if bundleID != "" {
d.bundleID = bundleID
}
if len(env) > 0 {
// The launch Start message carries an env map, but this backend does
// not pass it through: passing it would change the app's process
// environment in ways the rest of the run does not account for. Reject
// loudly rather than silently dropping the request.
return errors.New("ios companion: launch with environment variables is unsupported on this backend")
}
if clearState && (d.clearedBundleID == "" || d.clearedBundleID != d.bundleID) {
// Clearing here would uninstall and reinstall the app underneath a live
// automation session, which is what races FrontBoard's registration and
// leaves the session launching a bundle FrontBoard has not registered.
return fmt.Errorf("ios companion: clear-state must be requested when the driver is created (Options.ClearState) "+
"for the bundle being launched; this backend cleared %q before its automation session existed, not %q",
d.clearedBundleID, d.bundleID)
}
// Terminate first so the launch is a clean cold start regardless of the
// app's prior state. A not-running app is not an error here.
_ = d.lifecycleCall(ctx, func(callCtx context.Context, companion transport.Companion) error {
return companion.Terminate(callCtx, d.bundleID)
})
// Grant the app pasteboard access before it runs so unicode input (which
// must go through the pasteboard, since HID cannot express it) never trips
// the iOS paste-permission prompt. clearState reinstall resets the grant,
// so it is reapplied on every launch. Best effort: if it fails, the paste
// path still handles the prompt, just slower. The hybrid path types
// natively and never touches the pasteboard, so it skips the grant.
if d.bundleID != "" && d.runnerTyper() == nil {
if err := d.grantPaste(ctx); err != nil {
fmt.Fprintf(d.output, "grant pasteboard access failed (continuing): %v\n", err)
}
}
if err := d.launchWithSessionRecovery(ctx); err != nil {
return fmt.Errorf("launch %s: %w", d.bundleID, err)
}
return nil
}
// launchWithSessionRecovery runs the launch RPC and, when it blows its own
// bound, replaces the session and launches again.
//
// A launch the simulator refuses, which is what a clear-state reinstall racing
// FrontBoard's registration produces, never comes back as an error: XCTest
// records the refusal as a test failure the runner cannot observe, then holds
// the session's main thread for about four minutes walking a diagnostic chain
// (a 120s accessibility wait, a spindump, an idle wait). So there is no error
// text to key a retry on, only the expired bound, and every later call queues
// behind the same wedge. Only a session that never served the refused launch
// can serve the retry, which is why this restarts rather than calls again.
func (d *Driver) launchWithSessionRecovery(ctx context.Context) error {
launch := func(callCtx context.Context, companion transport.Companion) error {
return companion.Launch(callCtx, d.bundleID, true)
}
err := d.lifecycleCall(ctx, launch)
// A caller whose own budget ran out gets no restart: the bound that expired
// was the caller's to spend, and the second attempt would inherit it dead.
if !isBudgetExpiry(err) || ctx.Err() != nil || d.restart == nil {
return err
}
fmt.Fprintf(d.output, "launch %s blew its %v bound (%v); restarting the session and launching once more\n",
d.bundleID, launchTimeout, err)
// The restart runs under the driver's own lifetime context for the same
// reason withRecovery's does, while the second attempt stays on the
// caller's. Both end at one deadline, so a launch that already spent
// launchTimeout cannot then wait out a session cold start on top of it.
recoveryDeadline := time.Now().Add(launchRecoveryTimeout)
restartCtx := d.processContext
if restartCtx == nil {
restartCtx = ctx
}
restartCtx, cancelRestart := context.WithDeadline(restartCtx, recoveryDeadline)
defer cancelRestart()
if restartErr := d.restart(restartCtx); restartErr != nil {
return fmt.Errorf("session restart failed: %w (original: %v)", restartErr, err)
}
relaunchCtx, cancelRelaunch := context.WithDeadline(ctx, recoveryDeadline)
defer cancelRelaunch()
return d.lifecycleCall(relaunchCtx, launch)
}
// lifecycleCall runs an app lifecycle RPC against lifecycleCompanion under a
// launchTimeout-bounded context, with the usual one-restart recovery. The
// companion is resolved inside the retry so a restart's replacement client
// serves the second attempt.
func (d *Driver) lifecycleCall(ctx context.Context, call func(context.Context, transport.Companion) error) error {
boundedCtx, cancel := context.WithTimeout(ctx, launchTimeout)
defer cancel()
return d.withRecovery(boundedCtx, func() error {
return call(boundedCtx, d.lifecycleCompanion())
})
}
// lifecycleCompanion is the transport that owns app launch and terminate: the
// in-simulator runner when the hybrid is active, otherwise the legacy
// companion. Lifecycle performed outside the runner's automation session
// leaves the session's app proxies bound to dead processes, after which
// snapshots hang and typing asserts.
func (d *Driver) lifecycleCompanion() transport.Companion {
if d.runnerClient != nil {
return d.runnerClient
}
return d.companion
}
// clearAppState resets the app to a first-launch state. With an app path it
// uninstalls and reinstalls; without one it falls back to wiping the app's data
// container and warns once that a full reinstall needs the app path. Called
// only from construction, before any automation session is attached to the app.
func (d *Driver) clearAppState(ctx context.Context) error {
// Nothing may be writing to the state while it goes, which is the ordering
// Launch used to hold: uninstall copes with a running app, deleting the
// data container out from under one does not. Best effort, because an app
// that is not running reports a failure that means nothing here.
_ = d.terminateApp(ctx)
if d.appPath != "" {
if err := d.reinstallApp(ctx); err != nil {
return fmt.Errorf("reinstall %s: %w", d.appPath, err)
}
return nil
}
if !d.clearStateWarned {
fmt.Fprintln(d.output, "clear-state requested without an app path: resetting the data container only; pass the app path for a full reinstall")
d.clearStateWarned = true
}
return d.resetContainer(ctx)
}
// simctlReinstall uninstalls and reinstalls the app bundle via simctl. App
// lifecycle stays with simctl: the companion's install RPC misreads current
// simulator targets' architectures and rejects valid bundles.
// A failed uninstall ends the reinstall: `simctl install` over an installed app
// carries its data container across, so clear-state would be reported without
// happening. Uninstalling an app that is not installed exits 0, so there is no
// benign failure here to sort out from a real one.
func (d *Driver) simctlReinstall(ctx context.Context) error {
if output, err := exec.CommandContext(ctx, "xcrun", "simctl", "uninstall", d.udid, d.bundleID).CombinedOutput(); err != nil {
return fmt.Errorf("simctl uninstall %s: %w: %s", d.bundleID, err, strings.TrimSpace(string(output)))
}
output, err := exec.CommandContext(ctx, "xcrun", "simctl", "install", d.udid, d.appPath).CombinedOutput()
if err != nil {
return fmt.Errorf("simctl install: %w: %s", err, strings.TrimSpace(string(output)))
}
return nil
}
// simctlTerminate stops the app under test. Launch used to terminate through
// the automation session before clearing; the clear now runs before any session
// exists, so simctl is what is left to stop the app with. An app that is not
// running reports a failure that means nothing to the caller, which is why
// clearAppState treats this as best effort.
func (d *Driver) simctlTerminate(ctx context.Context) error {
if output, err := exec.CommandContext(ctx, "xcrun", "simctl", "terminate", d.udid, d.bundleID).CombinedOutput(); err != nil {
return fmt.Errorf("simctl terminate %s: %w: %s", d.bundleID, err, strings.TrimSpace(string(output)))
}
return nil
}
// grantPasteboardAccess authorizes the app to read the pasteboard without the
// iOS permission prompt, by writing an allow row into the simulator's privacy
// (TCC) database. This is the simulator counterpart to `simctl privacy grant`,
// which does not expose the pasteboard service. Without it, every unicode input
// (which must paste, since HID cannot express unicode) blocks on a modal that
// costs seconds; with it the paste lands in one frame.
func (d *Driver) grantPasteboardAccess(ctx context.Context) error {
databasePath := filepath.Join(
os.Getenv("HOME"), "Library", "Developer", "CoreSimulator", "Devices",
d.udid, "data", "Library", "TCC", "TCC.db",
)
if _, err := os.Stat(databasePath); err != nil {
return fmt.Errorf("locate privacy database: %w", err)
}
statement := fmt.Sprintf(
"INSERT OR REPLACE INTO access "+
"(service,client,client_type,auth_value,auth_reason,auth_version,indirect_object_identifier) "+
"VALUES ('kTCCServicePasteboard','%s',0,2,4,1,'UNUSED');",
d.bundleID,
)
output, err := exec.CommandContext(ctx, "sqlite3", databasePath, statement).CombinedOutput()
if err != nil {
return fmt.Errorf("write privacy grant: %w: %s", err, strings.TrimSpace(string(output)))
}
return nil
}
// resetDataContainer deletes the contents of the app's data container so the
// next launch starts with empty storage.
func (d *Driver) resetDataContainer(ctx context.Context) error {
output, err := exec.CommandContext(ctx, "xcrun", "simctl", "get_app_container", d.udid, d.bundleID, "data").Output()
if err != nil {
return fmt.Errorf("get app container: %w", err)
}
container := string(bytes.TrimSpace(output))
if container == "" {
return nil
}
entries, err := os.ReadDir(container)
if err != nil {
return fmt.Errorf("read app container: %w", err)
}
for _, entry := range entries {
if err := os.RemoveAll(filepath.Join(container, entry.Name())); err != nil {
return fmt.Errorf("clear app container: %w", err)
}
}
return nil
}
func (d *Driver) Terminate(ctx context.Context) error {
return d.lifecycleCall(ctx, func(callCtx context.Context, companion transport.Companion) error {
return companion.Terminate(callCtx, d.bundleID)
})
}
func (d *Driver) Tap(ctx context.Context, x, y int) error {
d.mu.Lock()
d.lastTap.x = float64(x)
d.lastTap.y = float64(y)
d.lastTap.set = true
d.mu.Unlock()
return d.withRecovery(ctx, func() error {
return d.companion.SendHID(ctx, tapEvents(float64(x), float64(y))...)
})
}
func (d *Driver) DoubleTap(ctx context.Context, x, y int) error {
return d.withRecovery(ctx, func() error {
return d.companion.SendHID(ctx, doubleTapEvents(float64(x), float64(y), d.doubleTapGapMilliseconds)...)
})
}
func (d *Driver) LongPress(ctx context.Context, x, y int) error {
return d.withRecovery(ctx, func() error {
return d.companion.SendHID(ctx, longPressEvents(float64(x), float64(y), longPressHoldMilliseconds)...)
})
}
func (d *Driver) Swipe(ctx context.Context, fromX, fromY, toX, toY int, duration time.Duration) error {
seconds := duration.Seconds()
if seconds <= 0 {
seconds = 0.25
}
return d.withRecovery(ctx, func() error {
return d.companion.SendHID(ctx, transport.SwipeEvent(
float64(fromX), float64(fromY), float64(toX), float64(toY), seconds))
})
}
func (d *Driver) PressKey(ctx context.Context, key string) error {
if d.textEditor() != nil {
return d.withRecovery(ctx, func() error {
return d.textEditor().PressKey(ctx, key)
})
}
usage, ok := pressKeyUsage(key)
if !ok {
return fmt.Errorf("ios companion: unsupported key %q", key)
}
return d.withRecovery(ctx, func() error {
return d.companion.SendHID(ctx, transport.KeyDown(usage), transport.KeyUp(usage))
})
}
// textEditor returns the companion's native text-editing capability, or nil
// when the transport does not implement it. Resolved per call because a
// restart replaces d.companion.
func (d *Driver) textEditor() transport.TextEditor {
if editor, ok := d.companion.(transport.TextEditor); ok {
return editor
}
return nil
}
// runnerTyper returns the in-simulator runner's native typing capability, or
// nil outside the hybrid path. Resolved per call because a restart replaces
// d.runnerClient.
func (d *Driver) runnerTyper() transport.TextTyper {
if typer, ok := d.runnerClient.(transport.TextTyper); ok {
return typer
}
return nil
}
// pressKeyUsage maps the logical key names mobile runs emit to a HID usage.
// Only Return/Enter has a hardware-keyboard equivalent on the simulator; other
// names (notably "back" and "home") have no HID key and report unsupported.
func pressKeyUsage(key string) (uint32, bool) {
switch key {
case "enter", "return", "Enter", "Return":
return usageReturn, true
default:
return 0, false
}
}
func (d *Driver) TapSelector(ctx context.Context, selector string) error {
x, y, err := d.resolveSelectorCenter(ctx, selector)
if err != nil {
return err
}
return d.Tap(ctx, x, y)
}
func (d *Driver) DoubleTapSelector(ctx context.Context, selector string) error {
x, y, err := d.resolveSelectorCenter(ctx, selector)
if err != nil {
return err
}
return d.DoubleTap(ctx, x, y)
}
// resolveSelectorCenter fetches a fresh hierarchy and returns the center of the
// first element matching selector.
func (d *Driver) resolveSelectorCenter(ctx context.Context, selector string) (int, int, error) {
hierarchyJSON, err := d.Hierarchy(ctx)
if err != nil {
return 0, 0, err
}
tree, err := hierarchy.Parse(hierarchyJSON)
if err != nil {
return 0, 0, fmt.Errorf("parse hierarchy: %w", err)
}
element := tree.Find(selector)
if element == nil {
return 0, 0, fmt.Errorf("selector %q matched no element", selector)
}
x, y := element.Bounds.Center()
return x, y, nil
}
func (d *Driver) InputText(ctx context.Context, text string) error {
// Hybrid path. Mappable text rides one HID stream: select-all chord plus
// keystrokes, atomic and strictly ordered on a single channel. Unicode
// (which HID cannot express) is typed natively by the runner after the
// chord; chord and typing ride different channels with no ordering
// guarantee between them, so the clear is verified through a snapshot
// before the first keystroke goes out.
if typer := d.runnerTyper(); typer != nil {
if !usesPasteboard(text) {
events := append(clearFieldEvents(), keyPressEvents(typeStringPresses(text))...)
return d.withRecovery(ctx, func() error {
return d.companion.SendHID(ctx, events...)
})
}
return d.withRecovery(ctx, func() error {
if err := d.companion.SendHID(ctx, clearFieldEvents()...); err != nil {
return fmt.Errorf("clear field: %w", err)
}
d.waitFieldCleared(ctx)
return d.runnerTyper().TypeText(ctx, text, false)
})
}
// A text-editing companion replaces the field's content natively, which
// covers unicode without the pasteboard and its permission dialog.
if d.textEditor() != nil {
return d.withRecovery(ctx, func() error {
return d.textEditor().InputText(ctx, text)
})
}
// The field target is only needed for the pasteboard path. Resolving it
// requires a describe-all, so the fast keyboard path skips that round-trip
// and lets inputText send the key presses directly.
var field fieldTarget
if usesPasteboard(text) {
field = d.resolveInputField(ctx)
}
return inputText(ctx, d.makeRunner(), text, field)
}
// fieldClearedWaitCap and fieldClearedPoll bound the verify-cleared loop
// between the HID clear chord and the runner's native typing.
const fieldClearedWaitCap = 1200 * time.Millisecond
const fieldClearedPoll = 150 * time.Millisecond
// waitFieldCleared polls the focused field (the editable element under the
// last tap) until its value reads empty, so the clear chord has demonstrably
// landed before typing starts on the other channel. Best effort: when the
// field cannot be resolved or the cap elapses, typing proceeds anyway.
func (d *Driver) waitFieldCleared(ctx context.Context) {
d.mu.Lock()
tap := d.lastTap
d.mu.Unlock()
if !tap.set {
return
}
deadline := time.Now().Add(fieldClearedWaitCap)
for time.Now().Before(deadline) {
dump, err := d.describeAllRaw(ctx)
if err != nil {
return
}
cleared := true
for _, element := range decodeDump(dump) {
if !isEditable(element.Type) {
continue
}
frame := element.Frame
if !finite(frame.X) || !finite(frame.Y) || !finite(frame.Width) || !finite(frame.Height) {
continue
}
if tap.x < frame.X || tap.x > frame.X+frame.Width ||
tap.y < frame.Y || tap.y > frame.Y+frame.Height {
continue
}
if value := stringValue(element.AXValue); value != "" && value != emptyFieldValueSentinel {
cleared = false
}
break
}
if cleared {
return
}
select {
case <-ctx.Done():
return
case <-time.After(fieldClearedPoll):
}
}
}
// resolveInputField finds the editable element under the last tap so the
// pasteboard fallback can confirm the paste landed and refocus after dismissing
// the permission dialog. The runner always taps a field before typing, so
// lastTap names the focus point. An empty fieldTarget is returned when no
// editable element contains the tap (the fast keyboard path ignores it).
func (d *Driver) resolveInputField(ctx context.Context) fieldTarget {
d.mu.Lock()
tap := d.lastTap
d.mu.Unlock()
if !tap.set {
return fieldTarget{}
}
dump, err := d.describeAll(ctx)
if err != nil {
return fieldTarget{}
}
for _, element := range decodeDump(dump) {
if !isEditable(element.Type) {
continue
}
frame := element.Frame
if !finite(frame.X) || !finite(frame.Y) || !finite(frame.Width) || !finite(frame.Height) {
continue
}
if tap.x < frame.X || tap.x > frame.X+frame.Width ||
tap.y < frame.Y || tap.y > frame.Y+frame.Height {
continue
}
return fieldTarget{
identifier: stringValue(element.AXUniqueID),
centerX: frame.X + frame.Width/2,
centerY: frame.Y + frame.Height/2,
}
}
return fieldTarget{}
}
func (d *Driver) EraseText(ctx context.Context, characterCount int) error {
if d.textEditor() != nil {
return d.withRecovery(ctx, func() error {
return d.textEditor().EraseText(ctx, characterCount)
})
}
return eraseText(ctx, d.makeRunner(), characterCount)
}
func (d *Driver) Hierarchy(ctx context.Context) (string, error) {
dump, err := d.describeAll(ctx)
if err != nil {
return "", err
}
mapped, err := MapHierarchy(dump, d.screenWidth, d.screenHeight)
if err != nil {
return "", err
}
return string(mapped), nil
}
func (d *Driver) Screenshot(ctx context.Context) (driver.Image, error) {
var data []byte
err := d.withRecovery(ctx, func() error {
var screenshotErr error
data, _, screenshotErr = d.companion.Screenshot(ctx)
return screenshotErr
})
if err != nil {
return driver.Image{}, fmt.Errorf("screenshot: %w", err)
}
return decodeScreenshot(data)
}
func (d *Driver) Snapshot(ctx context.Context) (string, driver.Image, error) {
d.mu.Lock()
defer d.mu.Unlock()
// The hierarchy and the screenshot ride different transports on the
// hybrid path, so they are captured concurrently. Only the hierarchy leg
// runs under withRecovery: two concurrent recoveries would race the
// restart bookkeeping, and a screenshot connection failure surfaces as a
// plain error that the next serialized call recovers from. The goroutine
// works through a captured local because a hierarchy-leg recovery
// reassigns d.companion mid-flight; a screenshot against the torn-down
// transport then fails as a plain error rather than racing the field.
var data []byte
screenshotDone := make(chan error, 1)
companion := d.companion
go func() {
imageData, _, callErr := companion.Screenshot(ctx)
data = imageData
screenshotDone <- callErr
}()
dump, err := d.describeAll(ctx)
screenshotErr := <-screenshotDone
if err != nil {
return "", driver.Image{}, err
}
if screenshotErr != nil {
return "", driver.Image{}, fmt.Errorf("screenshot: %w", screenshotErr)
}
mapped, err := MapHierarchy(dump, d.screenWidth, d.screenHeight)
if err != nil {
return "", driver.Image{}, err
}
image, err := decodeScreenshot(data)
if err != nil {
return string(mapped), driver.Image{}, err
}
return string(mapped), image, nil
}
// WaitForIdle polls the hierarchy until it settles. The duration argument is
// ignored: the ported settle constants (StabilityPollCap and friends) own the
// cap, matching the companion's own settle behavior.
func (d *Driver) WaitForIdle(ctx context.Context, _ time.Duration) error {
clock := d.idleClock
if clock == nil {
clock = SystemClock()
}
PollUntilStable(ctx, clock, func() *hierarchy.Tree {
dump, err := d.describeAllRaw(ctx)
if err != nil || dumpIsCollapsed(dump) {
// A collapsed dump is the bridge mid-transition; report it
// transitional so the streak resets and the poll waits for the
// real tree rather than settling on the empty shell.
return nil
}
mapped, err := MapHierarchy(dump, d.screenWidth, d.screenHeight)
if err != nil {
return nil
}
tree, err := hierarchy.Parse(string(mapped))
if err != nil {
return nil
}
return tree
})
return nil
}
// RecentLogs returns no entries: the companion log RPC is a follow-up, so v1
// reports an empty slice rather than failing. Every property reading state.logs
// therefore holds vacuously on iOS and nothing says so; closing it means
// tailing idb's streaming log RPC and mapping os_log levels onto the
// single-letter scale driver.LogEntry declares.
func (d *Driver) RecentLogs(_ context.Context, _ time.Time, _ string) ([]driver.LogEntry, error) {
return []driver.LogEntry{}, nil
}
func (d *Driver) Metrics(_ context.Context, _ string) (driver.Metrics, error) {
return driver.Metrics{}, nil
}
func (d *Driver) Health(_ context.Context) (driver.Health, error) {
return driver.Health{Ready: true, Platform: "ios"}, nil
}
// ForegroundApp reports the foreground app. It returns the app under test when
// it is running; otherwise it names another running user app, or "" when none
// is. The companion exposes process state but not a foreground flag, so "the
// app under test is running" stands in for "in the foreground".
func (d *Driver) ForegroundApp(ctx context.Context) (string, error) {
var apps []transport.InstalledApp
if err := d.withRecovery(ctx, func() error {
var listErr error
apps, listErr = d.companion.ListApps(ctx)
return listErr
}); err != nil {
return "", err
}
other := ""
for _, app := range apps {
if app.ProcessState != transport.ProcessStateRunning {
continue
}
if app.BundleID == d.bundleID {
return d.bundleID, nil
}
if app.InstallType == "user" && other == "" {
other = app.BundleID
}
}
return other, nil
}
// collapsedDumpRetries and collapsedDumpDelay bound how long describeAll waits
// out a collapsed accessibility dump. The bridge briefly reports only the app
// shell (no UI content) during cold start and screen transitions; it recovers
// within a few hundred milliseconds. Re-fetching past the collapse keeps the
// runner from acting on, and snapshotting, an empty tree.
const collapsedDumpRetries = 6
const collapsedDumpDelay = 150 * time.Millisecond
// snapshotCompanion is the transport that serves accessibility dumps: the
// in-simulator runner when the hybrid is active (its snapshots never collapse),
// otherwise the legacy companion.
func (d *Driver) snapshotCompanion() transport.Companion {
if d.runnerClient != nil {
return d.runnerClient
}
return d.companion
}
// describeAllRaw fetches the flat accessibility dump with one-restart recovery
// and no collapse handling. The settle loop uses it: it treats a collapsed dump
// as transitional itself, so an inner retry here would double the wait.
func (d *Driver) describeAllRaw(ctx context.Context) ([]byte, error) {
var dump []byte
err := d.withRecovery(ctx, func() error {
info, infoErr := d.snapshotCompanion().AccessibilityInfo(ctx)
if infoErr != nil {
return infoErr
}
dump = []byte(info)
return nil
})
return dump, err
}
// describeAll fetches the flat accessibility dump, retrying past a transient
// collapsed dump so one-shot reads (Snapshot, Hierarchy) see real UI content.
func (d *Driver) describeAll(ctx context.Context) ([]byte, error) {
dump, err := d.describeAllRaw(ctx)
if err != nil {
return dump, err
}
for attempt := 0; attempt < collapsedDumpRetries && dumpIsCollapsed(dump); attempt++ {
select {
case <-ctx.Done():
return dump, nil
case <-time.After(collapsedDumpDelay):
}
next, nextErr := d.describeAllRaw(ctx)
if nextErr != nil {
return dump, nil
}
dump = next
}
return dump, nil
}
// makeRunner builds the input runner backed by the current transport. The text
// runner does not route through withRecovery: it is invoked synchronously
// inside a single InputText call and a mid-paste connection drop surfaces as a
// normal error the runner retries.
func (d *Driver) makeRunner() runner {
return simctlRunner{companion: d.companion, udid: d.udid}
}
// Close stops the companion and runner children and releases the transports.
func (d *Driver) Close() {
if d.companion != nil {
_ = d.companion.Close()
d.companion = nil
}
d.stopChild()
if d.runnerClient != nil {
_ = d.runnerClient.Close()
d.runnerClient = nil
}
d.stopRunnerChild()
d.stopTunnel()
if d.processCancel != nil {
d.processCancel()
}
if d.deviceLock != nil {
_ = d.deviceLock.Close()
d.deviceLock = nil
}
}
// stopTunnel closes the in-process usbmux forwarder on the device path. Closing
// its listener ends the accept loop and lets the open bridges drain; a nil
// tunnel (the simulator path) is a no-op.
func (d *Driver) stopTunnel() {
tunnel := d.tunnel
d.tunnel = nil
if tunnel != nil {
_ = tunnel.Close()
}
}
// stopChild terminates the companion child gracefully (SIGTERM, grace window,
// then SIGKILL) so it leaves no orphan behind.
func (d *Driver) stopChild() {
child := d.child
d.child = nil
stopProcess(child)
}
// stopRunnerChild terminates the runner's hosting session the same way. The
// session tears down its in-simulator children on SIGTERM; killing it outright
// would orphan them.
func (d *Driver) stopRunnerChild() {
child := d.runnerChild
d.runnerChild = nil
stopProcess(child)
}
func stopProcess(child *exec.Cmd) {
if child == nil || child.Process == nil {
return
}
if err := child.Process.Signal(syscall.SIGTERM); err != nil {
_ = child.Process.Kill()
_ = child.Wait()
return
}
done := make(chan struct{})
go func() {
_ = child.Wait()
close(done)
}()
select {
case <-done:
case <-time.After(shutdownGrace):
_ = child.Process.Kill()
<-done
}
}
// decodeScreenshot sniffs the PNG magic and decodes the pixel dimensions. The
// companion leaves image_format empty in practice, so the magic bytes are the
// only reliable format signal. No scaling is applied: the dimensions are pixels.
func decodeScreenshot(data []byte) (driver.Image, error) {
if len(data) < 8 || !bytes.HasPrefix(data, []byte("\x89PNG\r\n\x1a\n")) {
return driver.Image{}, errors.New("screenshot: response is not a PNG")
}
config, _, err := image.DecodeConfig(bytes.NewReader(data))
if err != nil {
return driver.Image{}, fmt.Errorf("decode screenshot: %w", err)
}
return driver.Image{PNG: data, Width: config.Width, Height: config.Height}, nil
}
// realSpawnChild extracts the embedded companion and starts it on the given
// address. Cancel sends SIGTERM so the companion detaches cleanly from the
// simulator; WaitDelay bounds the grace before the runtime kills it.
func (d *Driver) realSpawnChild(ctx context.Context, address string) (*exec.Cmd, error) {
extractDirectory := filepath.Join(os.TempDir(), "sanderling-companion")
binaryPath, err := companionassets.Extract(extractDirectory)
if err != nil {
return nil, fmt.Errorf("extract companion: %w", err)
}
_, port, err := net.SplitHostPort(address)
if err != nil {
return nil, err
}
command := exec.CommandContext(ctx, binaryPath, "--udid", d.udid, "--grpc-port", port)
command.Stdout = d.output
command.Stderr = d.output
// The companion echoes its whole environment into the run log at startup,
// so it gets a minimal one: secrets in the parent environment must never
// reach run artifacts.
command.Env = []string{
"HOME=" + os.Getenv("HOME"),
"PATH=/usr/bin:/bin",
"TMPDIR=" + os.TempDir(),
}
command.Cancel = func() error { return command.Process.Signal(syscall.SIGTERM) }
command.WaitDelay = shutdownGrace
if err := command.Start(); err != nil {
return nil, fmt.Errorf("start companion: %w", err)
}
fmt.Fprintf(d.output, "companion pid=%d listening on %s\n", command.Process.Pid, address)
return command, nil
}
// pickLoopbackAddress reserves a free loopback port and returns its address.
func pickLoopbackAddress() (string, error) {
listener, err := net.Listen("tcp", "127.0.0.1:0")
if err != nil {
return "", err
}
defer listener.Close()
return listener.Addr().String(), nil
}
// waitForListener blocks until address accepts a TCP connection or ctx expires.
func waitForListener(ctx context.Context, address string) error {
ticker := time.NewTicker(100 * time.Millisecond)
defer ticker.Stop()
for {
dialer := net.Dialer{Timeout: time.Second}
conn, err := dialer.DialContext(ctx, "tcp", address)
if err == nil {
_ = conn.Close()
return nil
}
select {
case <-ctx.Done():
return ctx.Err()
case <-ticker.C:
}
}
}
// ReplacesTextOnInput reports that InputText replaces the field's content, so
// the runner skips its pre-erase. The driver clears the field inside InputText,
// which is robust even when a collapsed accessibility bridge would make the
// runner read the field length as zero and wrongly skip erasing.
func (d *Driver) ReplacesTextOnInput() bool { return true }
var (
_ driver.DeviceDriver = (*Driver)(nil)
_ driver.ForegroundChecker = (*Driver)(nil)
_ driver.TextReplacer = (*Driver)(nil)
)