|
10 | 10 | `session.name` (`default`) instead of the store address (`cwd:<hash>:default`), so the record |
11 | 11 | read as missing, the `log stream` child leaked, and the next `logs start` on that device failed |
12 | 12 | with "has not reached a confirmed terminal state" (#2647). |
| 13 | +- Fixed (ios): Simulator AX bridge snapshots report `enabled`. The bridge requested no state |
| 14 | + attribute, so a disabled control — a React Native `Pressable` with `disabled`, for example — read |
| 15 | + as a plain button while the XCTest runner answered the same screen with `enabled: false`. The |
| 16 | + bridge now reads the element's accessibility traits (sent as a decimal string, since the word has |
| 17 | + bits past 2^53) and derives `enabled` from `UIAccessibilityTraitNotEnabled`. Source version |
| 18 | + `agent-device-simulator-ax-v1.6.0` rebuilds the cached bridge on first use. |
| 19 | +- Changed (ios): a node the source declares disabled is presented `hittable: false` even when the |
| 20 | + capture has no hittability evidence, so on the Simulator bridge a disabled control stops counting |
| 21 | + as an interactive node and as a Maestro atomic-dispatch candidate. The navigation title affordance |
| 22 | + (a disabled title field presented as an enabled Button for the whole row) no longer carries the |
| 23 | + field's `hittable: false`; on the XCTest runner path that Button now counts as interactive and |
| 24 | + becomes a Maestro atomic-dispatch candidate where it was excluded before. |
13 | 25 | - Added (limrun): `longpress` on Limrun iOS direct sessions. The interactor refused it as |
14 | 26 | unsupported although the SDK exposes the HID primitives; it now holds one touch as a |
15 | 27 | `performActions` batch of `touchDown`, `wait`, `touchUp`, defaulting to the 800 ms the Android |
|
165 | 177 | against the instant that attempt signalled the recorder rather than the time of the retry. A stop |
166 | 178 | that fails and is then abandoned keeps the pulled set on the host beside `--out`. Chunk paths and |
167 | 179 | `--client-output-path` naming are unchanged. |
| 180 | +- Fixed: a process lock that could not be given back no longer waits for the daemon to restart. A |
| 181 | + release that cannot verify ownership — a refused `unlink`, an unreadable record — left its record |
| 182 | + standing and naming the live daemon, and the next acquire read that as a live owner and timed out |
| 183 | + after 30 s on every runner build or launch until the process restarted. The claim inside is spent |
| 184 | + the moment the release is asked for, so a reclaim here now reads it as dead and takes the path |
| 185 | + back, and the failed release is recorded in the request log as `process_lock_release_unverified`. |
| 186 | + A record is only read that way when this loading of the lock code issued it: a second bundled copy |
| 187 | + of the module in the same process shares the pid and cannot have a live claim cleared under it. |
| 188 | +- Changed (lock errors): code that takes a process lock now answers one question in one helper, which |
| 189 | + of its two failures the caller hears. The work inside the lock outranks a lock that could not be |
| 190 | + handed back, so a build or a publish that failed keeps its own error instead of being replaced by |
| 191 | + `Timed out waiting for …`, and work that succeeded still reports the lock it could not give back. |
| 192 | + The Apple runner's artifact, cache, lease and disposal paths, the managed-allocation store, the |
| 193 | + device-claim store, the iOS snapshot bridge cache, the Swift recording cache and the agent-browser setup |
| 194 | + moved onto it, replacing hand-written try/catch pairs that each chose differently. |
| 195 | +- Fixed: the redirect of `~/Library/Developer/XCTestDevices` gives itself back in one order — restore |
| 196 | + the host's own device set, then release the lock — and a restore that was refused is what the caller |
| 197 | + is told, whichever give-back door a teardown used. The lock's own complaint used to replace it in a |
| 198 | + `finally` and the best-effort door then dropped it, leaving the symlink pointed at the agent-device |
| 199 | + simulator set with nothing said about why. A simulator whose set already is `XCTestDevices` is no |
| 200 | + longer failed by a lock it could not verify, either. A leftover from an interrupted build — the host's |
| 201 | + set renamed aside and `XCTestDevices` symlinked into a simulator's own set — is now put back before the |
| 202 | + redirect decides whether it is needed, so the first build after an interruption still gets its own |
| 203 | + device set instead of the host's. A redirect that could not be installed reports the restore that failed |
| 204 | + with it, and names a backup path only when that backup is really on disk. |
168 | 205 | - Changed (sessions): the implicit session is now keyed by workspace **and platform**, so one checkout |
169 | 206 |
|
170 | 207 | can drive iOS and Android without inventing a `--session` name for every command (#2580). An |
|
0 commit comments