This is an automated email from the ASF dual-hosted git repository.
github-merge-queue[bot] pushed a commit to branch gh-readonly-queue/main/pr-7635-649a27d8bce33d8857ce6999dc0919f5b36c89ff
in repository https://gitbox.apache.org/repos/asf/texera.git
commit 65f1f668acf38d83aa17991359477d3f2f44db3e
Author: Eugene Gu <eu...@outlook.com>
AuthorDate: Sun Aug 23 06:50:15 2026 +0000
fix(test, computing-unit): restore ComputingUnitManagingServiceSpec compilation after the avatar rename (#7635)
### What changes were proposed in this PR?
`ComputingUnitManagingResourceSpec` references a field named
`ownerGoogleAvatar` in two assertions, but
`DashboardWorkflowComputingUnit` declares it as `ownerAvatar`
(`ComputingUnitManagingResource.scala:162`). `main` therefore fails to
compile its test sources:
```
[error] .../ComputingUnitManagingResourceSpec.scala:179:10: value ownerGoogleAvatar is not a member of
org.apache.texera.service.resource.ComputingUnitManagingResource.DashboardWorkflowComputingUnit
[error] info.ownerGoogleAvatar shouldBe "owner-avatar"
[error] .../ComputingUnitManagingResourceSpec.scala:285:13: value ownerGoogleAvatar is not a member of ...
[error] created.ownerGoogleAvatar shouldBe null
[error] two errors found
[error] (ComputingUnitManagingService / Test / compileIncremental) Compilation failed
```
This renames the two references. The diff is two lines; no production
file is touched and no assertion's meaning changes.
Because Scala test compilation is per module, the impact is wider than
two assertions: **no test in `computing-unit-managing-service` has been
able to run.** `ComputingUnitHelpersSpec`,
`AdminComputingUnitResourceSpec`, `KubernetesClientSpec` and
`ComputingUnitAccessResourceSpec` were all taken down with it.
This is merge skew. #7563 (`06845321b`) renamed the field and correctly
updated the two specs that referenced it at the time —
`AdminComputingUnitResourceSpec` and `ComputingUnitHelpersSpec`, both of
which use `ownerAvatar` on `main` today.
`ComputingUnitManagingResourceSpec` had no avatar assertions then, so
there was nothing to update. #7580 (`c7e7362a4`) then added them from a
branch based before the rename, and its CI ran against that older base.
`git merge-base --is-ancestor 06845321b c7e7362a4` returns true, so the
rename was already on `main` when #7580 landed.
### Any related issues, documentation, discussions?
Closes #7634
Introduced by the interaction of #7563 (the `ownerGoogleAvatar` →
`ownerAvatar` rename) and #7580 (which added the avatar assertions from
a pre-rename base).
### How was this PR tested?
`sbt "ComputingUnitManagingService/Test/compile"` fails on `main` with
the two errors above and succeeds on this branch.
`sbt "ComputingUnitManagingService/test"` on this branch: **116 tests
across 8 suites, all passing.** None of those 116 could run on `main`.
`ComputingUnitManagingResourceSpec` itself passes 31/31 — the two
assertions were already correct, so nothing about their meaning had to
change, only the field name they read.
No other module is affected: the change is confined to one test file in
one module and renames a field reference to match an existing
declaration. A repo-wide `sbt Test/compile` across all sixteen
aggregated modules succeeds on this branch.
### Was this PR authored or co-authored using generative AI tooling?
Generated-by: Claude Code (Claude Opus 5)