Repository navigation
Conversation
Backported from upstream `84be3c1f`. This branch forked before that fix, and the dev server bundled with the current Core writes the header with Go's duration syntax, which the old parser rejects, so `nexus_metrics` hangs.
Backported from upstream `c87e2060`. The dev server needs `system.enableCancelActivityWorkerCommand`, and the eager case needs a server newer than the one the pinned CLI bundles.
Backported from upstream `42c7bfe9`. This branch forked before that fix, and the unscoped query also matches the shared-namespace worker's series, whose order is not the test's to choose.
The matrix asks for 40 minutes on `macos-intel`, but the job never read that value, so the runner stayed on the 25-minute default it exceeds.
The nexgen the Python SDK drives rejects @nexus.type on a native declaration, so generation failed for every consumer. Upstream declares both policies as placeholders, which keeps the annotations valid.
The `string` declaration made the generated Python Nexus model require `SignalWithStartWorkflowResponse.first_execution_run_id`, which the dev server bundled with this Core never sends. Upstream omits the field instead.
The reporter panics on any machine name it has no visualizer for, so every state machine has to be listed there.
`clippy::type_complexity` rejects the inline nested type, and the lint is denied through `-D warnings` in `cargo test-lint`.
This was referenced Oct 3, 2026
Owner
Author
|
Replaced by #29 in the v4 series: one lineage through the channel and the interface, with Max's Option 7 prototype replayed onto it. The branch stays as a pin. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This PR fixes the chores Max's external-stream Core base needs before anything builds or tests cleanly on it today.
What changed?
request-timeoutheader outside the Nexus duration grammar. Backported from upstream by James Watkins-Harvey.system.enableCancelActivityWorkerCommand, and the Prometheus query in the metrics test is scoped to the test's task queue.timeoutMinutes, somacos-intelgets the 40 minutes it asks for.model.witandworkflow-service.witmatch upstream, so nexgen accepts them and the generated Python Nexus model stops requiringfirst_execution_run_id.ExternalStreamMachineis registered with the transition coverage reporter.Part of AI-198 (epic AI-37).
Why?
Each one keeps the suite honest on Max's older base. Without the Nexus fix,
nexus_metricshangs. Without the coverage registration, the reporter panics on a machine it doesn't know.cargo test-lintfails on the base itself until the type alias lands. None of this touches the channel or the stream machinery, so it sits at the bottom of the chain where every later PR can rely on a green suite.How did you test it?
Link to a test plan if any -
cargo fmt --check,cargo lintandcargo test-lintare clean, and the Core lib suite passes. Every commit builds. The integ fixes weren't run against a dev server here.