bugfix(ci): bump live-test tag v2026.04.0 → v2026.04.2 (#249 follow-up)#250
Merged
bugfix(ci): bump live-test tag v2026.04.0 → v2026.04.2 (#249 follow-up)#250
Conversation
…llow-up) The ci.yml `client-tests` job was passing `--tag=v2026.04.0` to the dagger `test-clients` fn, and the dagger client-test fns themselves defaulted to `v2026.04.0` per signature. Both pre-date the v2026.04.2 release-cut. v2026.04.0 doesn't exist on `gerchowl/mat-vis` (refs are just `[v2026.04.2, main]` after the v2026.04.1 sunset cleanup), so the client tests' live blocks fail with `Failed to fetch .../resolve/v2026.04.0/release-manifest.json`. Surfaced on PR #249's CI run (run 25172585927) where the new manifest-driven coverage suite fired the live blocks for the first time and exposed the stale tag. Bumps both the workflow invocation and all five dagger fn defaults (test_client_python/_js/_shell/_rust + test_clients) to v2026.04.2. Future bumps will need to update both surfaces in lock-step until we add a CalVer-injection mechanism (worth filing as a follow-up).
github-actions
Bot
added
area:ci
Merged
2 tasks
gerchowl
added a commit
that referenced
this pull request
Apr 30, 2026
…04.2 (#250 follow-up) (#252) * bugfix(ci): bump live-test tag from v2026.04.0 to v2026.04.2 (#249 follow-up) The ci.yml `client-tests` job was passing `--tag=v2026.04.0` to the dagger `test-clients` fn, and the dagger client-test fns themselves defaulted to `v2026.04.0` per signature. Both pre-date the v2026.04.2 release-cut. v2026.04.0 doesn't exist on `gerchowl/mat-vis` (refs are just `[v2026.04.2, main]` after the v2026.04.1 sunset cleanup), so the client tests' live blocks fail with `Failed to fetch .../resolve/v2026.04.0/release-manifest.json`. Surfaced on PR #249's CI run (run 25172585927) where the new manifest-driven coverage suite fired the live blocks for the first time and exposed the stale tag. Bumps both the workflow invocation and all five dagger fn defaults (test_client_python/_js/_shell/_rust + test_clients) to v2026.04.2. Future bumps will need to update both surfaces in lock-step until we add a CalVer-injection mechanism (worth filing as a follow-up). * bugfix(test): bump physicallybased proof + LIVE_TAG default to v2026.04.2 (#250 follow-up) #250's CI fix bumped the workflow + dagger fn defaults from v2026.04.0 to v2026.04.2 but missed two stale tag references in clients/python/test_client.py: - LIVE_TAG default = "v2026.04.1" (a sunset CalVer; refs on prod are [v2026.04.2, main] now). - test_proof_phase_2_fetch_physicallybased_index_from_hf hardcoded tag="v2026.04.1" instead of using LIVE_TAG. Both surfaced when #248's manifest-driven coverage suite finally fired the @LiVe blocks in CI and they 404'd against the deleted v2026.04.1 ref. Two changes: 1. LIVE_TAG default → "v2026.04.2". 2. The proof test now reads LIVE_TAG so it rolls forward with the release tag (env-overridable for staging dispatches). A separate concern surfaced by the same CI failure — bake_scalar_source in src/mat_vis_baker/hf_bake.py doesn't update release-manifest.json (legacy comment says "clients derive from tree", which broke as a contract once #239 made the client read the manifest file directly). Filed as a follow-up; not in this PR's scope.
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
1 participant
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.
Summary
#249's manifest-driven coverage suite fires the `@live` test blocks via `MAT_VIS_LIVE_TESTS=1`. The `ci.yml` `client-tests` job was passing `--tag=v2026.04.0`, and the dagger `test_client_*` fns defaulted to `v2026.04.0` per signature. `v2026.04.0` doesn't exist on `gerchowl/mat-vis` — refs are `[v2026.04.2, main]` after the v2026.04.1 sunset cleanup. Live blocks fail with:
```
error: Failed to fetch https://huggingface.co/datasets/gerchowl/mat-vis/resolve/v2026.04.0/release-manifest.json
```
Surfaced on run 25172585927 — PR #249's own CI exposed the stale tag the moment its new live-coverage suite ran.
Fix
Bump both surfaces to `v2026.04.2`:
Future CalVer bumps will need to touch both surfaces until we add a CalVer-injection mechanism (worth filing as a follow-up — bake.yml already accepts release-tag as input; the test-clients dispatch path doesn't yet).
Test plan