Conversation
The SDK's spec/public-api.yaml had drifted several commits behind zuuul's
reef/public/static/public-api.yaml. Recent SDK PRs hand-added just the
schemas they needed rather than re-copying the spec, so the file was never
a clean copy of any upstream revision and the gap accumulated silently.
This syncs spec/public-api.yaml to an exact copy of zuuul main and
regenerates. No functional SDK changes beyond what the spec dictates.
Picked up from zuuul:
- #6729 drop expires_at from token create: ApiTokenRequest is now
ApiTokenCreationInputRequest, without the expires_at field
- #6399 durable ONNX pipeline builds: /v1/edge/gll-pipeline and
/v1/edge/model-info, plus GLLPipelineInfo and GLLModelInfo. Also
renames StatusEnum to DetectorStatusEnum and adds StaleFromStatusEnum
- #6400 durable TensorRT engine builds: /v1/edge/gll-engine, plus
GLLEngineInfo, GLLEngineInfoRequest and Status638Enum
- #6768 priming-group id description on detector creation
- doc-only text from #6519, #6590 and #6652 that earlier partial syncs
had missed
The two renames touch hand-written code, so client.py and token_manager.py
are updated to match. Both are pure renames: DetectorStatusEnum carries the
same ON/OFF values as StatusEnum, and the token create call site already
omitted expires_at deliberately, so dropping the field changes nothing.
Me and Group models are also newly committed under generated/. They were
already in the spec as of #468 but the regenerated client was never checked
in alongside it.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What
spec/public-api.yamlhad drifted several commits behind zuuul'sreef/public/static/public-api.yaml. This syncs it to an exact copy of zuuulmainand regenerates viamake generate.Deliberately excludes the in-flight
POST /v1/synthetic-imageswork (zuuulfrancine/mirage-vce-to-janzu-routing) — this is a catch-up sync only.Why it drifted
The SDK's spec was never a clean copy of any upstream revision. Recent SDK PRs (#442, #445, #468) hand-added just the schemas they needed instead of re-copying the file — you can see it in the diff, where
VlmVerificationsits at the end of the file rather than in alphabetical position. So the gap accumulated quietly, including doc-only text from commits that were otherwise "synced."zuuul's
auto-update-public-api.yamldoes fail any PR that changes the spec, warning "be prepared to update the SDK" — but nothing verifies the SDK actually got updated, so #6399, #6400 and #6729 all shipped without it.What this picks up
expires_atfrom token create →ApiTokenRequest→ApiTokenCreationInputRequest/v1/edge/gll-pipeline/,/v1/edge/model-info/,GLLPipelineInfo,GLLModelInfo,StaleFromStatusEnum; renamesStatusEnum→DetectorStatusEnum/v1/edge/gll-engine/,GLLEngineInfo,GLLEngineInfoRequest,Status638EnumHand-written code touched
The two renames hit live imports, so
src/is updated to match. Both are pure renames:client.py—StatusEnum→DetectorStatusEnum, sameON/OFFvaluestoken_manager.py—ApiTokenRequest→ApiTokenCreationInputRequest. The call site already omitsexpires_aton purpose ("so the server applies the identity's token lifetime policy"), so dropping the field is a no-op here.MeandGroupalso appear as new files undergenerated/. They were already in the spec as of #468, but the regenerated client was never committed alongside it.Testing
groundlight.*modules import cleanly; newEdgeApioperations andMeresolveruff check srcpasses;pylint src9.85/10 andmypy srcshow only pre-existing findings in files this PR doesn't touchtest/conftest.pyconstructs a liveGroundlight()inpytest_configure, so collection fails without aGROUNDLIGHT_API_TOKEN, which I don't have in this environment. Needsmake testagainst a real token before merge — the token-rotation tests in particular exercise theApiTokenCreationInputRequestchange.Follow-up, not addressed here
generated/carries a pile of pre-existing orphan files no longer in the generator'sFILESmanifest (~15 models, ~100 tests), now includingstatus_enum.pyandapi_token_request.py. I left them alone to keep this diff to the sync — notedetector_request.py(itself an orphan) still importsStatusEnum, so a cleanup should remove them together.🤖 Generated with Claude Code