scriptoriumadmin/litellm
1
0
Fork 0
mirror of https://github.com/BerriAI/litellm.git synced 2026-09-05 08:07:05 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 136

test(e2e/batches): run the list assertion when a batch completes before cancel

Browse source 
The completed-batch early return skipped both the cancel and the list
assertion while the lifecycle's covers markers still credited both cells.
List does not depend on the batch being cancellable, so it now runs either
way; cancel on a completed batch stays a documented vacuous pass
This commit is contained in:
mateo-berri 2026-09-04 18:50:47 -07:00
parent 1748dd81a7
commit ee50f2bc44
2 changed files with 8 additions and 9 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

2
tests/e2e/batches/COVERAGE.md
View file

@ -25,7 +25,7 @@ Bedrock cancel maps to `StopModelInvocationJob` and comes back `cancelling`; the
lifecycle asserts it the same way it does for OpenAI (`_CANCEL_ASSERTED_PROVIDERS`).
Bedrock has no provider-side list, so list is the proxy's DB-backed managed view: the
unified lifecycle lists with the plain `GET /v1/batches` and the batch must appear
there. Both were gated off until LIT-5730, after LIT-4774 landed cancel support.
there. Both were gated off until LIT-5730, after LIT-4774 landed cancel support. A batch that completes inside the 2 s pre-cancel window skips the cancel assertion (a documented vacuous pass for the cancel cell, same as OpenAI); the list assertion runs either way.
Bedrock file upload requires a model on the request (`encoded` / `unified` scenarios only);
`model_param` and `provider_fallback` are omitted because `POST /bedrock/v1/files` has no
model-less passthrough path.

15
tests/e2e/batches/test_batches_e2e.py
View file

@ -293,14 +293,13 @@ def test_batch_lifecycle(
f"batch reached {pre_cancel.status!r} before cancel; "
"provider likely rejected the input"
)
if pre_cancel.status == "completed":
return
cancelled = cancel_batch(client, batch.id, key=key, provider=provider)
assert cancelled.id == batch.id
assert cancelled.object == "batch"
assert cancelled.status in {"cancelling", "cancelled"}, (
f"unexpected post-cancel status {cancelled.status!r}"
)
if pre_cancel.status != "completed":
cancelled = cancel_batch(client, batch.id, key=key, provider=provider)
assert cancelled.id == batch.id
assert cancelled.object == "batch"
assert cancelled.status in {"cancelling", "cancelled"}, (
f"unexpected post-cancel status {cancelled.status!r}"
)
if cap.can_list:
list_result = client.list_batches(key=key, provider=provider)