scriptoriumadmin/litellm
1
0
Fork 0
mirror of https://github.com/BerriAI/litellm.git synced 2026-10-26 06:03:19 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 35

fix(azure_ai): use builtin tuple[...] to stay within UP006 budget

Browse source 
The two new module-level constants used typing.Tuple[str, ...], which ruff
UP006 flags in favor of the PEP 585 builtin. The repo's UP006 ceiling in
ruff-strict-budget.json was already at its limit (12792), so the 2 added
violations pushed the total to 12793 and failed the strict-rule budget gate
in CI (lint job). Switch to tuple[str, ...] (requires-python >=3.10), which
is runtime-equivalent and removes both new violations. The pre-existing
typing.Tuple usage later in the file is left unchanged as it is already
counted in the base budget.
This commit is contained in:
vrlo 2026-07-20 16:01:28 +02:00
parent 6979225b61
commit 727206405d
1 changed files with 2 additions and 2 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

4
litellm/llms/azure_ai/chat/transformation.py
View file

@ -35,7 +35,7 @@ class AzureFoundryErrorStrings(str, enum.Enum):
# ``reasoning_content`` on input, but the direct ``fireworks_ai`` provider
# strips ``thinking_blocks`` and Azure AI Foundry fronts multiple backends, so
# drop both as a defensive default for a generic multi-backend provider.
_TOP_LEVEL_OUTPUT_ONLY_FIELDS: Tuple[str, ...] = (
_TOP_LEVEL_OUTPUT_ONLY_FIELDS: tuple[str, ...] = (
"thinking_blocks",
"reasoning_content",
)
@ -49,7 +49,7 @@ _TOP_LEVEL_OUTPUT_ONLY_FIELDS: Tuple[str, ...] = (
# strict upstreams still reject it with
# ``Extra inputs are not permitted, field:
# 'messages[n].tool_calls[m].function.provider_specific_fields'``.
_RECURSIVELY_STRIPPED_FIELDS: Tuple[str, ...] = (
_RECURSIVELY_STRIPPED_FIELDS: tuple[str, ...] = (
"cache_control",
"provider_specific_fields",
)