[CI] Ignore ruff D209, rejoin the docstrings it split, and silence incompatible-rule warnings (#57212)

Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
This commit is contained in:
Harry Mellor
2026-09-16 10:57:24 -07:00
committed by GitHub
co-authored by Claude Opus 5
parent 1085b64425
commit 651a88c09a
693 changed files with 2088 additions and 4171 deletions
@@ -111,8 +111,7 @@ def test_basic_embed(server: RemoteOpenAIServer, model_name: str):
def test_unsupported_input_type_rejected(server: RemoteOpenAIServer, model_name: str):
"""An input_type not defined in the model's prompt config should be
rejected with a 400 error.
"""
rejected with a 400 error."""
body = {
"model": model_name,
"input_type": "nonexistent_type",
@@ -139,8 +138,7 @@ def test_omitted_input_type_accepted(server: RemoteOpenAIServer, model_name: str
def test_v1_v2_parity(server: RemoteOpenAIServer, model_name: str):
"""v1 (OpenAI) and v2 (Cohere) endpoints should produce the same
float embeddings for a generic model.
"""
float embeddings for a generic model."""
texts = ["hello world"]
v2 = _cohere_embed(server, model_name, texts=texts, embedding_types=["float"])
v1 = _openai_embed(server, model_name, texts)
@@ -76,8 +76,7 @@ def test_single_text_parity(server: RemoteOpenAIServer):
def test_batch_parity(server: RemoteOpenAIServer):
"""A batch of texts should produce equivalent embeddings via both APIs,
in the same order.
"""
in the same order."""
texts = [
"machine learning",
"deep learning",
@@ -241,8 +241,7 @@ class TestResolveTruncation:
def test_truncate_none_with_max_tokens(self):
"""truncate=NONE should NOT set truncate_prompt_tokens; the
max_tokens limit is enforced separately via _check_max_tokens.
"""
max_tokens limit is enforced separately via _check_max_tokens."""
req = self._make_request(truncate="NONE", max_tokens=10)
tokens, side = EmbedIOProcessor._resolve_cohere_truncation(req)
assert tokens is None
@@ -240,8 +240,7 @@ async def test_long_text_embedding_1500_chars(
client_with_chunked_processing: openai.AsyncOpenAI, model_name: str
):
"""Test embedding processing for ~1500 character long text
(~1028 tokens, exceeding 512 token limit).
"""
(~1028 tokens, exceeding 512 token limit)."""
# Verify text length
# Verify text has sufficient word count (approximately 1500 words)
word_count = len(LONG_TEXT_1500_WORDS.split())
@@ -289,8 +288,7 @@ async def test_long_text_embedding_2500_chars(
client_with_chunked_processing: openai.AsyncOpenAI, model_name: str
):
"""Test embedding processing for ~2500 character long text
(~2048 tokens, requiring multiple chunks).
"""
(~2048 tokens, requiring multiple chunks)."""
# Verify text length
# Verify text has sufficient word count (approximately 2500 words)
word_count = len(LONG_TEXT_2500_WORDS.split())
@@ -382,8 +380,7 @@ async def test_chunked_vs_normal_consistency(
client_with_chunked_processing: openai.AsyncOpenAI, model_name: str
):
"""Test consistency between chunked and
normal processing (using short text).
"""
normal processing (using short text)."""
# Use a short text within the 512 token limit
short_text = (
"Artificial intelligence technology is changing our world, "
@@ -1,8 +1,7 @@
# SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
"""Unit tests for Cohere embed protocol: build_typed_embeddings and its
underlying packing helpers, plus Cohere-specific serving helpers.
"""
underlying packing helpers, plus Cohere-specific serving helpers."""
import struct