[Frontend] Move offline_utils.py to entrypoints/common. (#58052)

Signed-off-by: wang.yuqi <yuqi.wang@daocloud.io>

.github/CODEOWNERS

@@ -45,7 +45,7 @@
 /vllm/entrypoints/serve @njhill
 /vllm/entrypoints/*.py @njhill
 /vllm/entrypoints/chat_utils.py @DarkLight1337
-/vllm/entrypoints/offline_utils.py @DarkLight1337
+/vllm/entrypoints/common @DarkLight1337
 /vllm/entrypoints/llm.py @DarkLight1337
 
 # Rust Frontend

tests/entrypoints/multimodal/llm/test_mm_processor_kwargs.py

@@ -212,7 +212,7 @@ def test_preprocess_cmpl_applies_mm_processor_kwargs_to_renderer(
     llm.renderer = renderer
 
     monkeypatch.setattr(
-        "vllm.entrypoints.offline_utils.parse_model_prompt",
+        "vllm.entrypoints.common.offline.parse_model_prompt",
         lambda _model_config, parsed_prompt: parsed_prompt,
     )
 
@@ -246,7 +246,7 @@ def test_preprocess_cmpl_keeps_prompt_mm_processor_kwargs_when_no_override(
     llm.renderer = renderer
 
     monkeypatch.setattr(
-        "vllm.entrypoints.offline_utils.parse_model_prompt",
+        "vllm.entrypoints.common.offline.parse_model_prompt",
         lambda _model_config, parsed_prompt: parsed_prompt,
     )
 

tests/entrypoints/unit_tests/test_offline_utils.py

@@ -4,7 +4,7 @@
 import pytest
 
 from vllm import SamplingParams
-from vllm.entrypoints.offline_utils import OfflineInferenceMixin
+from vllm.entrypoints.common.offline import OfflineInferenceMixin
 from vllm.exceptions import VLLMValidationError
 from vllm.lora.request import LoRARequest
 

vllm/entrypoints/generate/beam_search/offline.py

@@ -8,7 +8,7 @@ import torch
 from tqdm import tqdm
 
 from vllm import RequestOutput, TextPrompt, TokensPrompt
-from vllm.entrypoints.offline_utils import OfflineInferenceMixin
+from vllm.entrypoints.common.offline import OfflineInferenceMixin
 from vllm.logger import init_logger
 from vllm.lora.request import LoRARequest
 from vllm.pooling_params import PoolingParams

vllm/entrypoints/llm.py

@@ -33,6 +33,7 @@ from vllm.entrypoints.chat_utils import (
     ChatTemplateContentFormatOption,
     load_chat_template,
 )
+from vllm.entrypoints.common.offline import _O, _R, OfflineInferenceMixin
 from vllm.entrypoints.generate.beam_search.offline import BeamSearchOfflineMixin
 from vllm.entrypoints.pooling.offline import PoolingOfflineMixin
 from vllm.entrypoints.rl.offline import RLOfflineMixin
@@ -52,7 +53,6 @@ from vllm.v1.engine.llm_engine import LLMEngine
 from vllm.v1.sample.logits_processor import LogitsProcessor
 
 from ..renderers import ChatParams
-from .offline_utils import _O, _R, OfflineInferenceMixin
 
 if TYPE_CHECKING:
     from vllm.v1.metrics.reader import Metric

vllm/entrypoints/pooling/offline.py

@@ -7,7 +7,7 @@ from typing import Any
 from tqdm.auto import tqdm
 
 from vllm.entrypoints.chat_utils import ChatTemplateConfig
-from vllm.entrypoints.offline_utils import OfflineInferenceMixin
+from vllm.entrypoints.common.offline import OfflineInferenceMixin
 from vllm.inputs import DataPrompt, PromptType
 from vllm.logger import init_logger
 from vllm.lora.request import LoRARequest

vllm/entrypoints/rl/offline.py

@@ -5,7 +5,7 @@ from vllm.distributed.weight_transfer.base import (
     WeightTransferInitRequest,
     WeightTransferUpdateRequest,
 )
-from vllm.entrypoints.offline_utils import OfflineInferenceMixin
+from vllm.entrypoints.common.offline import OfflineInferenceMixin
 
 
 class RLOfflineMixin(OfflineInferenceMixin):