EngineX-Hygon/sglang
5
0
Fork 0
Code Issues Pull Requests Actions 7 Projects Releases Wiki Activity

Tiny move files to utils folder (#11166)

Browse Source
This commit is contained in:
fzyzcjy
2025-10-03 22:40:06 +08:00
committed by GitHub
parent 04b86b3c5c
commit fdc4e1e570
66 changed files with 91 additions and 79 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
test/srt/openai_server/basic/test_openai_server.py
View File

@@ -13,8 +13,8 @@ import numpy as np
import openai
import requests
from sglang.srt.hf_transformers_utils import get_tokenizer
from sglang.srt.utils import kill_process_tree
from sglang.srt.utils.hf_transformers_utils import get_tokenizer
from sglang.test.runners import TEST_RERANK_QUERY_DOCS
from sglang.test.test_utils import (
DEFAULT_SMALL_CROSS_ENCODER_MODEL_NAME_FOR_TEST,

2
test/srt/openai_server/features/test_enable_thinking.py
View File

@@ -16,8 +16,8 @@ import unittest
import openai
import requests
from sglang.srt.hf_transformers_utils import get_tokenizer
from sglang.srt.utils import kill_process_tree
from sglang.srt.utils.hf_transformers_utils import get_tokenizer
from sglang.test.test_utils import (
DEFAULT_ENABLE_THINKING_MODEL_NAME_FOR_TEST,
DEFAULT_TIMEOUT_FOR_SERVER_LAUNCH,

2
test/srt/openai_server/features/test_openai_server_ebnf.py
View File

@@ -2,8 +2,8 @@ import re
import openai
from sglang.srt.hf_transformers_utils import get_tokenizer
from sglang.srt.utils import kill_process_tree
from sglang.srt.utils.hf_transformers_utils import get_tokenizer
from sglang.test.test_utils import (
DEFAULT_SMALL_MODEL_NAME_FOR_TEST,
DEFAULT_TIMEOUT_FOR_SERVER_LAUNCH,

2
test/srt/openai_server/features/test_openai_server_hidden_states.py
View File

@@ -8,8 +8,8 @@ import numpy as np
import openai
import torch
from sglang.srt.hf_transformers_utils import get_tokenizer
from sglang.srt.utils import kill_process_tree
from sglang.srt.utils.hf_transformers_utils import get_tokenizer
from sglang.test.test_utils import (
DEFAULT_EAGLE_DRAFT_MODEL_FOR_TEST,
DEFAULT_EAGLE_TARGET_MODEL_FOR_TEST,

2
test/srt/openai_server/function_call/test_openai_function_calling.py
View File

@@ -4,8 +4,8 @@ import unittest
import openai
from sglang.srt.hf_transformers_utils import get_tokenizer
from sglang.srt.utils import kill_process_tree
from sglang.srt.utils.hf_transformers_utils import get_tokenizer
from sglang.test.test_utils import (
DEFAULT_SMALL_MODEL_NAME_FOR_TEST,
DEFAULT_TIMEOUT_FOR_SERVER_LAUNCH,

2
test/srt/openai_server/function_call/test_tool_choice.py
View File

@@ -12,8 +12,8 @@ import unittest
import openai
from sglang.srt.hf_transformers_utils import get_tokenizer
from sglang.srt.utils import kill_process_tree
from sglang.srt.utils.hf_transformers_utils import get_tokenizer
from sglang.test.test_utils import (
DEFAULT_TIMEOUT_FOR_SERVER_LAUNCH,
DEFAULT_URL_FOR_TEST,

2
test/srt/openai_server/validation/test_large_max_new_tokens.py
View File

@@ -9,8 +9,8 @@ from concurrent.futures import ThreadPoolExecutor
import openai
from sglang.srt.hf_transformers_utils import get_tokenizer
from sglang.srt.utils import kill_process_tree
from sglang.srt.utils.hf_transformers_utils import get_tokenizer
from sglang.test.test_utils import (
DEFAULT_SMALL_MODEL_NAME_FOR_TEST,
DEFAULT_TIMEOUT_FOR_SERVER_LAUNCH,

2
test/srt/openai_server/validation/test_openai_server_ignore_eos.py
View File

@@ -1,7 +1,7 @@
import openai
from sglang.srt.hf_transformers_utils import get_tokenizer
from sglang.srt.utils import kill_process_tree
from sglang.srt.utils.hf_transformers_utils import get_tokenizer
from sglang.test.test_utils import (
DEFAULT_SMALL_MODEL_NAME_FOR_TEST,
DEFAULT_TIMEOUT_FOR_SERVER_LAUNCH,

2
test/srt/rl/test_verl_engine_2_gpu.py
View File

@@ -19,8 +19,8 @@ from torch.distributed.fsdp.api import (
from transformers import AutoModelForCausalLM
from sglang.srt.entrypoints.verl_engine import VerlEngine
from sglang.srt.hf_transformers_utils import get_tokenizer
from sglang.srt.utils import is_port_available
from sglang.srt.utils.hf_transformers_utils import get_tokenizer
from sglang.test.runners import (
HFRunner,
SRTRunner,

2
test/srt/rl/test_verl_engine_4_gpu.py
View File

@@ -19,8 +19,8 @@ from torch.distributed.fsdp.api import (
from transformers import AutoModelForCausalLM
from sglang.srt.entrypoints.verl_engine import VerlEngine
from sglang.srt.hf_transformers_utils import get_tokenizer
from sglang.srt.utils import is_port_available
from sglang.srt.utils.hf_transformers_utils import get_tokenizer
from sglang.test.runners import (
HFRunner,
SRTRunner,

2
test/srt/test_bench_serving.py
View File

@@ -4,7 +4,7 @@ import unittest
import requests
from sglang.srt.hf_transformers_utils import get_tokenizer
from sglang.srt.utils.hf_transformers_utils import get_tokenizer
from sglang.test.test_utils import (
DEFAULT_EAGLE_DRAFT_MODEL_FOR_TEST,
DEFAULT_EAGLE_TARGET_MODEL_FOR_TEST,

2
test/srt/test_eagle_infer_a.py
View File

@@ -4,8 +4,8 @@ import requests
import torch
import sglang as sgl
from sglang.srt.hf_transformers_utils import get_tokenizer
from sglang.srt.utils import kill_process_tree
from sglang.srt.utils.hf_transformers_utils import get_tokenizer
from sglang.test.test_utils import (
DEFAULT_EAGLE_DRAFT_MODEL_FOR_TEST,
DEFAULT_EAGLE_TARGET_MODEL_FOR_TEST,

2
test/srt/test_fim_completion.py
View File

@@ -2,8 +2,8 @@ import unittest
import openai
from sglang.srt.hf_transformers_utils import get_tokenizer
from sglang.srt.utils import kill_process_tree
from sglang.srt.utils.hf_transformers_utils import get_tokenizer
from sglang.test.test_utils import (
DEFAULT_TIMEOUT_FOR_SERVER_LAUNCH,
DEFAULT_URL_FOR_TEST,

2
test/srt/test_forward_split_prefill.py
View File

@@ -13,13 +13,13 @@ import numpy as np
import torch
from sglang.srt.configs.model_config import ModelConfig
from sglang.srt.hf_transformers_utils import get_tokenizer
from sglang.srt.managers.schedule_batch import Req, ScheduleBatch
from sglang.srt.model_executor.forward_batch_info import ForwardBatch
from sglang.srt.model_executor.model_runner import ModelRunner
from sglang.srt.sampling.sampling_params import SamplingParams
from sglang.srt.server_args import PortArgs, ServerArgs
from sglang.srt.speculative.spec_info import SpeculativeAlgorithm
from sglang.srt.utils.hf_transformers_utils import get_tokenizer
from sglang.test.test_utils import DEFAULT_SMALL_MODEL_NAME_FOR_TEST, CustomTestCase

2
test/srt/test_function_call_parser.py
View File

@@ -15,7 +15,7 @@ from sglang.srt.function_call.mistral_detector import MistralDetector
from sglang.srt.function_call.pythonic_detector import PythonicDetector
from sglang.srt.function_call.qwen3_coder_detector import Qwen3CoderDetector
from sglang.srt.function_call.qwen25_detector import Qwen25Detector
from sglang.srt.hf_transformers_utils import get_tokenizer
from sglang.srt.utils.hf_transformers_utils import get_tokenizer
from sglang.test.test_utils import DEFAULT_SMALL_MODEL_NAME_FOR_TEST

2
test/srt/test_patch_torch.py
View File

@@ -6,7 +6,7 @@ from typing import Dict, List
import torch
import torch.multiprocessing as mp
from sglang.srt.patch_torch import monkey_patch_torch_reductions
from sglang.srt.utils.patch_torch import monkey_patch_torch_reductions
class TestReleaseMemoryOccupation(unittest.TestCase):

2
test/srt/test_sagemaker_server.py
View File

@@ -7,8 +7,8 @@ import unittest
import requests
from sglang.srt.hf_transformers_utils import get_tokenizer
from sglang.srt.utils import kill_process_tree
from sglang.srt.utils.hf_transformers_utils import get_tokenizer
from sglang.test.test_utils import (
DEFAULT_SMALL_MODEL_NAME_FOR_TEST,
DEFAULT_TIMEOUT_FOR_SERVER_LAUNCH,

2
test/srt/test_session_control.py
View File

@@ -13,8 +13,8 @@ import unittest
import aiohttp
import requests
from sglang.srt.hf_transformers_utils import get_tokenizer
from sglang.srt.utils import kill_process_tree
from sglang.srt.utils.hf_transformers_utils import get_tokenizer
from sglang.test.test_utils import (
DEFAULT_SMALL_MODEL_NAME_FOR_TEST,
DEFAULT_TIMEOUT_FOR_SERVER_LAUNCH,

2
test/srt/test_srt_engine.py
View File

@@ -12,8 +12,8 @@ import torch
import sglang as sgl
from sglang.bench_offline_throughput import BenchArgs, throughput_test
from sglang.srt.hf_transformers_utils import get_tokenizer
from sglang.srt.server_args import ServerArgs
from sglang.srt.utils.hf_transformers_utils import get_tokenizer
from sglang.test.few_shot_gsm8k_engine import run_eval
from sglang.test.test_utils import (
DEFAULT_SMALL_EMBEDDING_MODEL_NAME_FOR_TEST,

4
test/srt/test_tokenizer_batch_encode.py
View File

@@ -34,7 +34,9 @@ class TestTokenizerBatchEncode(unittest.TestCase):
with patch("zmq.asyncio.Context"), patch(
"sglang.srt.utils.get_zmq_socket"
), patch("sglang.srt.hf_transformers_utils.get_tokenizer") as mock_tokenizer:
), patch(
"sglang.srt.utils.hf_transformers_utils.get_tokenizer"
) as mock_tokenizer:
mock_tokenizer.return_value = Mock(vocab_size=32000)
self.tokenizer_manager = TokenizerManager(self.server_args, self.port_args)

16
test/srt/test_tokenizer_manager.py
View File

@@ -31,7 +31,9 @@ class TestInputFormatDetection(unittest.TestCase):
with patch("zmq.asyncio.Context"), patch(
"sglang.srt.utils.get_zmq_socket"
), patch("sglang.srt.hf_transformers_utils.get_tokenizer") as mock_tokenizer:
), patch(
"sglang.srt.utils.hf_transformers_utils.get_tokenizer"
) as mock_tokenizer:
mock_tokenizer.return_value = Mock(vocab_size=32000)
self.tokenizer_manager = TokenizerManager(self.server_args, self.port_args)
@@ -125,7 +127,9 @@ class TestTokenizerInputPreparation(unittest.TestCase):
with patch("zmq.asyncio.Context"), patch(
"sglang.srt.utils.get_zmq_socket"
), patch("sglang.srt.hf_transformers_utils.get_tokenizer") as mock_tokenizer:
), patch(
"sglang.srt.utils.hf_transformers_utils.get_tokenizer"
) as mock_tokenizer:
mock_tokenizer.return_value = Mock(vocab_size=32000)
self.tokenizer_manager = TokenizerManager(self.server_args, self.port_args)
@@ -177,7 +181,9 @@ class TestTokenizerResultExtraction(unittest.TestCase):
with patch("zmq.asyncio.Context"), patch(
"sglang.srt.utils.get_zmq_socket"
), patch("sglang.srt.hf_transformers_utils.get_tokenizer") as mock_tokenizer:
), patch(
"sglang.srt.utils.hf_transformers_utils.get_tokenizer"
) as mock_tokenizer:
mock_tokenizer.return_value = Mock(vocab_size=32000)
self.tokenizer_manager = TokenizerManager(self.server_args, self.port_args)
@@ -279,7 +285,9 @@ class TestTokenizerManagerIntegration(unittest.TestCase):
with patch("zmq.asyncio.Context"), patch(
"sglang.srt.utils.get_zmq_socket"
), patch("sglang.srt.hf_transformers_utils.get_tokenizer") as mock_tokenizer:
), patch(
"sglang.srt.utils.hf_transformers_utils.get_tokenizer"
) as mock_tokenizer:
mock_tokenizer.return_value = Mock(vocab_size=32000)
self.tokenizer_manager = TokenizerManager(self.server_args, self.port_args)