[Bugfix] Migrate to REGEX Library to prevent catastrophic backtracking (#18454)

Signed-off-by: Crucifixion-Fxl <xmufxl@gmail.com>
Co-authored-by: Crucifixion-Fxl <xmufxl@gmail.com>

vllm/model_executor/layers/quantization/utils/gptq_utils.py

 # SPDX-License-Identifier: Apache-2.0
-import re
 from copy import deepcopy
 from typing import Optional, Union
 
+import regex as re
 import torch
 
 from vllm.config import QuantizationConfig

vllm/model_executor/model_loader/tensorizer.py

@@ -7,7 +7,6 @@ import dataclasses
 import io
 import json
 import os
-import re
 import threading
 import time
 from collections.abc import Generator
@@ -15,6 +14,7 @@ from dataclasses import dataclass
 from functools import partial
 from typing import Any, BinaryIO, Optional, Union
 
+import regex as re
 import torch
 from torch import nn
 from torch.utils._python_dispatch import TorchDispatchMode

vllm/model_executor/models/mimo_mtp.py

@@ -250,7 +250,7 @@ class MiMoMTP(nn.Module):
         return loaded_params
 
     def map_model_name_to_mtp_param_name(self, name: str) -> str:
-        import re
+        import regex as re
         name_without_prefix = [
             "token_layernorm", "hidden_layernorm", "input_proj",
             "final_layernorm"

vllm/model_executor/models/minimax_text_01.py

@@ -2,10 +2,10 @@
 """Inference-only MiniMaxText01 model."""
 import copy
 import math
-import re
 from collections.abc import Iterable
 from typing import Optional, Union
 
+import regex as re
 import torch
 import torch.distributed
 import torch.nn.functional as F

vllm/model_executor/models/phi3v.py

@@ -14,10 +14,10 @@
 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 # See the License for the specific language governing permissions and
 # limitations under the License.
-import re
 from collections.abc import Iterable, Mapping, Sequence
 from typing import Any, Literal, Optional, TypedDict, Union
 
+import regex as re
 import torch
 import torch.nn as nn
 from transformers import (BatchFeature, CLIPVisionConfig, PretrainedConfig,

vllm/model_executor/models/qwen_vl.py

@@ -7,12 +7,12 @@
 
 import copy
 import math
-import re
 import unicodedata
 from collections.abc import Collection, Mapping, Sequence, Set
 from functools import lru_cache, partial
 from typing import Callable, Literal, Optional, TypedDict, Union
 
+import regex as re
 import torch
 from torch import nn
 from torchvision import transforms

vllm/model_executor/models/transformers.py

@@ -14,11 +14,11 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 """Wrapper around `transformers` models"""
-import re
 from collections.abc import Iterable
 from contextlib import nullcontext
 from typing import Literal, Optional, Union
 
+import regex as re
 import torch
 from torch import nn
 from transformers import AutoModel, PretrainedConfig, PreTrainedModel

vllm/multimodal/processing.py

 # SPDX-License-Identifier: Apache-2.0
 import json
-import re
 import sys
 from abc import ABC, abstractmethod
 from collections import defaultdict
@@ -12,6 +11,7 @@ from functools import lru_cache
 from typing import (TYPE_CHECKING, Generic, NamedTuple, Optional, Protocol,
                     TypeVar, Union, cast)
 
+import regex as re
 import torch
 from typing_extensions import assert_never
 

vllm/reasoning/granite_reasoning_parser.py

 # SPDX-License-Identifier: Apache-2.0
 
-import re
 from collections.abc import Sequence
 from typing import Optional, Union
 
+import regex as re
 from transformers import PreTrainedTokenizerBase
 
 from vllm.entrypoints.openai.protocol import (ChatCompletionRequest,

vllm/transformers_utils/tokenizers/mistral.py

 # SPDX-License-Identifier: Apache-2.0
 
 import os
-import re
 from dataclasses import dataclass
 from pathlib import Path
 from typing import TYPE_CHECKING, Any, Optional, Union, cast
 
 import huggingface_hub
+import regex as re
 from huggingface_hub import HfApi, hf_hub_download
 
 from vllm.logger import init_logger

vllm/utils.py

@@ -19,7 +19,6 @@ import json
 import multiprocessing
 import os
 import pickle
-import re
 import signal
 import socket
 import subprocess
@@ -54,6 +53,7 @@ import cloudpickle
 import numpy as np
 import numpy.typing as npt
 import psutil
+import regex as re
 import torch
 import torch.types
 import yaml

vllm/v1/structured_output/utils.py

@@ -2,7 +2,7 @@
 
 from __future__ import annotations
 
-import re
+import regex as re
 
 
 def grammar_is_likely_lark(grammar_str: str) -> bool: