[CI] Try introducing isort. (vllm-project#3495)

Author: rkooo567

.github/workflows/ruff.yml

@@ -25,10 +25,13 @@ jobs:
     - name: Install dependencies
       run: |
         python -m pip install --upgrade pip
-        pip install ruff==0.1.5 codespell==2.2.6 tomli==2.0.1
+        pip install ruff==0.1.5 codespell==2.2.6 tomli==2.0.1 isort==5.13.2
     - name: Analysing the code with ruff
       run: |
         ruff .
     - name: Spelling check with codespell
       run: |
-        codespell --toml pyproject.toml
+        codespell --toml pyproject.toml
+    - name: Run isort
+      run: |
+        isort . --check-only

benchmarks/benchmark_prefix_caching.py

@@ -1,8 +1,7 @@
 import argparse
 import time
 
-from vllm import LLM
-from vllm import SamplingParams
+from vllm import LLM, SamplingParams
 
 PROMPT = "You are a helpful assistant in recognizes the content of tables in markdown format. Here is a table as fellows. You need to answer my question about the table.\n# Table\n|Opening|Opening|Sl. No.|Film|Cast|Director|Music Director|Notes|\n|----|----|----|----|----|----|----|----|\n|J A N|9|1|Agni Pushpam|Jayabharathi, Kamalahasan|Jeassy|M. K. Arjunan||\n|J A N|16|2|Priyamvada|Mohan Sharma, Lakshmi, KPAC Lalitha|K. S. Sethumadhavan|V. Dakshinamoorthy||\n|J A N|23|3|Yakshagaanam|Madhu, Sheela|Sheela|M. S. Viswanathan||\n|J A N|30|4|Paalkkadal|Sheela, Sharada|T. K. Prasad|A. T. Ummer||\n|F E B|5|5|Amma|Madhu, Srividya|M. Krishnan Nair|M. K. Arjunan||\n|F E B|13|6|Appooppan|Thikkurissi Sukumaran Nair, Kamal Haasan|P. Bhaskaran|M. S. Baburaj||\n|F E B|20|7|Srishti|Chowalloor Krishnankutty, Ravi Alummoodu|K. T. Muhammad|M. S. Baburaj||\n|F E B|20|8|Vanadevatha|Prem Nazir, Madhubala|Yusufali Kechery|G. Devarajan||\n|F E B|27|9|Samasya|Madhu, Kamalahaasan|K. Thankappan|Shyam||\n|F E B|27|10|Yudhabhoomi|K. P. Ummer, Vidhubala|Crossbelt Mani|R. K. Shekhar||\n|M A R|5|11|Seemantha Puthran|Prem Nazir, Jayabharathi|A. B. Raj|M. K. Arjunan||\n|M A R|12|12|Swapnadanam|Rani Chandra, Dr. Mohandas|K. G. George|Bhaskar Chandavarkar||\n|M A R|19|13|Thulavarsham|Prem Nazir, sreedevi, Sudheer|N. Sankaran Nair|V. Dakshinamoorthy||\n|M A R|20|14|Aruthu|Kaviyoor Ponnamma, Kamalahasan|Ravi|G. Devarajan||\n|M A R|26|15|Swimming Pool|Kamal Haasan, M. G. Soman|J. Sasikumar|M. K. Arjunan||\n\n# Question\nWhat' s the content in the (1,1) cells\n"  # noqa: E501
 

benchmarks/benchmark_serving.py

@@ -25,15 +25,12 @@
 from typing import AsyncGenerator, List, Tuple
 
 import numpy as np
+from backend_request_func import (ASYNC_REQUEST_FUNCS, RequestFuncInput,
+                                  RequestFuncOutput)
 from tqdm.asyncio import tqdm
 from transformers import PreTrainedTokenizerBase
-from vllm.transformers_utils.tokenizer import get_tokenizer
 
-from backend_request_func import (
-    ASYNC_REQUEST_FUNCS,
-    RequestFuncInput,
-    RequestFuncOutput,
-)
+from vllm.transformers_utils.tokenizer import get_tokenizer
 
 
 @dataclass

benchmarks/benchmark_throughput.py

@@ -6,9 +6,9 @@
 from typing import List, Optional, Tuple
 
 import torch
+from tqdm import tqdm
 from transformers import (AutoModelForCausalLM, AutoTokenizer,
                           PreTrainedTokenizerBase)
-from tqdm import tqdm
 
 
 def sample_requests(

benchmarks/kernels/benchmark_mixtral_moe.py

@@ -2,11 +2,13 @@
 import os
 import sys
 
-from vllm.model_executor.layers.fused_moe import fused_moe, get_config_file_name
 import torch
 import torch.nn.functional as F
 import triton
 
+from vllm.model_executor.layers.fused_moe import (fused_moe,
+                                                  get_config_file_name)
+
 os.environ['CUDA_VISIBLE_DEVICES'] = '0'
 
 

benchmarks/kernels/benchmark_paged_attention.py

@@ -1,12 +1,12 @@
-from typing import Optional
 import argparse
 import random
 import time
+from typing import Optional
 
 import torch
 
-from vllm.utils import STR_DTYPE_TO_TORCH_DTYPE, create_kv_caches_with_random
 from vllm._C import ops
+from vllm.utils import STR_DTYPE_TO_TORCH_DTYPE, create_kv_caches_with_random
 
 NUM_BLOCKS = 1024
 PARTITION_SIZE = 512

benchmarks/kernels/benchmark_rope.py

@@ -1,9 +1,10 @@
+import argparse
+from itertools import accumulate
 from typing import Optional
 
-import argparse
-import torch
 import nvtx
-from itertools import accumulate
+import torch
+
 from vllm.model_executor.layers.rotary_embedding import get_rope
 
 

cmake/hipify.py

@@ -9,8 +9,8 @@
 #
 
 import argparse
-import shutil
 import os
+import shutil
 
 from torch.utils.hipify.hipify_python import hipify
 

collect_env.py

@@ -6,10 +6,10 @@
 # Run it with `python collect_env.py` or `python -m torch.utils.collect_env`
 import datetime
 import locale
+import os
 import re
 import subprocess
 import sys
-import os
 from collections import namedtuple
 
 try:

docs/source/conf.py

@@ -10,10 +10,11 @@
 # add these directories to sys.path here. If the directory is relative to the
 # documentation root, use os.path.abspath to make it absolute, like shown here.
 
+import logging
 import os
 import sys
+
 from sphinx.ext import autodoc
-import logging
 
 sys.path.insert(0, os.path.abspath(os.path.join('..', '..')))
 

examples/gradio_openai_chatbot_webserver.py

@@ -1,6 +1,7 @@
 import argparse
-from openai import OpenAI
+
 import gradio as gr
+from openai import OpenAI
 
 # Argument parser setup
 parser = argparse.ArgumentParser(

examples/llm_engine_example.py

@@ -1,7 +1,7 @@
 import argparse
 from typing import List, Tuple
 
-from vllm import EngineArgs, LLMEngine, SamplingParams, RequestOutput
+from vllm import EngineArgs, LLMEngine, RequestOutput, SamplingParams
 
 
 def create_test_prompts() -> List[Tuple[str, SamplingParams]]:

examples/multilora_inference.py

@@ -5,11 +5,11 @@
 Requires HuggingFace credentials for access to Llama2.
 """
 
-from typing import Optional, List, Tuple
+from typing import List, Optional, Tuple
 
 from huggingface_hub import snapshot_download
 
-from vllm import EngineArgs, LLMEngine, SamplingParams, RequestOutput
+from vllm import EngineArgs, LLMEngine, RequestOutput, SamplingParams
 from vllm.lora.request import LoRARequest
 
 

examples/offline_inference_distributed.py

@@ -5,11 +5,13 @@
 Learn more about Ray Data in https://docs.ray.io/en/latest/data/data.html
 """
 
-from vllm import LLM, SamplingParams
 from typing import Dict
+
 import numpy as np
 import ray
 
+from vllm import LLM, SamplingParams
+
 # Create a sampling params object.
 sampling_params = SamplingParams(temperature=0.8, top_p=0.95)
 

format.sh

@@ -25,6 +25,7 @@ YAPF_VERSION=$(yapf --version | awk '{print $2}')
 RUFF_VERSION=$(ruff --version | awk '{print $2}')
 MYPY_VERSION=$(mypy --version | awk '{print $2}')
 CODESPELL_VERSION=$(codespell --version)
+ISORT_VERSION=$(isort --vn)
 
 # # params: tool name, tool version, required version
 tool_version_check() {
@@ -37,6 +38,7 @@ tool_version_check() {
 tool_version_check "yapf" $YAPF_VERSION "$(grep yapf requirements-dev.txt | cut -d'=' -f3)"
 tool_version_check "ruff" $RUFF_VERSION "$(grep "ruff==" requirements-dev.txt | cut -d'=' -f3)"
 tool_version_check "mypy" "$MYPY_VERSION" "$(grep mypy requirements-dev.txt | cut -d'=' -f3)"
+tool_version_check "isort" "$ISORT_VERSION" "$(grep isort requirements-dev.txt | cut -d'=' -f3)"
 tool_version_check "codespell" "$CODESPELL_VERSION" "$(grep codespell requirements-dev.txt | cut -d'=' -f3)"
 
 YAPF_FLAGS=(
@@ -178,6 +180,46 @@ else
    lint_changed
 fi
 
+# check spelling of specified files
+isort_check() {
+    isort "$@"
+}
+
+isort_check_all(){
+  isort .
+}
+
+# Spelling  check of files that differ from main branch.
+isort_check_changed() {
+    # The `if` guard ensures that the list of filenames is not empty, which
+    # could cause ruff to receive 0 positional arguments, making it hang
+    # waiting for STDIN.
+    #
+    # `diff-filter=ACM` and $MERGEBASE is to ensure we only lint files that
+    # exist on both branches.
+    MERGEBASE="$(git merge-base origin/main HEAD)"
+
+    if ! git diff --diff-filter=ACM --quiet --exit-code "$MERGEBASE" -- '*.py' '*.pyi' &>/dev/null; then
+        git diff --name-only --diff-filter=ACM "$MERGEBASE" -- '*.py' '*.pyi' | xargs \
+             isort
+    fi
+}
+
+# Run Isort
+# This flag runs spell check of individual files. --files *must* be the first command line
+# arg to use this option.
+if [[ "$1" == '--files' ]]; then
+   isort_check "${@:2}"
+   # If `--all` is passed, then any further arguments are ignored and the
+   # entire python directory is linted.
+elif [[ "$1" == '--all' ]]; then
+   isort_check_all
+else
+   # Check spelling only of the files that changed in last commit.
+   isort_check_changed
+fi
+echo 'vLLM isort: Done'
+
 if ! git diff --quiet &>/dev/null; then
     echo 'Reformatted files. Please review and stage the changes.'
     echo 'Changes not staged for commit:'

pyproject.toml

@@ -51,3 +51,7 @@ exclude = "vllm/model_executor/parallel_utils/|vllm/model_executor/models/"
 [tool.codespell]
 ignore-words-list = "dout, te, indicies"
 skip = "./tests/prompts"
+
+[tool.isort]
+use_parentheses = true
+skip_gitignore = true

requirements-dev.txt

@@ -4,6 +4,7 @@ toml==0.10.2
 tomli==2.0.1
 ruff==0.1.5
 codespell==2.2.6
+isort==5.13.2
 
 # type checking
 mypy==0.991

setup.py

@@ -1,16 +1,16 @@
 import io
+import logging
 import os
 import re
-import logging
 import subprocess
 import sys
+from shutil import which
 from typing import List
 
-from packaging.version import parse, Version
-from setuptools import setup, find_packages, Extension
-from setuptools.command.build_ext import build_ext
-from shutil import which
 import torch
+from packaging.version import Version, parse
+from setuptools import Extension, find_packages, setup
+from setuptools.command.build_ext import build_ext
 from torch.utils.cpp_extension import CUDA_HOME
 
 ROOT_DIR = os.path.dirname(__file__)

tests/async_engine/test_chat_template.py

@@ -1,12 +1,12 @@
-from dataclasses import dataclass
 import os
 import pathlib
+from dataclasses import dataclass
 
 import pytest
 
-from vllm.transformers_utils.tokenizer import get_tokenizer
-from vllm.entrypoints.openai.serving_chat import OpenAIServingChat
 from vllm.entrypoints.openai.protocol import ChatCompletionRequest
+from vllm.entrypoints.openai.serving_chat import OpenAIServingChat
+from vllm.transformers_utils.tokenizer import get_tokenizer
 
 chatml_jinja_path = pathlib.Path(os.path.dirname(os.path.abspath(
     __file__))).parent.parent / "examples/template_chatml.jinja"

tests/conftest.py

@@ -6,8 +6,8 @@
 from transformers import AutoModelForCausalLM
 
 from vllm import LLM, SamplingParams
-from vllm.transformers_utils.tokenizer import get_tokenizer
 from vllm.config import TokenizerPoolConfig
+from vllm.transformers_utils.tokenizer import get_tokenizer
 
 _TEST_DIR = os.path.dirname(__file__)
 _TEST_PROMPTS = [os.path.join(_TEST_DIR, "prompts", "example.txt")]

tests/core/test_block_manager.py

@@ -1,13 +1,14 @@
-import pytest
 import time
 from typing import List
 
+import pytest
+
 from vllm import SamplingParams
 from vllm.block import PhysicalTokenBlock
-from vllm.core.block_manager import (UncachedBlockAllocator, BlockSpaceManager,
-                                     AllocStatus)
+from vllm.core.block_manager import (AllocStatus, BlockSpaceManager,
+                                     UncachedBlockAllocator)
+from vllm.sequence import Logprob, Sequence, SequenceGroup, SequenceStatus
 from vllm.utils import Device
-from vllm.sequence import Sequence, SequenceGroup, SequenceStatus, Logprob
 
 from .utils import create_dummy_prompt
 

tests/core/test_scheduler.py

@@ -1,10 +1,11 @@
+import time
 from typing import List
+
 import pytest  # noqa
-import time
 
 from vllm.config import CacheConfig, SchedulerConfig
 from vllm.core.scheduler import Scheduler
-from vllm.sequence import SequenceGroup, Logprob
+from vllm.sequence import Logprob, SequenceGroup
 
 from .utils import create_dummy_prompt
 

tests/distributed/test_comm_ops.py

@@ -3,14 +3,12 @@
 Run `pytest tests/distributed/test_comm_ops.py --forked`.
 """
 import pytest
-import torch
 import ray
+import torch
 
 from vllm.model_executor.parallel_utils.communication_op import (
-    tensor_model_parallel_all_reduce,
-    tensor_model_parallel_all_gather,
-    broadcast_tensor_dict,
-)
+    broadcast_tensor_dict, tensor_model_parallel_all_gather,
+    tensor_model_parallel_all_reduce)
 from vllm.test_utils import (init_test_distributed_environment,
                              multi_process_tensor_parallel)
 

tests/distributed/test_custom_all_reduce.py

@@ -1,6 +1,6 @@
+import os
 import random
 
-import os
 import pytest
 import ray
 import torch

tests/entrypoints/test_guided_processors.py

@@ -1,11 +1,11 @@
 # This unit test should be moved to a new
 # tests/test_guided_decoding directory.
 
-from transformers import AutoTokenizer
 import torch
+from transformers import AutoTokenizer
 
-from vllm.model_executor.guided_logits_processors import (RegexLogitsProcessor,
-                                                          JSONLogitsProcessor)
+from vllm.model_executor.guided_logits_processors import (JSONLogitsProcessor,
+                                                          RegexLogitsProcessor)
 
 TEST_SCHEMA = {
     "type": "object",