fix(infra): conform ruff to 150 LL (#781)

Generally correctly format it with ruff format and manual style

Signed-off-by: Aaron <29749331+aarnphm@users.noreply.github.com>

.pre-commit-config.yaml

@@ -8,7 +8,7 @@ default_language_version:
 exclude: '.*\.(css|js|svg)$'
 repos:
   - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: 'v0.1.7'
+    rev: 'v0.1.8'
     hooks:
       - id: ruff
         alias: r

.ruff.toml

@@ -1,77 +1,38 @@
-extend-exclude = [
-    "tools",
-    "examples",
-    "openllm-python/src/openllm/__init__.py",
-    "openllm-python/src/openllm/_version.py",
-    "openllm-python/src/openllm/models/__init__.py",
-    "openllm-python/src/openllm_cli/playground",
-    "openllm-client/src/openllm_client/pb/**",
-]
+exclude = ["tools", "examples", "openllm-python/src/openllm_cli/playground/"]
 extend-include = ["*.ipynb"]
-extend-select = [
-    "E",
+preview = true
+select = [
     "F",
-    "B",
-    "PIE",
     "G",    # flake8-logging-format
-    "W",    # pycodestyle
-    "Q",    # flake8-quotes
-    "FA",   # flake8-future-annotations
-    "TCH",  # flake8-type-checking
-    "PLW",  # pylint-warning
-    "PLR",  # pylint-refactor
-    "PT",   # flake8-pytest-style
     "PERF", # perflint
     "RUF",  # Ruff-specific rules
-    "YTT",  # flake8-2020
-]
-fix = true
-ignore = [
-    "PLR0911",
-    "PLR0912",
-    "PLR0913",
-    "PLR0915",
-    "ANN",     # Use mypy
-    "PLR2004", # magic value to use constant
-    "E501",    # ignore line length violation
-    "E401",    # ignore multiple line import
+    "W6",
+    "E71",
+    "E72",
+    "E112",
+    "E113",
+    # "E124",
+    "E203",
+    "E272",
+    # "E303",
+    # "E304",
+    # "E501",
+    # "E502",
     "E702",
-    "TCH004",  # don't move runtime import out, just warn about it
-    "RUF012",  # mutable attributes to be used with ClassVar
-    "E701",    # multiple statement on single line
+    "E703",
+    "E731",
+    "W191",
+    "W291",
+    "W293",
+    "UP039", # unnecessary-class-parentheses
 ]
-line-length = 119
+ignore = ["RUF012"]
+line-length = 150
 indent-width = 2
-target-version = "py38"
 typing-modules = [
     "openllm_core._typing_compat",
     "openllm_client._typing_compat",
 ]
-unfixable = ["TCH004"]
-
-[lint.flake8-type-checking]
-exempt-modules = [
-    "typing",
-    "typing_extensions",
-    "openllm_core._typing_compat",
-    "openllm_client._typing_compat",
-]
-runtime-evaluated-base-classes = [
-    "openllm_core._configuration.LLMConfig",
-    "openllm_core._configuration.GenerationConfig",
-    "openllm_core._configuration.SamplingParams",
-    "openllm_core._configuration.ModelSettings",
-    "openllm.LLMConfig",
-]
-runtime-evaluated-decorators = [
-    "attrs.define",
-    "attrs.frozen",
-    "trait",
-    "attr.attrs",
-    'attr.define',
-    '_attr.define',
-    'attr.frozen',
-]
 
 [format]
 preview = true
@@ -81,20 +42,3 @@ skip-magic-trailing-comma = true
 
 [lint.pydocstyle]
 convention = "google"
-
-[lint.pycodestyle]
-ignore-overlong-task-comments = true
-max-line-length = 119
-
-[lint.flake8-quotes]
-avoid-escape = false
-inline-quotes = "single"
-multiline-quotes = "single"
-docstring-quotes = "single"
-
-[lint.extend-per-file-ignores]
-"openllm-python/tests/**/*" = ["S101", "TID252", "PT011", "S307"]
-"openllm-python/src/openllm/_llm.py" = ["F811"]
-"openllm-core/src/openllm_core/utils/import_utils.py" = ["PLW0603", "F811"]
-"openllm-core/src/openllm_core/_configuration.py" = ["F811", "Q001"]
-"openllm-python/src/openllm/_service_vars_pkg.py" = ["F821"]

cz.py

@@ -22,21 +22,13 @@ def run_cz(args):
       with tokenize.open(filepath) as file_:
         tokens = [t for t in tokenize.generate_tokens(file_.readline) if t.type in TOKEN_WHITELIST]
         token_count, line_count = len(tokens), len(set([t.start[0] for t in tokens]))
-        table.append(
-          [
-            filepath.replace(os.path.join(args.dir, 'src'), ''),
-            line_count,
-            token_count / line_count if line_count != 0 else 0,
-          ]
-        )
+        table.append([filepath.replace(os.path.join(args.dir, 'src'), ''), line_count, token_count / line_count if line_count != 0 else 0])
   print(tabulate([headers, *sorted(table, key=lambda x: -x[1])], headers='firstrow', floatfmt='.1f') + '\n')
   print(
     tabulate(
       [
         (dir_name, sum([x[1] for x in group]))
-        for dir_name, group in itertools.groupby(
-          sorted([(x[0].rsplit('/', 1)[0], x[1]) for x in table]), key=lambda x: x[0]
-        )
+        for dir_name, group in itertools.groupby(sorted([(x[0].rsplit('/', 1)[0], x[1]) for x in table]), key=lambda x: x[0])
       ],
       headers=['Directory', 'LOC'],
       floatfmt='.1f',
@@ -54,10 +46,6 @@ def run_cz(args):
 
   parser = argparse.ArgumentParser()
   parser.add_argument(
-    '--dir',
-    choices=['openllm-python', 'openllm-core', 'openllm-client'],
-    help='directory to check',
-    default='openllm-python',
-    required=False,
+    '--dir', choices=['openllm-python', 'openllm-core', 'openllm-client'], help='directory to check', default='openllm-python', required=False
   )
   raise SystemExit(run_cz(parser.parse_args()))

openllm-client/src/openllm_client/__init__.pyi

@@ -15,17 +15,11 @@ class HTTPClient:
   address: str
   helpers: _Helpers
   @overload
-  def __init__(
-    self, address: str, timeout: int = ..., verify: bool = ..., max_retries: int = ..., api_version: str = ...
-  ) -> None: ...
+  def __init__(self, address: str, timeout: int = ..., verify: bool = ..., max_retries: int = ..., api_version: str = ...) -> None: ...
   @overload
-  def __init__(
-    self, address: str = ..., timeout: int = ..., verify: bool = ..., max_retries: int = ..., api_version: str = ...
-  ) -> None: ...
+  def __init__(self, address: str = ..., timeout: int = ..., verify: bool = ..., max_retries: int = ..., api_version: str = ...) -> None: ...
   @overload
-  def __init__(
-    self, address: None = ..., timeout: int = ..., verify: bool = ..., max_retries: int = ..., api_version: str = ...
-  ) -> None: ...
+  def __init__(self, address: None = ..., timeout: int = ..., verify: bool = ..., max_retries: int = ..., api_version: str = ...) -> None: ...
   @property
   def is_ready(self) -> bool: ...
   def health(self) -> bool: ...
@@ -66,17 +60,11 @@ class AsyncHTTPClient:
   address: str
   helpers: _AsyncHelpers
   @overload
-  def __init__(
-    self, address: str, timeout: int = ..., verify: bool = ..., max_retries: int = ..., api_version: str = ...
-  ) -> None: ...
+  def __init__(self, address: str, timeout: int = ..., verify: bool = ..., max_retries: int = ..., api_version: str = ...) -> None: ...
   @overload
-  def __init__(
-    self, address: str = ..., timeout: int = ..., verify: bool = ..., max_retries: int = ..., api_version: str = ...
-  ) -> None: ...
+  def __init__(self, address: str = ..., timeout: int = ..., verify: bool = ..., max_retries: int = ..., api_version: str = ...) -> None: ...
   @overload
-  def __init__(
-    self, address: None = ..., timeout: int = ..., verify: bool = ..., max_retries: int = ..., api_version: str = ...
-  ) -> None: ...
+  def __init__(self, address: None = ..., timeout: int = ..., verify: bool = ..., max_retries: int = ..., api_version: str = ...) -> None: ...
   @property
   def is_ready(self) -> bool: ...
   async def health(self) -> bool: ...

openllm-client/src/openllm_client/_http.py

@@ -70,14 +70,10 @@ def query(self, prompt, **attrs):
     return self.generate(prompt, **attrs)
 
   def health(self):
-    response = self._get(
-      '/readyz', response_cls=None, options={'return_raw_response': True, 'max_retries': self._max_retries}
-    )
+    response = self._get('/readyz', response_cls=None, options={'return_raw_response': True, 'max_retries': self._max_retries})
     return response.status_code == 200
 
-  def generate(
-    self, prompt, llm_config=None, stop=None, adapter_name=None, timeout=None, verify=None, **attrs
-  ) -> Response:
+  def generate(self, prompt, llm_config=None, stop=None, adapter_name=None, timeout=None, verify=None, **attrs) -> Response:
     if timeout is None:
       timeout = self._timeout
     if verify is None:
@@ -100,9 +96,7 @@ def generate_stream(
     for response_chunk in self.generate_iterator(prompt, llm_config, stop, adapter_name, timeout, verify, **attrs):
       yield StreamingResponse.from_response_chunk(response_chunk)
 
-  def generate_iterator(
-    self, prompt, llm_config=None, stop=None, adapter_name=None, timeout=None, verify=None, **attrs
-  ) -> t.Iterator[Response]:
+  def generate_iterator(self, prompt, llm_config=None, stop=None, adapter_name=None, timeout=None, verify=None, **attrs) -> t.Iterator[Response]:
     if timeout is None:
       timeout = self._timeout
     if verify is None:
@@ -152,9 +146,7 @@ def _build_auth_headers(self) -> t.Dict[str, str]:
   @property
   async def _metadata(self) -> t.Awaitable[Metadata]:
     if self.__metadata is None:
-      self.__metadata = await self._post(
-        f'/{self._api_version}/metadata', response_cls=Metadata, json={}, options={'max_retries': self._max_retries}
-      )
+      self.__metadata = await self._post(f'/{self._api_version}/metadata', response_cls=Metadata, json={}, options={'max_retries': self._max_retries})
     return self.__metadata
 
   @property
@@ -167,14 +159,10 @@ async def query(self, prompt, **attrs):
     return await self.generate(prompt, **attrs)
 
   async def health(self):
-    response = await self._get(
-      '/readyz', response_cls=None, options={'return_raw_response': True, 'max_retries': self._max_retries}
-    )
+    response = await self._get('/readyz', response_cls=None, options={'return_raw_response': True, 'max_retries': self._max_retries})
     return response.status_code == 200
 
-  async def generate(
-    self, prompt, llm_config=None, stop=None, adapter_name=None, timeout=None, verify=None, **attrs
-  ) -> Response:
+  async def generate(self, prompt, llm_config=None, stop=None, adapter_name=None, timeout=None, verify=None, **attrs) -> Response:
     if timeout is None:
       timeout = self._timeout
     if verify is None:
@@ -195,9 +183,7 @@ async def generate(
   async def generate_stream(
     self, prompt, llm_config=None, stop=None, adapter_name=None, timeout=None, verify=None, **attrs
   ) -> t.AsyncGenerator[StreamingResponse, t.Any]:
-    async for response_chunk in self.generate_iterator(
-      prompt, llm_config, stop, adapter_name, timeout, verify, **attrs
-    ):
+    async for response_chunk in self.generate_iterator(prompt, llm_config, stop, adapter_name, timeout, verify, **attrs):
       yield StreamingResponse.from_response_chunk(response_chunk)
 
   async def generate_iterator(

openllm-client/src/openllm_client/_schemas.py

@@ -22,9 +22,9 @@
 
 @attr.define
 class Metadata(_SchemaMixin):
-  '''NOTE: Metadata is a modified version of the original MetadataOutput from openllm-core.
+  """NOTE: Metadata is a modified version of the original MetadataOutput from openllm-core.
 
-  The configuration is now structured into a dictionary for easy of use.'''
+  The configuration is now structured into a dictionary for easy of use."""
 
   model_id: str
   timeout: int
@@ -42,11 +42,7 @@ def _structure_metadata(data: t.Dict[str, t.Any], cls: type[Metadata]) -> Metada
     raise RuntimeError(f'Malformed metadata configuration (Server-side issue): {e}') from None
   try:
     return cls(
-      model_id=data['model_id'],
-      timeout=data['timeout'],
-      model_name=data['model_name'],
-      backend=data['backend'],
-      configuration=configuration,
+      model_id=data['model_id'], timeout=data['timeout'], model_name=data['model_name'], backend=data['backend'], configuration=configuration
     )
   except Exception as e:
     raise RuntimeError(f'Malformed metadata (Server-side issue): {e}') from None
@@ -65,10 +61,7 @@ class StreamingResponse(_SchemaMixin):
   @classmethod
   def from_response_chunk(cls, response: Response) -> StreamingResponse:
     return cls(
-      request_id=response.request_id,
-      index=response.outputs[0].index,
-      text=response.outputs[0].text,
-      token_ids=response.outputs[0].token_ids[0],
+      request_id=response.request_id, index=response.outputs[0].index, text=response.outputs[0].text, token_ids=response.outputs[0].token_ids[0]
     )
 
 
@@ -95,17 +88,11 @@ def async_client(self):
     return self._async_client
 
   def messages(self, messages, add_generation_prompt=False):
-    return self.client._post(
-      '/v1/helpers/messages',
-      response_cls=str,
-      json=dict(messages=messages, add_generation_prompt=add_generation_prompt),
-    )
+    return self.client._post('/v1/helpers/messages', response_cls=str, json=dict(messages=messages, add_generation_prompt=add_generation_prompt))
 
   async def async_messages(self, messages, add_generation_prompt=False):
     return await self.async_client._post(
-      '/v1/helpers/messages',
-      response_cls=str,
-      json=dict(messages=messages, add_generation_prompt=add_generation_prompt),
+      '/v1/helpers/messages', response_cls=str, json=dict(messages=messages, add_generation_prompt=add_generation_prompt)
     )
 
   @classmethod