add more ruff rules

docs/examples/gallery_streaming.py

@@ -1,5 +1,4 @@
-"""
-# Streaming messages
+"""# Streaming messages
 
 Ragna supports streaming responses from the assistant. This example showcases how this
 is performed using the Python and REST API.

docs/tutorials/gallery_custom_components.py

@@ -1,5 +1,4 @@
-"""
-# Custom Components
+"""# Custom Components
 
 While Ragna has builtin support for a few [source storages][ragna.source_storages]
 and [assistants][ragna.assistants], its real strength lies in allowing users
@@ -84,7 +83,7 @@ def retrieve(
 #     [streaming example](../../generated/examples/gallery_streaming.md) for more
 #     information.
 
-from typing import Iterator
+from collections.abc import Iterator
 
 from ragna.core import Assistant, Source
 
@@ -402,7 +401,7 @@ def answer(
 
 import asyncio
 import time
-from typing import AsyncIterator
+from collections.abc import AsyncIterator
 
 
 class AsyncAssistant(Assistant):

docs/tutorials/gallery_python_api.py

@@ -1,5 +1,4 @@
-"""
-# Python API
+"""# Python API
 
 The [Python API](../../references/python-api.md) is the best place to get started with
 Ragna and understand its key components. It's also the best way to continue

docs/tutorials/gallery_rest_api.py

@@ -1,5 +1,4 @@
-"""
-# REST API
+"""# REST API
 
 Ragna was designed to help you quickly build custom RAG powered web
 applications. For this you can leverage the built-in
@@ -117,10 +116,13 @@
 # - The field name is the ID of the document returned by step 1.
 # - The field value is the binary content of the document.
 
-client.put(
-    "/api/documents",
-    files=[("documents", (documents[0]["id"], open(document_path, "rb")))],
-)
+with open(document_path, "rb") as f:
+    print(
+        client.put(
+            "/api/documents",
+            files=[("documents", (documents[0]["id"], f))],
+        )
+    )
 
 # %%
 # ## Step 4: Select a source storage and assistant

docs/tutorials/gallery_web_ui.py

@@ -1,5 +1,4 @@
-"""
-# Web UI
+"""# Web UI
 
 Ragna comes with a web UI where you can try out all of the features!
 

pyproject.toml

@@ -177,22 +177,51 @@ depends-on = [
     "build",
 ]
 
+[tool.ruff]
+target-version = "py310"
+
 [tool.ruff.lint]
+
+pydocstyle = { convention = "google" }
+
 select = [
     "E",
     "F",
-    # import sorting
-    "I001"
+    "I001",
+    "B",
+    "C4",
+    "ISC",
+    "RET",
+    "SIM",
+    "PTH",
+    "D2",
+    "UP",
+    "ASYNC",
+]
+
+ignore = [
+    # Conflicts with ruff formatter
+    "ISC001",
+    # Ignore line too long, because due to black, the error can only occur for strings
+    "E501",
+    # cache has its purpose
+    "B019",
+    # built-in open() is well understood
+    "PTH123",
+    # mutually-exclusive with D211
+    "D203",
+    # mutually-exclusive with D212
+    "D213",
 ]
-# Ignore line too long, because due to black, the error can only occur for strings
-ignore = ["E501"]
 
 [tool.ruff.lint.per-file-ignores]
 # ignore unused imports and imports not at the top of the file in __init__.py files
 "__init__.py" = ["F401", "E402"]
-# The examples often have imports below the top of the file to follow the narrative
-"docs/examples/**/*.py" = ["E402", "F704", "I001"]
-"docs/tutorials/**/*.py" = ["E402", "F704", "I001"]
+# The examples and tutorials need to have a good the narrative rather than follow our code-style rules
+"docs/examples/**/*.py" = ["E402", "F704", "I001", "D"]
+"docs/tutorials/**/*.py" = ["E402", "F704", "I001", "D"]
+# blocking code in async tests is not an issue
+"tests/**/*.py" = ["ASYNC101"]
 
 [tool.pytest.ini_options]
 minversion = "6.0"

ragna/__init__.py

@@ -3,7 +3,7 @@
 except ModuleNotFoundError:
     import warnings
 
-    warnings.warn("ragna was not properly installed!")
+    warnings.warn("ragna was not properly installed!", stacklevel=2)
     del warnings
 
     __version__ = "UNKNOWN"

ragna/_cli/config.py

@@ -1,8 +1,9 @@
 import itertools
 from collections import defaultdict
+from collections.abc import Iterable
 from pathlib import Path
 from types import ModuleType
-from typing import Annotated, Iterable, Type, TypeVar, cast
+from typing import Annotated, TypeVar, cast
 
 import pydantic
 import questionary
@@ -27,19 +28,19 @@
 def parse_config(value: str) -> Config:
     try:
         config = Config.from_file(value)
-    except RagnaException:
+    except RagnaException as exc:
         rich.print(f"The configuration file {value} does not exist.")
         if value == "./ragna.toml":
             rich.print(
                 "If you don't have a configuration file yet, "
                 "run [bold]ragna init[/bold] to generate one."
             )
-        raise typer.Exit(1)
+        raise typer.Exit(1) from exc
     except pydantic.ValidationError as validation:
         # FIXME: pretty formatting!
         for error in validation.errors():
             rich.print(error)
-        raise typer.Exit(1)
+        raise typer.Exit(1) from validation
     # This stores the original value so we can pass it on to subprocesses that we might
     # start.
     config.__ragna_cli_config_path__ = value  # type: ignore[attr-defined]
@@ -157,8 +158,8 @@ def _wizard_builtin() -> Config:
 def _select_components(
     title: str,
     module: ModuleType,
-    base_cls: Type[T],
-) -> list[Type[T]]:
+    base_cls: type[T],
+) -> list[type[T]]:
     components = sorted(
         (
             obj
@@ -170,7 +171,7 @@ def _select_components(
         key=lambda component: component.display_name(),
     )
     return cast(
-        list[Type[T]],
+        list[type[T]],
         questionary.checkbox(
             f"Which {title} do you want to use?",
             choices=[
@@ -186,13 +187,13 @@ def _select_components(
     )
 
 
-def _handle_unmet_requirements(components: Iterable[Type[Component]]) -> None:
-    unmet_requirements = set(
+def _handle_unmet_requirements(components: Iterable[type[Component]]) -> None:
+    unmet_requirements = {
         requirement
         for component in components
         for requirement in component.requirements()
         if not requirement.is_available()
-    )
+    }
     if not unmet_requirements:
         return
 
@@ -331,7 +332,7 @@ def check_config(config: Config) -> bool:
         ("source storages", config.source_storages),
         ("assistants", config.assistants),
     ]:
-        components = cast(list[Type[Component]], components)
+        components = cast(list[type[Component]], components)
 
         table = Table(
             "",
@@ -361,7 +362,7 @@ def check_config(config: Config) -> bool:
 
 def _split_requirements(
     requirements: Iterable[Requirement],
-) -> dict[Type[Requirement], list[Requirement]]:
+) -> dict[type[Requirement], list[Requirement]]:
     split_reqs = defaultdict(list)
     for req in requirements:
         split_reqs[type(req)].append(req)

ragna/_cli/core.py

@@ -1,5 +1,5 @@
 from pathlib import Path
-from typing import Annotated, Optional
+from typing import Annotated
 
 import rich
 import typer
@@ -30,7 +30,7 @@ def version_callback(value: bool) -> None:
 @app.callback()
 def _main(
     version: Annotated[
-        Optional[bool],
+        bool | None,
         typer.Option(
             "--version", callback=version_callback, help="Show version and exit."
         ),
@@ -97,7 +97,7 @@ def deploy(
         ),
     ] = False,
     open_browser: Annotated[
-        Optional[bool],
+        bool | None,
         typer.Option(
             help="Open a browser when Ragna is deployed.",
             show_default="value of ui / no-ui",

ragna/_cli/corpus.py

@@ -1,7 +1,7 @@
 import json
 import sys
 from pathlib import Path
-from typing import Annotated, Optional
+from typing import Annotated
 
 import rich
 import typer
@@ -45,7 +45,7 @@ def experimental_warning() -> None:
 def ingest(
     documents: list[Path],
     metadata_fields: Annotated[
-        Optional[Path],
+        Path | None,
         typer.Option(
             help="JSON file that contains mappings from document name "
             "to metadata fields associated with a document.",
@@ -57,7 +57,7 @@ def ingest(
     ] = "default",
     config: ConfigOption = "./ragna.toml",  # type: ignore[assignment]
     user: Annotated[
-        Optional[str],
+        str | None,
         typer.Option(help="User to link the documents to in the ragna database."),
     ] = None,
     report_failures: Annotated[
@@ -69,12 +69,12 @@ def ingest(
     ] = False,
 ) -> None:
     try:
-        document_factory = getattr(config.document, "from_path")
-    except AttributeError:
+        document_factory = config.document.from_path  # type: ignore[attr-defined]
+    except AttributeError as exc:
         raise typer.BadParameter(
             f"{config.document.__name__} does not support creating documents from a"
             f"path. Please implement a `from_path` method."
-        )
+        ) from exc
 
     database = Database(config.database_url)
     core_to_schema_document = CoreToSchemaConverter().document
@@ -83,10 +83,10 @@ def ingest(
         try:
             with open(metadata_fields) as file:
                 metadata = json.load(file)
-        except Exception:
+        except Exception as exc:
             raise typer.BadParameter(
                 f"Could not read the metadata fields file: {metadata_fields}"
-            )
+            ) from exc
     else:
         metadata = {}
 
@@ -132,14 +132,10 @@ def ingest(
                 document_instances = []
 
                 if source_storage.display_name() in ingestion_log:
-                    batch_doc_set = set(
-                        [
-                            str(doc)
-                            for doc in documents[
-                                batch_number : batch_number + BATCH_SIZE
-                            ]
-                        ]
-                    )
+                    batch_doc_set = {
+                        str(doc)
+                        for doc in documents[batch_number : batch_number + BATCH_SIZE]
+                    }
                     if batch_doc_set.issubset(
                         ingestion_log[source_storage.display_name()]
                     ):
@@ -150,12 +146,7 @@ def ingest(
                     try:
                         document_instances.append(
                             document_factory(
-                                document,
-                                metadata=(
-                                    metadata[str(document)]
-                                    if str(document) in metadata
-                                    else None
-                                ),
+                                document, metadata=metadata.get(str(document))
                             )
                         )
                     except Exception:

ragna/_docs.py

@@ -7,7 +7,7 @@
 import tempfile
 import textwrap
 from pathlib import Path
-from typing import Any, Optional, cast
+from typing import Any, cast
 
 import httpx
 
@@ -114,7 +114,7 @@ def get_http_client(
         *,
         authenticate: bool = False,
         upload_sample_document: bool = False,
-    ) -> tuple[httpx.Client, Optional[dict[str, Any]]]:
+    ) -> tuple[httpx.Client, dict[str, Any] | None]:
         if upload_sample_document and not authenticate:
             raise RagnaException(
                 "Cannot upload a document without authenticating first. "