fix: CI lint and test issues (#102)

* fix: CI problems

- [x] Update ruff to match CI and apply fmt
- [x] bump timeouts
- [x] fix openapi client

.github/workflows/ci.yml

@@ -38,6 +38,7 @@ jobs:
         run: poetry run pytest -v
         env:
           OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
+          PYTEST_TIMEOUT: 120
       - name: check openapi client up to date
         run: |
           poetry run poe extract-openapi

dewy-client/dewy_client/__init__.py

@@ -1,4 +1,5 @@
-""" A client library for accessing Dewy Knowledge Base API """
+"""A client library for accessing Dewy Knowledge Base API"""
+
 from .client import AuthenticatedClient, Client
 
 __all__ = (

dewy-client/dewy_client/api/__init__.py

@@ -1 +1 @@
-""" Contains methods for accessing the API """
+"""Contains methods for accessing the API"""

dewy-client/dewy_client/api/kb/add_document.py

@@ -66,7 +66,7 @@ def sync_detailed(
 ) -> Response[Union[Document, HTTPValidationError]]:
     """Add Document
 
-     Add a document from a URL.
+     Add a document to a collection.
 
     Args:
         body (AddDocumentRequest):
@@ -97,7 +97,7 @@ def sync(
 ) -> Optional[Union[Document, HTTPValidationError]]:
     """Add Document
 
-     Add a document from a URL.
+     Add a document to a collection.
 
     Args:
         body (AddDocumentRequest):
@@ -123,7 +123,7 @@ async def asyncio_detailed(
 ) -> Response[Union[Document, HTTPValidationError]]:
     """Add Document
 
-     Add a document from a URL.
+     Add a document to a collection.
 
     Args:
         body (AddDocumentRequest):
@@ -152,7 +152,7 @@ async def asyncio(
 ) -> Optional[Union[Document, HTTPValidationError]]:
     """Add Document
 
-     Add a document from a URL.
+     Add a document to a collection.
 
     Args:
         body (AddDocumentRequest):

dewy-client/dewy_client/errors.py

@@ -1,4 +1,4 @@
-""" Contains shared errors types that can be raised from API functions """
+"""Contains shared errors types that can be raised from API functions"""
 
 
 class UnexpectedStatus(Exception):

dewy-client/dewy_client/models/__init__.py

@@ -1,4 +1,4 @@
-""" Contains all the data models used in inputs/outputs """
+"""Contains all the data models used in inputs/outputs"""
 
 from .add_document_request import AddDocumentRequest
 from .body_upload_document_content import BodyUploadDocumentContent

dewy-client/dewy_client/types.py

@@ -1,4 +1,5 @@
-""" Contains some shared types for properties """
+"""Contains some shared types for properties"""
+
 from http import HTTPStatus
 from typing import BinaryIO, Generic, Literal, MutableMapping, Optional, Tuple, TypeVar
 

dewy-client/pyproject.toml

@@ -21,5 +21,7 @@ requires = ["poetry-core>=1.0.0"]
 build-backend = "poetry.core.masonry.api"
 
 [tool.ruff]
-select = ["F", "I", "UP"]
 line-length = 120
+
+[tool.ruff.lint]
+select = ["F", "I", "UP"]

dewy/common/collection_embeddings.py

@@ -250,7 +250,7 @@ def encode_chunk(c: str) -> str:
                     # (https://github.com/DewyKB/dewy/issues/20). We do not know
                     # that all of this is truly necessary.
                     encoded = c.encode("utf-8").decode("utf-8", "ignore")
-                    return encoded.replace("\x00", "\uFFFD")
+                    return encoded.replace("\x00", "\ufffd")
 
                 # First, insert the chunks.
                 await conn.executemany(

openapi.yaml

@@ -114,7 +114,7 @@ paths:
       tags:
       - kb
       summary: Add Document
-      description: Add a document from a URL.
+      description: Add a document to a collection.
       operationId: addDocument
       requestBody:
         required: true

pyproject.toml

@@ -70,9 +70,10 @@ langchain-core = "^0.1.27"
 langchain-openai = "^0.0.8"
 langchain-community = "^0.0.24"
 langchain = "^0.1.9"
+sentence-transformers = "^2.5.0"
 
 [tool.poetry.group.dev.dependencies]
-ruff = "^0.1.11"
+ruff = "^0.3.0"
 mypy = "^1.8.0"
 ipykernel = "^6.28.0"
 asyncpg-stubs = "^0.29.1"
@@ -93,6 +94,10 @@ build-backend = "poetry.core.masonry.api"
 [tool.ruff]
 line-length = 98
 
+# Assume Python 3.11.
+target-version = "py311"
+
+[tool.ruff.lint]
 # Enable Pyflakes `E` and `F` codes by default.
 select = [
     # Pyflakes
@@ -108,10 +113,7 @@ ignore = []
 # Allow unused variables when underscore-prefixed.
 dummy-variable-rgx = "^(_+|(_+[a-zA-Z0-9_]*[a-zA-Z0-9]+?))$"
 
-# Assume Python 3.11.
-target-version = "py311"
-
-[tool.ruff.mccabe]
+[tool.ruff.lint.mccabe]
 # Unlike Flake8, default to a complexity level of 10.
 max-complexity = 10
 
@@ -125,4 +127,4 @@ show_error_context = true
 [tool.pytest.ini_options]
 testpaths = ["tests"]
 asyncio_mode = "auto"
-timeout = 10
+timeout = 20

tests/test_e2e.py

@@ -30,7 +30,6 @@
     "embedding_model",
     ["openai:text-embedding-ada-002", "hf:BAAI/bge-small-en", "hf:BAAI/bge-small-en-v1.5"],
 )
-@pytest.mark.timeout(120)  # slow due to embedding (especially in CI)
 async def test_index_retrieval(client, embedding_model):
     collection_name = "".join(random.choices(string.ascii_lowercase, k=5))