Merge pull request #24 from i-dot-ai/feature/REDBOX-4-setup-mypy-for-…

…type-hinting-testing

Feature/redbox 4 setup mypy for type hinting testing

.github/workflows/python-app.yml

@@ -47,3 +47,7 @@ jobs:
     - name: Format code
       run: |
         make format
+
+    - name: Type check
+      run: |
+        make checktypes

Makefile

@@ -36,3 +36,7 @@ format:
 	poetry run ruff format .
 	# additionally we format, but not lint, the notebooks
 	poetry run ruff format **/*.ipynb
+
+checktypes:
+	poetry run mypy redbox app tests --ignore-missing-imports
+	poetry run mypy legacy_app --follow-imports skip --ignore-missing-imports

app/workers/embed/app.py

@@ -2,7 +2,6 @@
 import os
 from contextlib import asynccontextmanager
 from datetime import datetime
-from typing import List
 from uuid import uuid4
 
 import pydantic
@@ -22,23 +21,23 @@ class ModelInfo(pydantic.BaseModel):
 class Embedding(pydantic.BaseModel):
     object: str = "embedding"
     index: int
-    embedding: List[float]
+    embedding: list[float]
 
 
 class EmbeddingResponse(pydantic.BaseModel):
     object: str = "list"
-    data: List[Embedding]
+    data: list[Embedding]
     embedding_id: str
     model: str
     model_info: ModelInfo
 
 
 class EmbeddingRequest(pydantic.BaseModel):
-    sentences: List[str]
+    sentences: list[str]
 
 
 class ModelListResponse(pydantic.BaseModel):
-    models: List[ModelInfo]
+    models: list[ModelInfo]
 
 
 class StatusResponse(pydantic.BaseModel):
@@ -174,12 +173,12 @@ def get_model(model: str):
 
 
 @app.post("/models/{model}/embed", response_model=EmbeddingResponse, tags=["models"])
-def embed_sentences(model: str, sentences: List[str]):
+def embed_sentences(model: str, sentences: list[str]):
     """Embeds a list of sentences using a given model
 
     Args:
         model (str): The name of the model
-        sentences (List[str]): A list of sentences
+        sentences (list[str]): A list of sentences
 
     Returns:
         EmbeddingResponse: The embeddings of the sentences

legacy_app/utils.py

@@ -5,7 +5,7 @@
 import uuid
 from datetime import datetime
 from io import BytesIO
-from typing import Dict, List, Optional, Union
+from typing import Optional
 
 import boto3
 import cognitojwt
@@ -86,6 +86,9 @@ def populate_user_info(ENV: dict) -> dict:
             st.sidebar.markdown("Running Locally")
             return {"name": "dev", "email": "dev@example.com"}
 
+    # TODO: is this the right thing to do here?
+    return {}
+
 
 def init_session_state() -> dict:
     """
@@ -257,7 +260,7 @@ def init_session_state() -> dict:
 
 
 def get_link_html(
-    page: str, text: str, query_dict: dict = None, target: str = "_self"
+    page: str, text: str, query_dict: Optional[dict] = None, target: str = "_self"
 ) -> str:
     """Returns a link in HTML format
 
@@ -279,7 +282,7 @@ def get_link_html(
     return f"<a href='/{page}{query}' target={target}><button style='background-color: white;border-radius: 8px;'>{text}</button></a>"
 
 
-def get_file_link(file: File, page: int = None) -> str:
+def get_file_link(file: File, page: Optional[int] = None) -> str:
     """Returns a link to a file
 
     Args:
@@ -296,7 +299,7 @@ def get_file_link(file: File, page: int = None) -> str:
         presentation_name = file.name
     query_dict = {"file_uuid": file.uuid}
     if page is not None:
-        query_dict["page_number"] = page[0]
+        query_dict["page_number"] = page
 
     link_html = get_link_html(
         page="Preview_Files",
@@ -360,7 +363,7 @@ def load_llm_handler(ENV, update=False) -> None:
             )
 
 
-def hash_list_of_files(list_of_files: List[File]) -> str:
+def hash_list_of_files(list_of_files: list[File]) -> str:
     """Returns a hash of the list of files
 
     Args:
@@ -425,7 +428,7 @@ def st_render(self, file: File) -> None:
         file_bytes = stream["Body"].read()
         render_method(file, file_bytes)
 
-    def _render_pdf(self, file: File, page_number: int = None) -> None:
+    def _render_pdf(self, file: File, page_number: Optional[int] = None) -> None:
         stream = st.session_state.s3_client.get_object(
             Bucket=st.session_state.BUCKET_NAME, Key=file.name
         )
@@ -480,8 +483,8 @@ def _render_docx(self, file: File, file_bytes: bytes) -> None:
 
 def replace_doc_ref(
     output_for_render: str = "",
-    files: List[File] = [],
-    page_numbers: List = [],
+    files: Optional[list[File]] = None,
+    page_numbers: Optional[list] = None,
     flexible=False,
 ) -> str:
     """Replaces references to files in the output text with links to the files
@@ -495,6 +498,9 @@ def replace_doc_ref(
     Returns:
         str: The modified text
     """
+    files = files or []
+    page_numbers = page_numbers or []
+
     if len(page_numbers) != len(files):
         page_numbers = [None for _ in files]
 
@@ -537,8 +543,8 @@ def eval_csv_to_squad_json(csv_path: str, json_path: str) -> None:
 
 
 def submit_feedback(
-    feedback: Dict,
-    input: Union[str, List[str]],
+    feedback: dict,
+    input: str | list[str],
     output: str,
     creator_user_uuid: str,
     chain: Optional[Chain] = None,

pyproject.toml

@@ -47,7 +47,9 @@ ruff = "^0.2.1"
 pytest = "^8.0.1"
 pre-commit = "^3.6.1"
 pylint = "^3.0.3"
-
+mypy = "^1.8.0"
+types-python-dateutil = "^2.8.19.20240106"
+types-markdown = "^3.5.0.20240129"
 
 
 [build-system]

redbox/export/docx.py

@@ -1,5 +1,5 @@
 import tempfile
-from typing import List, Optional
+from typing import Optional
 
 import markdown
 from dateutil import parser
@@ -20,7 +20,7 @@ def lookup_indentedness(raw: str, line_str_to_match: str):
 
 def spotlight_complete_to_docx(
     spotlight_complete: SpotlightComplete,
-    files: List[File],
+    files: list[File],
     title: Optional[str] = None,
 ):
     document = Document()