Update

Author: Mr-Milk

.github/workflows/build.yaml

@@ -35,7 +35,7 @@ jobs:
           enable-cache: true
           cache-dependency-glob: "uv.lock"
       - name: Install project
-        run: uv sync --all-extras --dev
+        run: uv sync --dev
       - name: Tests
         run: uv run task test
 

.gitignore

@@ -69,7 +69,9 @@ instance/
 .scrapy
 
 # Sphinx documentation
-docs/_build/
+docs/build/
+docs/.jupyter_cache/
+docs/jupyter_execute
 
 # PyBuilder
 .pybuilder/
@@ -169,3 +171,4 @@ work/
 # pixi environments
 .pixi
 *.egg-info
+

pyproject.toml

@@ -4,14 +4,29 @@ build-backend = "hatchling.build"
 
 [project]
 name = "wsidata"
-authors = []
+description = "Data structures and I/O functions for whole-slide images (WSIs)."
 readme = "README.md"
+requires-python = ">=3.10"
+license = "MIT"
+authors = [
+    {name = "Yimin Zheng", email = "yzheng@cemm.at"},
+    {name = "Ernesto Abila", email = "eabila@cemm.at"},
+    {name = "André F. Rendeiro", email = "arendeiro@cemm.at"},
+]
+keywords = ["histopathology", "whole slide image", "image analysis", "data structure"]
 classifiers = [
+    "Development Status :: 3 - Alpha",
+    "Intended Audience :: Science/Research",
     "License :: OSI Approved :: MIT License",
+    "Natural Language :: English",
+    "Operating System :: OS Independent",
+    "Programming Language :: Python :: 3",
+    "Topic :: File Formats",
+    "Topic :: Scientific/Engineering :: Bio-Informatics",
 ]
-requires-python = ">=3.10"
-dynamic = ["version", "description"]
-
+Documentation = "https://wsidata.readthedocs.io"
+repository = "https://github.com/rendeirolab/wsidata"
+dynamic = ["version"]
 dependencies = [
     "spatialdata>=0.3.0",
     "opencv-python-headless",
@@ -20,7 +35,7 @@ dependencies = [
     "tiffslide",
     "rich",
     "numba>=0.61.0",
-    "torch>=2.6.0",
+    "torch>=2.0.0",
 ]
 
 [project.optional-dependencies]
@@ -50,6 +65,9 @@ include = [
     "wsidata",
 ]
 
+[tool.hatch.build.targets.wheel]
+packages = ["wsidata", "README.md", "LICENSE", "pyproject.toml"]
+
 [tool.ruff]
 lint.ignore = ["F401"]
 line-length = 88
@@ -58,7 +76,8 @@ line-length = 88
 hello = "echo Hello, World!"
 test = "pytest tests"
 doc-build = "sphinx-build -b html docs/source docs/build"
-doc-serve = "python -m http.server -d docs/build/html"
+doc-clean-build = "python docs/clean_up.py && sphinx-build -b html docs/source docs/build"
+doc-serve = "python -m http.server -d docs/build"
 fmt = "ruff format docs/source wsidata tests"
 
 [tool.uv]
@@ -82,4 +101,5 @@ dev = [
     "torchstain>=1.4.1",
     "matplotlib>=3.10.0",
     "legendkit>=0.3.4",
+    "pyright>=1.1.393",
 ]

uv.lock

@@ -11,4 +11,4 @@
     IterAccessor,
     DatasetAccessor,
 )
-from .reader import get_reader
+from .reader import get_reader, SlideProperties

wsidata/__init__.py

@@ -1,7 +1,11 @@
+from __future__ import annotations
+
 from concurrent.futures import ThreadPoolExecutor
 from pathlib import Path
+from typing import Literal
 
 import numpy as np
+import pandas as pd
 from anndata import AnnData
 from rich.progress import track
 from spatialdata import read_zarr, SpatialData
@@ -13,25 +17,24 @@
 
 
 def open_wsi(
-    wsi,
-    store=None,
-    reader=None,
-    attach_images=False,
-    image_key="wsi",
-    save_images=False,
-    attach_thumbnail=True,
-    thumbnail_key="wsi_thumbnail",
-    thumbnail_size=2000,
-    save_thumbnail=True,
+    wsi: str | Path,
+    store: str = None,
+    reader: Literal["openslide", "tiffslide", "bioformats"] = None,
+    attach_images: bool = False,
+    image_key: str = "wsi",
+    save_images: bool = False,
+    attach_thumbnail: bool = True,
+    thumbnail_key: str = "wsi_thumbnail",
+    thumbnail_size: int = 2000,
+    save_thumbnail: bool = True,
     **kwargs,
 ):
     """Open a whole slide image.
 
-    You can open a whole slide image from a URL or a local file.
-    If load from remote URL, the image will be downloaded and cached (default to current the working directory).
-    You can also attach images and thumbnail to the SpatialData object. By default, only the thumbnail is attached,
-    the thumbnail is a downsampled version of the whole slide image, the original image is not attached to save disk space
-    when you save the WSIData object on disk.
+    You can attach images and thumbnail to the SpatialData object. By default, only the thumbnail is attached,
+    the thumbnail is a downsampled version of the whole slide image,
+    the original image is not attached as it will make unnecessary copies of the data on disk
+    when saving the SpatialData object.
 
     Parameters
     ----------
@@ -45,8 +48,6 @@ def open_wsi(
         This is useful when you want to store all zarr files in a specific location.
     reader : str, optional
         Reader to use, by default "auto", choosing available reader, first openslide, then tifffile.
-    name : str, optional
-        The name of the slide.
     attach_images : bool, optional, default: False
         Whether to attach whole slide image to image slot in the spatial data object.
     image_key : str, optional
@@ -147,13 +148,13 @@ def open_wsi(
 
 
 def agg_wsi(
-    slides_table,
-    feature_key,
-    tile_key="tiles",
-    agg_key="agg",
-    wsi_col=None,
-    store_col=None,
-    error="raise",
+    slides_table: pd.DataFrame,
+    feature_key: str,
+    tile_key: str = "tiles",
+    agg_key: str = "agg",
+    wsi_col: str = None,
+    store_col: str = None,
+    error: Literal["raise", "skip"] = "raise",
 ):
     """
     Aggregate feature from a whole slide image.
@@ -212,7 +213,6 @@ def agg_wsi(
         mask = np.asarray([r is not None for r in results])
         X = np.vstack(np.asarray(results, dtype=object)[mask])
         slides_table = slides_table[mask]
-
     else:
         X = np.vstack(results)