Merge branch 'main' into ig/fix_set_index

benchmarks/benchmarks/readwrite.py

@@ -127,8 +127,9 @@ def track_read_full_memratio(self, *_):
     def peakmem_read_backed(self, *_):
         anndata.read_h5ad(self.filepath, backed="r")
 
-    def mem_read_backed_object(self, *_):
-        return anndata.read_h5ad(self.filepath, backed="r")
+    # causes benchmarking to break from: https://github.com/pympler/pympler/issues/151
+    # def mem_read_backed_object(self, *_):
+    #     return anndata.read_h5ad(self.filepath, backed="r")
 
 
 class H5ADWriteSuite:

docs/conf.py

@@ -4,7 +4,7 @@
 from datetime import datetime
 from functools import partial
 from importlib import metadata
-from pathlib import Path
+from pathlib import Path, PurePosixPath
 from typing import TYPE_CHECKING
 
 from docutils import nodes
@@ -60,6 +60,7 @@
     "sphinx.ext.linkcode",
     "nbsphinx",
     "IPython.sphinxext.ipython_console_highlighting",
+    "patch_sphinx_toolbox_autoprotocol",
     "sphinx_toolbox.more_autodoc.autoprotocol",
 ]
 myst_enable_extensions = [
@@ -71,6 +72,7 @@
 autosummary_generate = True
 autodoc_member_order = "bysource"
 issues_github_path = "scverse/anndata"
+rtd_links_prefix = PurePosixPath("src")
 # autodoc_default_flags = ['members']
 napoleon_google_docstring = False
 napoleon_numpy_docstring = True
@@ -190,40 +192,3 @@ def setup(app: Sphinx):
         "Miscellaneous",
     )
 ]
-
-
-def patch_sphinx_toolbox_autoprotocol():
-    """Compat hack: https://github.com/sphinx-toolbox/sphinx-toolbox/issues/168"""
-
-    from sphinx.ext.autodoc import ObjectMember
-    from sphinx_toolbox.more_autodoc.autoprotocol import ProtocolDocumenter
-
-    if TYPE_CHECKING:
-        from typing import Self
-
-    class ObjectMemberCompat(ObjectMember):
-        @classmethod
-        def from_other(cls, other: ObjectMember) -> Self:
-            return cls(
-                other.__name__,
-                other.object,
-                docstring=other.docstring,
-                class_=other.class_,
-                skipped=other.skipped,
-            )
-
-        def __iter__(self):
-            return iter([self.__name__, self.object])
-
-    filter_orig = ProtocolDocumenter.filter_members
-
-    def filter_members(
-        self, members: list[ObjectMember], want_all: bool
-    ) -> list[tuple[str, object, bool]]:
-        member_tuples = [ObjectMemberCompat.from_other(m) for m in members]
-        return filter_orig(self, member_tuples, want_all)
-
-    ProtocolDocumenter.filter_members = filter_members
-
-
-patch_sphinx_toolbox_autoprotocol()

docs/extensions/patch_sphinx_toolbox_autoprotocol.py

@@ -0,0 +1,43 @@
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+from sphinx.ext.autodoc import ObjectMember
+from sphinx_toolbox.more_autodoc.autoprotocol import ProtocolDocumenter
+
+if TYPE_CHECKING:
+    from typing import Self
+
+    from sphinx.application import Sphinx
+
+
+def patch_sphinx_toolbox_autoprotocol():
+    """Compat hack: https://github.com/sphinx-toolbox/sphinx-toolbox/issues/168"""
+
+    class ObjectMemberCompat(ObjectMember):
+        @classmethod
+        def from_other(cls, other: ObjectMember) -> Self:
+            return cls(
+                other.__name__,
+                other.object,
+                docstring=other.docstring,
+                class_=other.class_,
+                skipped=other.skipped,
+            )
+
+        def __iter__(self):
+            return iter([self.__name__, self.object])
+
+    filter_orig = ProtocolDocumenter.filter_members
+
+    def filter_members(
+        self, members: list[ObjectMember], want_all: bool
+    ) -> list[tuple[str, object, bool]]:
+        member_tuples = [ObjectMemberCompat.from_other(m) for m in members]
+        return filter_orig(self, member_tuples, want_all)
+
+    ProtocolDocumenter.filter_members = filter_members
+
+
+def setup(_app: Sphinx) -> None:
+    patch_sphinx_toolbox_autoprotocol()

docs/release-notes/0.10.9.md

@@ -6,9 +6,9 @@
 * Add warning for setting `X` on a view with repeated indices {pr}`1501` {user}`ilan-gold`
 * Coerce {class}`numpy.matrix` classes to arrays when trying to store them in `AnnData` {pr}`1516` {user}`flying-sheep`
 * Fix for setting a dense `X` view with a sparse matrix {pr}`1532` {user}`ilan-gold`
-* Upper bound {mod}`numpy` for `gpu` installation on account of https://github.com/cupy/cupy/issues/8391 {pr}`1540` {user}`ilan-gold`
+* Upper bound {mod}`numpy` for `gpu` installation on account of {issue}`cupy/cupy#8391` {pr}`1540` {user}`ilan-gold`
 * Fix writing large number of columns for `h5` files {pr}`1147` {user}`ilan-gold` {user}`selmanozleyen`
-* Upper bound dask on account of https://github.com/scverse/anndata/issues/1579 {pr}`1580` {user}`ilan-gold`
+* Upper bound dask on account of {issue}`1579` {pr}`1580` {user}`ilan-gold`
 * Ensure setting {attr}`pandas.DataFrame.index` on a view of a {class}`~anndata.AnnData` instantiates the {class}`~pandas.DataFrame` from the view {pr}`1586` {user}`ilan-gold`
 
 #### Documentation

docs/release-notes/0.11.0.md

@@ -5,11 +5,12 @@
 
 * Allow `axis` parameter of e.g. :func:`anndata.concat` to accept `'obs'` and `'var` {pr}`1244` {user}`flying-sheep`
 * Add `settings` object with methods for altering internally-used options, like checking for uniqueness on `obs`' index {pr}`1270` {user}`ilan-gold`
-* Add `should_remove_unused_categories` option to `anndata.settings` to override current behavior.  Default is `True` (i.e., previous behavior).  Please refer to the [documentation](https://anndata.readthedocs.io/en/latest/generated/anndata.settings.html) for usage.  {pr}`1340` {user}`ilan-gold`
 * `scipy.sparse.csr_array` and `scipy.sparse.csc_array` are now supported when constructing `AnnData` objects {pr}`1028` {user}`ilan-gold` {user}`isaac-virshup`
-* Add `should_check_uniqueness` option to `anndata.settings` to override current behavior.  Default is `True` (i.e., previous behavior).  Please refer to the [documentation](https://anndata.readthedocs.io/en/latest/generated/anndata.settings.html) for usage.  {pr}`1507` {user}`ilan-gold`
+* Add {attr}`~anndata.settings.should_remove_unused_categories` option to {attr}`anndata.settings` to override current behavior {pr}`1340` {user}`ilan-gold`
+* Add {attr}`~anndata.settings.should_check_uniqueness` option to {attr}`anndata.settings` to override current behavior {pr}`1507` {user}`ilan-gold`
 * Add :func:`~anndata.experimental.read_elem_as_dask` function to handle i/o with sparse and dense arrays {pr}`1469` {user}`ilan-gold`
 * Add functionality to write from GPU {class}`dask.array.Array` to disk {pr}`1550` {user}`ilan-gold`
+* Add ability to convert strings to categoricals on write in {meth}`~anndata.AnnData.write_h5ad` and {meth}`~anndata.AnnData.write_zarr` via `convert_strings_to_categoricals` parameter {pr}`1474` {user}` falexwolf`
 
 #### Bugfix
 

src/anndata/_io/h5ad.py

@@ -47,7 +47,7 @@ def write_h5ad(
     adata: AnnData,
     *,
     as_dense: Sequence[str] = (),
-    strings_to_categoricals: bool = True,
+    convert_strings_to_categoricals: bool = True,
     dataset_kwargs: Mapping[str, Any] = MappingProxyType({}),
     **kwargs,
 ) -> None:
@@ -63,7 +63,7 @@ def write_h5ad(
     if "raw/X" in as_dense and adata.raw is None:
         raise ValueError("Cannot specify writing `raw/X` to dense if it doesn’t exist.")
 
-    if strings_to_categoricals:
+    if convert_strings_to_categoricals:
         adata.strings_to_categoricals()
         if adata.raw is not None:
             adata.strings_to_categoricals(adata.raw.var)

src/anndata/_io/zarr.py

@@ -28,12 +28,12 @@ def write_zarr(
     adata: AnnData,
     *,
     chunks: tuple[int, ...] | None = None,
-    strings_to_categoricals: bool = True,
+    convert_strings_to_categoricals: bool = True,
     **ds_kwargs,
 ) -> None:
     if isinstance(store, Path):
         store = str(store)
-    if strings_to_categoricals:
+    if convert_strings_to_categoricals:
         adata.strings_to_categoricals()
         if adata.raw is not None:
             adata.strings_to_categoricals(adata.raw.var)