Fix int csr_matrix

.github/dependabot.yml

@@ -1,6 +1,6 @@
 version: 2
 updates:
-  - package-ecosystem: "github-actions"
+  - package-ecosystem: github-actions
     directory: /
     schedule:
       interval: weekly

.github/workflows/run_tests.yml

@@ -2,11 +2,8 @@ name: Unit Tests
 
 on:
   push:
-    branches:
-      - main
+    branches: [main]
   pull_request:
-    branches:
-      - "*"
 
 concurrency:
   group: ${{ github.workflow }}-${{ github.ref }}
@@ -47,14 +44,12 @@ jobs:
           packages: pandoc gfortran libblas-dev liblapack-dev libedit-dev llvm-dev libcurl4-openssl-dev ffmpeg libhdf5-dev
           version: 1.0
 
-      - name: Set up Python
-        uses: actions/setup-python@v5
+      - uses: actions/setup-python@v5
         with:
           python-version: ${{ matrix.config.python }}
           cache: pip
 
-      - name: Set up R
-        id: setup-r
+      - id: setup-r
         uses: r-lib/actions/setup-r@v2
         with:
           r-version: ${{ matrix.config.r }}

src/anndata2ri/_r2py.py

@@ -11,7 +11,7 @@
 
 from . import _conv_name
 from ._conv import converter, mat_rpy2py
-from ._rpy2_ext import importr
+from ._rpy2_ext import R_INT_BYTES, importr
 from .scipy2ri import supported_r_matrix_classes
 from .scipy2ri._r2py import rmat_to_spmat
 
@@ -23,9 +23,6 @@
     from scipy.sparse import spmatrix
 
 
-R_INT_BYTES = 4
-
-
 @converter.rpy2py.register(SexpS4)
 def rpy2py_s4(obj: SexpS4) -> pd.DataFrame | AnnData | None:
     """Convert known S4 class instance to Python object.

src/anndata2ri/_rpy2_ext.py

@@ -1,16 +1,19 @@
 from __future__ import annotations
 
-from functools import lru_cache
+from functools import cache
 
 from rpy2.robjects import Environment, packages
 
 
-@lru_cache
+R_INT_BYTES = 4
+
+
+@cache
 def importr(name: str) -> packages.Package:
     return packages.importr(name)
 
 
-@lru_cache
+@cache
 def data(package: str, name: str | None = None) -> packages.PackageData | Environment:
     if name is None:
         return packages.data(importr(package))

src/anndata2ri/scipy2ri/_py2r.py

@@ -1,16 +1,16 @@
 from __future__ import annotations
 
-from functools import lru_cache, wraps
+from functools import cache
 from importlib.resources import files
 from typing import TYPE_CHECKING
 
 import numpy as np
 from rpy2.robjects import default_converter, numpy2ri
 from rpy2.robjects.conversion import localconverter
-from rpy2.robjects.packages import Package, SignatureTranslatedAnonymousPackage
+from rpy2.robjects.packages import InstalledSTPackage, SignatureTranslatedAnonymousPackage
 from scipy import sparse
 
-from anndata2ri._rpy2_ext import importr
+from anndata2ri._rpy2_ext import R_INT_BYTES, importr
 
 from ._conv import converter
 
@@ -21,55 +21,46 @@
     from rpy2.rinterface import Sexp
 
 
-matrix: SignatureTranslatedAnonymousPackage | None = None
-base: Package | None = None
+@cache
+def baseenv() -> InstalledSTPackage:
+    with localconverter(default_converter):
+        return importr('base')
 
 
-@lru_cache
-def get_r_code() -> str:
-    return files('anndata2ri').joinpath('scipy2ri', '_py2r_helpers.r').read_text()
+@cache
+def matrixenv() -> SignatureTranslatedAnonymousPackage:
+    with localconverter(default_converter):
+        importr('Matrix')  # make class available
+        r_code = files('anndata2ri').joinpath('scipy2ri', '_py2r_helpers.r').read_text()
+        return SignatureTranslatedAnonymousPackage(r_code, 'matrix')
 
 
 def get_type_conv(dtype: np.dtype) -> Callable[[np.ndarray], Sexp]:
-    global base  # noqa: PLW0603
-    if base is None:
-        base = importr('base')
+    base = baseenv()
     if np.issubdtype(dtype, np.floating):
         return base.as_double
+    if np.issubdtype(dtype, np.integer):
+        if dtype.itemsize <= R_INT_BYTES:
+            return base.as_integer
+        return base.as_numeric  # maybe uses R_xlen_t?
     if np.issubdtype(dtype, np.bool_):
         return base.as_logical
     msg = f'Unknown dtype {dtype!r} cannot be converted to ?gRMatrix.'
     raise ValueError(msg)
 
 
-def py2r_context(f: Callable[[sparse.spmatrix], Sexp]) -> Callable[[sparse.spmatrix], Sexp]:
-    """R globalenv context with some helper functions."""
-
-    @wraps(f)
-    def wrapper(obj: sparse.spmatrix) -> Sexp:
-        global matrix  # noqa: PLW0603
-        if matrix is None:
-            importr('Matrix')  # make class available
-            r_code = get_r_code()
-            matrix = SignatureTranslatedAnonymousPackage(r_code, 'matrix')
-
-        return f(obj)
-
-    return wrapper
-
-
 @converter.py2rpy.register(sparse.csc_matrix)
-@py2r_context
 def csc_to_rmat(csc: sparse.csc_matrix) -> Sexp:
+    matrix = matrixenv()
     csc.sort_indices()
     conv_data = get_type_conv(csc.dtype)
     with localconverter(default_converter + numpy2ri.converter):
         return matrix.from_csc(i=csc.indices, p=csc.indptr, x=csc.data, dims=list(csc.shape), conv_data=conv_data)
 
 
 @converter.py2rpy.register(sparse.csr_matrix)
-@py2r_context
 def csr_to_rmat(csr: sparse.csr_matrix) -> Sexp:
+    matrix = matrixenv()
     csr.sort_indices()
     conv_data = get_type_conv(csr.dtype)
     with localconverter(default_converter + numpy2ri.converter):
@@ -83,8 +74,8 @@ def csr_to_rmat(csr: sparse.csr_matrix) -> Sexp:
 
 
 @converter.py2rpy.register(sparse.coo_matrix)
-@py2r_context
 def coo_to_rmat(coo: sparse.coo_matrix) -> Sexp:
+    matrix = matrixenv()
     conv_data = get_type_conv(coo.dtype)
     with localconverter(default_converter + numpy2ri.converter):
         return matrix.from_coo(
@@ -97,8 +88,8 @@ def coo_to_rmat(coo: sparse.coo_matrix) -> Sexp:
 
 
 @converter.py2rpy.register(sparse.dia_matrix)
-@py2r_context
 def dia_to_rmat(dia: sparse.dia_matrix) -> Sexp:
+    matrix = matrixenv()
     conv_data = get_type_conv(dia.dtype)
     if len(dia.offsets) > 1:
         msg = (

tests/test_py2rpy.py

@@ -1,7 +1,7 @@
 from __future__ import annotations
 
 from typing import TYPE_CHECKING
-from warnings import catch_warnings, simplefilter
+from warnings import catch_warnings, filterwarnings, simplefilter
 
 import numpy as np
 import pytest
@@ -10,6 +10,7 @@
 from pandas import DataFrame
 from rpy2.robjects import baseenv, globalenv
 from rpy2.robjects.conversion import localconverter
+from scipy import sparse
 
 import anndata2ri
 from anndata2ri._rpy2_ext import importr
@@ -31,6 +32,26 @@ def mk_ad_simple() -> AnnData:
     )
 
 
+@pytest.mark.parametrize('dtype', [np.float32, np.float64, np.int32, np.int64])
+@pytest.mark.parametrize('mat_type', [np.asarray, sparse.csr_matrix])
+def test_simple(
+    py2r: Py2R,
+    dtype: np.dtype,
+    mat_type: Callable[[np.ndarray, np.dtype], np.ndarray | sparse.spmatrix],
+) -> None:
+    data = mk_ad_simple()
+    if data.X is not None:
+        data.X = mat_type(data.X, dtype=dtype)
+    ex = py2r(anndata2ri, data)
+    assert tuple(baseenv['dim'](ex)[::-1]) == data.shape
+
+
+def krumsiek() -> AnnData:
+    adata = sc.datasets.krumsiek11()
+    adata.obs_names_make_unique()
+    return adata
+
+
 def check_empty(_: Sexp) -> None:
     pass
 
@@ -45,33 +66,31 @@ def check_pca(ex: Sexp) -> None:
 datasets = [
     pytest.param(check_empty, (0, 0), AnnData, id='empty'),
     pytest.param(check_pca, (2, 3), mk_ad_simple, id='simple'),
-    pytest.param(check_empty, (640, 11), sc.datasets.krumsiek11, id='krumsiek'),
+    pytest.param(check_empty, (640, 11), krumsiek, id='krumsiek'),
 ]
 
 
 @pytest.mark.parametrize(('check', 'shape', 'dataset'), datasets)
-def test_py2rpy(
+def test_datasets(
     py2r: Py2R,
     check: Callable[[Sexp], None],
     shape: tuple[int, ...],
     dataset: Callable[[], AnnData],
 ) -> None:
-    if dataset is sc.datasets.krumsiek11:
-        with (
-            pytest.warns(UserWarning, match=r'Duplicated obs_names'),
-            pytest.warns(UserWarning, match=r'Observation names are not unique'),
-            # TODO(flying-sheep): Adapt to rpy2 changes instead
-            # https://github.com/theislab/anndata2ri/issues/109
-            pytest.warns(DeprecationWarning, match=r'rpy2\.robjects\.conversion is deprecated'),
-        ):
+    if dataset is krumsiek:
+        # TODO(flying-sheep): Adapt to rpy2 changes instead
+        # https://github.com/theislab/anndata2ri/issues/109
+        with pytest.warns(DeprecationWarning, match=r'rpy2\.robjects\.conversion is deprecated'):
+            filterwarnings('ignore', r'Duplicated obs_names', UserWarning)
+            filterwarnings('ignore', r'Observation names are not unique', UserWarning)
             ex = py2r(anndata2ri, dataset())
     else:
         ex = py2r(anndata2ri, dataset())
     assert tuple(baseenv['dim'](ex)[::-1]) == shape
     check(ex)
 
 
-def test_py2rpy2_numpy_pbmc68k() -> None:
+def test_numpy_pbmc68k() -> None:
     """Not tested above as the pbmc68k dataset has some weird metadata."""
     from scanpy.datasets import pbmc68k_reduced
 

tests/test_rpy2py.py

@@ -69,7 +69,7 @@ def check_example(adata: AnnData) -> None:
 
 
 @pytest.mark.parametrize(('check', 'shape', 'dataset'), expression_sets)
-def test_convert_manual(
+def test_manual(
     r2py: R2Py,
     check: Callable[[AnnData], None],
     shape: tuple[int, ...],
@@ -81,15 +81,15 @@ def test_convert_manual(
     check(ad)
 
 
-def test_convert_empty_df_with_rows(r2py: R2Py) -> None:
+def test_empty_df_with_rows(r2py: R2Py) -> None:
     df = r('S4Vectors::DataFrame(a=1:10)[, -1]')
     assert df.slots['nrows'][0] == 10
 
     df_py = r2py(anndata2ri, lambda: conversion.get_conversion().rpy2py(df))
     assert isinstance(df_py, pd.DataFrame)
 
 
-def test_convert_factor(r2py: R2Py) -> None:
+def test_factor(r2py: R2Py) -> None:
     code = """
     SingleCellExperiment::SingleCellExperiment(
         assays = list(counts = matrix(rpois(6*4, 5), ncol=4)),

tests/test_scipy_py2rpy.py

@@ -27,7 +27,7 @@
 
 @pytest.mark.parametrize('typ', ['l', 'd'])
 @pytest.mark.parametrize(('shape', 'dataset', 'cls'), mats)
-def test_py2rpy(
+def test_mats(
     py2r: Py2R,
     typ: Literal['l', 'd'],
     shape: tuple[int, ...],

tests/test_scipy_rpy2py.py

@@ -63,7 +63,7 @@
 
 
 @pytest.mark.parametrize(('shape', 'cls', 'dtype', 'arr', 'dataset'), mats)
-def test_py2rpy(
+def test_mats(
     r2py: R2Py,
     shape: tuple[int, int],
     cls: type[sparse.spmatrix],