Merge branch 'dev' into rly-patch-1

.github/workflows/run_coverage.yml

@@ -56,7 +56,7 @@ jobs:
       - name: Install package with optional dependencies
         if: ${{ matrix.opt_req }}
         run: |
-          python -m pip install ".[test,tqdm,zarr,termset]"
+          python -m pip install ".[test,tqdm,sparse,zarr,termset]"
 
       - name: Run tests and generate coverage report
         run: |

.readthedocs.yaml

@@ -24,7 +24,7 @@ formats: all
 # Optionally set the version of Python and requirements required to build your docs
 python:
   install:
-    - path: .[docs,tqdm,zarr,termset] # path to the package relative to the root
+    - path: .[docs,tqdm,sparse,zarr,termset] # path to the package relative to the root
 
 # Optionally include all submodules
 submodules:

CHANGELOG.md

@@ -7,6 +7,7 @@
 - Importing from `hdmf.build.map` is no longer supported. Import from `hdmf.build` instead. @rly [#1221](https://github.com/hdmf-dev/hdmf/pull/1221)
 - Python 3.8 has reached end of life. Dropped support for Python 3.8 and add support for Python 3.13. @mavaylon1 [#1209](https://github.com/hdmf-dev/hdmf/pull/1209)
 - Support for Zarr is limited to versions < 3. @rly [#1229](https://github.com/hdmf-dev/hdmf/pull/1229)
+- Scipy is no longer a required dependency. Users using the `CSRMatrix` data type should install `scipy` separately or with `pip install "hdmf[sparse]"`. @rly [#1140](https://github.com/hdmf-dev/hdmf/pull/1140)
 
 ### Changed
 - Added checks to ensure that group and dataset spec names and default names do not contain slashes. @bendichter [#1219](https://github.com/hdmf-dev/hdmf/pull/1219)

pyproject.toml

@@ -35,13 +35,13 @@ dependencies = [
     'numpy>=1.19.3',
     "pandas>=1.2.0",
     "ruamel.yaml>=0.16",
-    "scipy>=1.7",
 ]
 dynamic = ["version"]
 
 [project.optional-dependencies]
 tqdm = ["tqdm>=4.41.0"]
 zarr = ["zarr>=2.12.0,<3"]
+sparse = ["scipy>=1.7"]
 termset = [
     "linkml-runtime>=1.5.5",
     "schemasheets>=0.1.23",
@@ -70,7 +70,7 @@ docs = [
 ]
 
 # all possible dependencies
-all = ["hdmf[tqdm,zarr,termset,test,docs]"]
+all = ["hdmf[tqdm,zarr,sparse,termset,test,docs]"]
 
 [project.urls]
 "Homepage" = "https://github.com/hdmf-dev/hdmf"

src/hdmf/common/sparse.py

@@ -1,4 +1,11 @@
-import scipy.sparse as sps
+try:
+    from scipy.sparse import csr_matrix
+    SCIPY_INSTALLED = True
+except ImportError:
+    SCIPY_INSTALLED = False
+    class csr_matrix:  # dummy class to prevent import errors
+        pass
+
 from . import register_class
 from ..container import Container
 from ..utils import docval, popargs, to_uint_array,  get_data_shape, AllowPositional
@@ -7,7 +14,7 @@
 @register_class('CSRMatrix')
 class CSRMatrix(Container):
 
-    @docval({'name': 'data', 'type': (sps.csr_matrix, 'array_data'),
+    @docval({'name': 'data', 'type': (csr_matrix, 'array_data'),
              'doc': 'the data to use for this CSRMatrix or CSR data array.'
                     'If passing CSR data array, *indices*, *indptr*, and *shape* must also be provided'},
             {'name': 'indices', 'type': 'array_data', 'doc': 'CSR index array', 'default': None},
@@ -16,13 +23,17 @@ class CSRMatrix(Container):
             {'name': 'name', 'type': str, 'doc': 'the name to use for this when storing', 'default': 'csr_matrix'},
             allow_positional=AllowPositional.WARNING)
     def __init__(self, **kwargs):
+        if not SCIPY_INSTALLED:
+            raise ImportError(
+                "scipy must be installed to use CSRMatrix. Please install scipy using `pip install scipy`."
+            )
         data, indices, indptr, shape = popargs('data', 'indices', 'indptr', 'shape', kwargs)
         super().__init__(**kwargs)
-        if not isinstance(data, sps.csr_matrix):
+        if not isinstance(data, csr_matrix):
             temp_shape = get_data_shape(data)
             temp_ndim = len(temp_shape)
             if temp_ndim == 2:
-                data = sps.csr_matrix(data)
+                data = csr_matrix(data)
             elif temp_ndim == 1:
                 if any(_ is None for _ in (indptr, indices, shape)):
                     raise ValueError("Must specify 'indptr', 'indices', and 'shape' arguments when passing data array.")
@@ -31,9 +42,10 @@ def __init__(self, **kwargs):
                 shape = self.__check_arr(shape, 'shape')
                 if len(shape) != 2:
                     raise ValueError("'shape' argument must specify two and only two dimensions.")
-                data = sps.csr_matrix((data, indices, indptr), shape=shape)
+                data = csr_matrix((data, indices, indptr), shape=shape)
             else:
                 raise ValueError("'data' argument cannot be ndarray of dimensionality > 2.")
+        # self.__data is a scipy.sparse.csr_matrix
         self.__data = data
 
     @staticmethod

tests/unit/common/test_sparse.py

@@ -1,10 +1,25 @@
 from hdmf.common import CSRMatrix
 from hdmf.testing import TestCase, H5RoundTripMixin
-
-import scipy.sparse as sps
 import numpy as np
+import unittest
+
+try:
+    import scipy.sparse as sps
+    SCIPY_INSTALLED = True
+except ImportError:
+    SCIPY_INSTALLED = False
+
+
+@unittest.skipIf(SCIPY_INSTALLED, "scipy is installed")
+class TestCSRMatrixNoScipy(TestCase):
 
+    def test_import_error(self):
+        data = np.array([[1, 0, 2], [0, 0, 3], [4, 5, 6]])
+        with self.assertRaises(ImportError):
+            CSRMatrix(data=data)
 
+
+@unittest.skipIf(not SCIPY_INSTALLED, "scipy is not installed")
 class TestCSRMatrix(TestCase):
 
     def test_from_sparse_matrix(self):
@@ -153,7 +168,7 @@ def test_array_bad_dim(self):
         with self.assertRaisesWith(ValueError, msg):
             CSRMatrix(data=data, indices=indices, indptr=indptr, shape=shape)
 
-
+@unittest.skipIf(not SCIPY_INSTALLED, "scipy is not installed")
 class TestCSRMatrixRoundTrip(H5RoundTripMixin, TestCase):
 
     def setUpContainer(self):
@@ -164,6 +179,7 @@ def setUpContainer(self):
         return CSRMatrix(data=data, indices=indices, indptr=indptr, shape=shape)
 
 
+@unittest.skipIf(not SCIPY_INSTALLED, "scipy is not installed")
 class TestCSRMatrixRoundTripFromLists(H5RoundTripMixin, TestCase):
     """Test that CSRMatrix works with lists as well"""
 

tox.ini

@@ -29,7 +29,7 @@ extras =
     # which optional dependency set(s) to use (default: none)
     pytest:    test
     gallery:   doc
-    optional:  tqdm,zarr,termset
+    optional:  tqdm,sparse,zarr,termset
 commands =
     # commands to run for every environment
     python --version     # print python version for debugging