Simplified layer code, improved tests

anndata/base.py

@@ -1,12 +1,11 @@
 """Main class and helper functions.
 """
-import os, sys
-import warnings
+import os
+import sys
 import logging as logg
 from enum import Enum
 from collections import OrderedDict
 from functools import reduce
-from pathlib import Path
 from textwrap import indent, dedent
 from typing import Union, Optional, Any, Iterable, Mapping, Sequence, Sized, Tuple, List
 from copy import deepcopy
@@ -26,6 +25,7 @@
     from zarr.core import Array as ZarrArray
 except ImportError:
     class ZarrArray:
+        @staticmethod
         def __rep__():
             return 'mock zarr.core.Array'
 
@@ -68,6 +68,9 @@ def classes(cls):
         return tuple(c.value for c in cls.__members__.values())
 
 
+Index = Union[slice, int, np.int64, np.ndarray, Sized]
+
+
 class BoundRecArr(np.recarray):
     """A `np.recarray` to which fields can be added using `.['key']`.
 
@@ -652,7 +655,7 @@ def __init__(
         filename: Optional[PathLike] = None,
         filemode: Optional[str] = None,
         asview: bool = False,
-        *, oidx=None, vidx=None):
+        *, oidx: Index = None, vidx: Index = None):
         if asview:
             if not isinstance(X, AnnData):
                 raise ValueError('`X` has to be an AnnData object.')
@@ -665,8 +668,8 @@ def __init__(
                 dtype=dtype, shape=shape,
                 filename=filename, filemode=filemode)
 
-    def _init_as_view(self, adata_ref, oidx, vidx):
-        def get_n_items_idx(idx):
+    def _init_as_view(self, adata_ref: 'AnnData', oidx: Index, vidx: Index):
+        def get_n_items_idx(idx: Index):
             if isinstance(idx, np.ndarray) and idx.dtype == bool:
                 return idx.sum()
             else:

anndata/layers.py

@@ -1,12 +1,23 @@
-#thats just start
+from typing import Mapping, Optional, Union
+
 import numpy as np
 from collections import OrderedDict
 from scipy.sparse import issparse
 
-class AnnDataLayers():
+if False:
+    from .base import AnnData, Index  # noqa
 
-    def __init__(self, adata, layers=None, dtype='float32', adata_ref=None, oidx=None, vidx=None):
 
+class AnnDataLayers:
+    def __init__(
+        self,
+        adata: 'AnnData',
+        layers: Optional[Mapping[str, np.ndarray]] = None,
+        dtype: Union[str, np.dtype] = 'float32',
+        adata_ref: 'AnnData' = None,
+        oidx: 'Index' = None,
+        vidx: 'Index' = None,
+    ):
         self._adata = adata
         self._adata_ref = adata_ref
         self._oidx = oidx
@@ -63,29 +74,21 @@ def __delattr__(self, key):
     def keys(self):
         if self.isview:
             return self._adata_ref.layers.keys()
-        else:
+        else:  # TODO @Koncopd: Why wrap this in list() and not the above?
             return list(self._layers.keys())
 
     def items(self, copy=True):
         if self.isview:
-            if copy:
-                return [(k, v[self._oidx, self._vidx].copy()) for (k, v) in self._adata_ref.layers.items()]
-            else:
-                return [(k, v[self._oidx, self._vidx]) for (k, v) in self._adata_ref.layers.items()]
+            pairs = [(k, v[self._oidx, self._vidx]) for (k, v) in self._adata_ref.layers.items()]
         else:
-            if copy:
-                return [(k, v.copy()) for (k, v) in self._layers.items()]
-            else:
-                return self._layers.items()
+            pairs = self._layers.items()
+        return [(k, v.copy()) for (k, v) in pairs] if copy else pairs
 
     def as_dict(self, copy=True):
-        return {k:v for (k, v) in self.items(copy)}
+        return dict(self.items(copy))
 
     def __len__(self):
-        if self.isview:
-            return len(self._adata_ref.layers)
-        else:
-            return len(self._layers)
+        return len(self._adata_ref.layers if self.isview else self._layers)
 
     @property
     def isview(self):

anndata/tests/readwrite.py

@@ -39,59 +39,60 @@
 # -------------------------------------------------------------------------------
 
 
-def test_readwrite_h5ad():
-    for typ in [np.array, csr_matrix]:
-        X = typ(X_list)
-        adata = ad.AnnData(X, obs=obs_dict, var=var_dict, uns=uns_dict)
-        assert pd.api.types.is_string_dtype(adata.obs['oanno1'])
-        adata.raw = adata
-        adata.write('./test.h5ad')
-        adata = ad.read('./test.h5ad')
-        assert pd.api.types.is_categorical(adata.obs['oanno1'])
-        assert pd.api.types.is_string_dtype(adata.obs['oanno2'])
-        assert adata.obs.index.tolist() == ['name1', 'name2', 'name3']
-        assert adata.obs['oanno1'].cat.categories.tolist() == ['cat1', 'cat2']
-        assert pd.api.types.is_categorical(adata.raw.var['vanno2'])
-
-
-def test_readwrite_dynamic():
-    for typ in [np.array, csr_matrix]:
-        X = typ(X_list)
-        adata = ad.AnnData(X, obs=obs_dict, var=var_dict, uns=uns_dict)
-        adata.filename = './test.h5ad'  # change to backed mode
-        adata.write()
-        adata = ad.read('./test.h5ad')
-        assert pd.api.types.is_categorical(adata.obs['oanno1'])
-        assert pd.api.types.is_string_dtype(adata.obs['oanno2'])
-        assert adata.obs.index.tolist() == ['name1', 'name2', 'name3']
-        assert adata.obs['oanno1'].cat.categories.tolist() == ['cat1', 'cat2']
-
-
-def test_readwrite_zarr():
-    for typ in [np.array, csr_matrix]:
-        X = typ(X_list)
-        adata = ad.AnnData(X, obs=obs_dict, var=var_dict, uns=uns_dict)
-        assert pd.api.types.is_string_dtype(adata.obs['oanno1'])
-        adata.write_zarr('./test_zarr_dir', chunks=True)
-        adata = ad.read_zarr('./test_zarr_dir')
-        assert pd.api.types.is_categorical(adata.obs['oanno1'])
-        assert pd.api.types.is_string_dtype(adata.obs['oanno2'])
-        assert adata.obs.index.tolist() == ['name1', 'name2', 'name3']
-        assert adata.obs['oanno1'].cat.categories.tolist() == ['cat1', 'cat2']
+@pytest.mark.parametrize('typ', [np.array, csr_matrix])
+def test_readwrite_h5ad(typ):
+    X = typ(X_list)
+    adata = ad.AnnData(X, obs=obs_dict, var=var_dict, uns=uns_dict)
+    assert pd.api.types.is_string_dtype(adata.obs['oanno1'])
+    adata.raw = adata
+    adata.write('./test.h5ad')
+    adata = ad.read('./test.h5ad')
+    assert pd.api.types.is_categorical(adata.obs['oanno1'])
+    assert pd.api.types.is_string_dtype(adata.obs['oanno2'])
+    assert adata.obs.index.tolist() == ['name1', 'name2', 'name3']
+    assert adata.obs['oanno1'].cat.categories.tolist() == ['cat1', 'cat2']
+    assert pd.api.types.is_categorical(adata.raw.var['vanno2'])
+
+
+@pytest.mark.parametrize('typ', [np.array, csr_matrix])
+def test_readwrite_dynamic(typ):
+    X = typ(X_list)
+    adata = ad.AnnData(X, obs=obs_dict, var=var_dict, uns=uns_dict)
+    adata.filename = './test.h5ad'  # change to backed mode
+    adata.write()
+    adata = ad.read('./test.h5ad')
+    assert pd.api.types.is_categorical(adata.obs['oanno1'])
+    assert pd.api.types.is_string_dtype(adata.obs['oanno2'])
+    assert adata.obs.index.tolist() == ['name1', 'name2', 'name3']
+    assert adata.obs['oanno1'].cat.categories.tolist() == ['cat1', 'cat2']
+
+
+@pytest.mark.skipif(not find_spec('zarr'), reason='Zarr is not installed')
+@pytest.mark.parametrize('typ', [np.array, csr_matrix])
+def test_readwrite_zarr(typ):
+    X = typ(X_list)
+    adata = ad.AnnData(X, obs=obs_dict, var=var_dict, uns=uns_dict)
+    assert pd.api.types.is_string_dtype(adata.obs['oanno1'])
+    adata.write_zarr('./test_zarr_dir', chunks=True)
+    adata = ad.read_zarr('./test_zarr_dir')
+    assert pd.api.types.is_categorical(adata.obs['oanno1'])
+    assert pd.api.types.is_string_dtype(adata.obs['oanno2'])
+    assert adata.obs.index.tolist() == ['name1', 'name2', 'name3']
+    assert adata.obs['oanno1'].cat.categories.tolist() == ['cat1', 'cat2']
 
 
 @pytest.mark.skipif(not find_spec('loompy'), reason='Loompy is not installed (expected on Python 3.5)')
-def test_readwrite_loom():
-    for i, typ in enumerate([np.array, csr_matrix]):
-        X = typ(X_list)
-        adata = ad.AnnData(X, obs=obs_dict, var=var_dict, uns=uns_dict)
-        adata.write_loom('./test.loom')
-        adata = ad.read_loom('./test.loom', sparse=(i == 1))
-        if isinstance(X, np.ndarray):
-            assert np.allclose(adata.X, X)
-        else:
-            # TODO: this should not be necessary
-            assert np.allclose(adata.X.toarray(), X.toarray())
+@pytest.mark.parametrize('typ', [np.array, csr_matrix])
+def test_readwrite_loom(typ):
+    X = typ(X_list)
+    adata = ad.AnnData(X, obs=obs_dict, var=var_dict, uns=uns_dict)
+    adata.write_loom('./test.loom')
+    adata = ad.read_loom('./test.loom', sparse=typ is csr_matrix)
+    if isinstance(X, np.ndarray):
+        assert np.allclose(adata.X, X)
+    else:
+        # TODO: this should not be necessary
+        assert np.allclose(adata.X.toarray(), X.toarray())
 
 
 def test_read_csv():
@@ -116,8 +117,8 @@ def test_read_tsv_iter():
         assert adata.X.tolist() == X_list
 
 
-def test_write_csv():
-    for typ in [np.array, csr_matrix]:
-        X = typ(X_list)
-        adata = ad.AnnData(X, obs=obs_dict, var=var_dict, uns=uns_dict)
-        adata.write_csvs('./test_csv_dir', skip_data=False)
+@pytest.mark.parametrize('typ', [np.array, csr_matrix])
+def test_write_csv(typ):
+    X = typ(X_list)
+    adata = ad.AnnData(X, obs=obs_dict, var=var_dict, uns=uns_dict)
+    adata.write_csvs('./test_csv_dir', skip_data=False)