Skip to content

Commit

Permalink
WIP: flake8 conformance
Browse files Browse the repository at this point in the history
  • Loading branch information
@tvandera
tvandera committed Jun 10, 2024
1 parent 688b1f5 commit 1abce2c
Show file tree
Hide file tree
Showing 2 changed files with 57 additions and 55 deletions.
64 changes: 33 additions & 31 deletions h5sparse/h5sparse.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -10,10 +10,11 @@
'coo': ss.coo_matrix,
}

indptr_dtype = np.int64
indptr_dtype = np.int64
indices_dtype = np.int64
row_dtype = np.int64
col_dtype = np.int64
row_dtype = np.int64
col_dtype = np.int64


def get_format_str(data):
for format_str, format_class in six.viewitems(FORMAT_DICT):
Expand All @@ -30,6 +31,7 @@ def get_format_class(format_str):
.format(format_str))
return format_class


def is_compressed_format(format_str):
return format_str in ('csc', 'csr')

Expand All @@ -52,27 +54,27 @@ def __getitem__(self, key):
raise ValueError("Unexpected item type.")

def create_dataset_compressed(self, name, sparse_format, shape, data, indices, indptr,
dtype, **kwargs):
dtype, **kwargs):
"""Create a dataset in csc or csr format"""
assert sparse_format in ("csc", "csr")

group = self.create_group(name)
group.attrs['h5sparse_format'] = sparse_format
group.attrs['h5sparse_shape'] = shape
group.create_dataset('data', data=data, dtype=dtype, **kwargs)
group.attrs['h5sparse_shape'] = shape
group.create_dataset('data', data=data, dtype=dtype, **kwargs)
group.create_dataset('indices', data=indices, dtype=indices_dtype, **kwargs)
group.create_dataset('indptr', data=indptr, dtype=indptr_dtype, **kwargs)
group.create_dataset('indptr', data=indptr, dtype=indptr_dtype, **kwargs)
return group

def create_dataset_coo(self, name, sparse_format, shape, data, row, col,
dtype, **kwargs):
dtype, **kwargs):
"""Create a dataset in csc or csr format"""
assert sparse_format == "coo"

group = self.create_group(name)
group.attrs['h5sparse_format'] = sparse_format
group.attrs['h5sparse_shape'] = shape
group.create_dataset('data', data=data, dtype=dtype, **kwargs)
group.attrs['h5sparse_shape'] = shape
group.create_dataset('data', data=data, dtype=dtype, **kwargs)
group.create_dataset('row', data=row, dtype=row_dtype, **kwargs)
group.create_dataset('col', data=col, dtype=col_dtype, **kwargs)
return group
Expand All @@ -81,13 +83,13 @@ def create_dataset_from_dataset(self, name, data, dtype, **kwargs):
sparse_format = data.attrs['h5sparse_format']
if (is_compressed_format(sparse_format)):
group = self.create_dataset_compressed(name,
data.attrs['h5sparse_format'],
data.attrs['h5sparse_shape'],
data.h5py_group['data'],
data.h5py_group['indices'],
data.h5py_group['indptr'],
dtype,
**kwargs)
data.attrs['h5sparse_format'],
data.attrs['h5sparse_shape'],
data.h5py_group['data'],
data.h5py_group['indices'],
data.h5py_group['indptr'],
dtype,
**kwargs)
else:
group = self.create_dataset_coo(name,
data.attrs['h5sparse_format'],
Expand All @@ -100,7 +102,7 @@ def create_dataset_from_dataset(self, name, data, dtype, **kwargs):
return group

def create_dataset_from_scipy(self, name, data, dtype, **kwargs):
sparse_format = get_format_str(data)
sparse_format = get_format_str(data)
if (is_compressed_format(sparse_format)):
group = self.create_dataset_compressed(name,
sparse_format,
Expand All @@ -112,13 +114,13 @@ def create_dataset_from_scipy(self, name, data, dtype, **kwargs):
**kwargs)
else:
group = self.create_dataset_coo(name,
sparse_format,
data.shape,
data.data,
data.row,
data.col,
dtype,
**kwargs)
sparse_format,
data.shape,
data.data,
data.row,
data.col,
dtype,
**kwargs)
return group

def create_dataset(self, name, shape=None, dtype=None, data=None,
Expand All @@ -137,9 +139,9 @@ def create_dataset(self, name, shape=None, dtype=None, data=None,
format_class = get_format_class(sparse_format)
data = format_class(data)
group = self.create_dataset_from_scipy(name,
data,
dtype,
**kwargs)
data,
dtype,
**kwargs)
elif data is None and sparse_format is not None:
format_class = get_format_class(sparse_format)
if dtype is None:
Expand All @@ -148,9 +150,9 @@ def create_dataset(self, name, shape=None, dtype=None, data=None,
shape = (0, 0)
data = format_class(shape, dtype=dtype)
group = self.create_dataset_from_scipy(name,
data,
dtype,
**kwargs)
data,
dtype,
**kwargs)
else:
# forward the arguments to h5py
assert sparse_format is None
Expand Down
48 changes: 24 additions & 24 deletions h5sparse/tests.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -8,10 +8,11 @@

import h5sparse


class AbstractTestH5Sparse():
def test_create_empty_sparse_dataset(self):
h5_path = mkstemp(suffix=".h5")[1]
format_str = h5sparse.get_format_str(self.sparse_class((0,0)))
format_str = h5sparse.get_format_str(self.sparse_class((0, 0)))
with h5sparse.File(h5_path, 'w') as h5f:
h5f.create_dataset('sparse/matrix', sparse_format=format_str)
with h5sparse.File(h5_path, 'r') as h5f:
Expand All @@ -27,15 +28,14 @@ def test_create_empty_sparse_dataset(self):

os.remove(h5_path)


def test_create_dataset_from_dataset(self):
from_h5_path = mkstemp(suffix=".h5")[1]
to_h5_path = mkstemp(suffix=".h5")[1]
sparse_matrix = self.sparse_class([[0, 1, 0],
[0, 0, 1],
[0, 0, 0],
[1, 1, 0]],
dtype=np.float64)
[0, 0, 1],
[0, 0, 0],
[1, 1, 0]],
dtype=np.float64)
with h5sparse.File(from_h5_path, 'w') as from_h5f:
from_dset = from_h5f.create_dataset('sparse/matrix', data=sparse_matrix)

Expand All @@ -57,7 +57,6 @@ def test_numpy_array(self):
np.testing.assert_equal(h5f['matrix'][()], matrix)
os.remove(h5_path)


def test_bytestring(self):
h5_path = mkstemp(suffix=".h5")[1]
strings = [str(i) for i in range(100)]
Expand All @@ -68,7 +67,6 @@ def test_bytestring(self):
assert strings == json.loads(h5f['strings'][()].decode('utf8'))
os.remove(h5_path)


def test_create_empty_dataset(self):
h5_path = mkstemp(suffix=".h5")[1]
with h5sparse.File(h5_path, 'w') as h5f:
Expand All @@ -84,10 +82,10 @@ class Test5HCSR(unittest.TestCase, AbstractTestH5Sparse):
def test_create_and_read_dataset(self):
h5_path = mkstemp(suffix=".h5")[1]
sparse_matrix = self.sparse_class([[0, 1, 0],
[0, 0, 1],
[0, 0, 0],
[1, 1, 0]],
dtype=np.float64)
[0, 0, 1],
[0, 0, 0],
[1, 1, 0]],
dtype=np.float64)
with h5sparse.File(h5_path, 'w') as h5f:
h5f.create_dataset('sparse/matrix', data=sparse_matrix)
with h5sparse.File(h5_path, 'r') as h5f:
Expand All @@ -105,18 +103,18 @@ def test_create_and_read_dataset(self):
def test_dataset_append(self):
h5_path = mkstemp(suffix=".h5")[1]
sparse_matrix = self.sparse_class([[0, 1, 0],
[0, 0, 1],
[0, 0, 0],
[1, 1, 0]],
dtype=np.float64)
[0, 0, 1],
[0, 0, 0],
[1, 1, 0]],
dtype=np.float64)
to_append = self.sparse_class([[0, 1, 1],
[1, 0, 0]],
dtype=np.float64)
[1, 0, 0]],
dtype=np.float64)
appended_matrix = ss.vstack((sparse_matrix, to_append))

with h5sparse.File(h5_path, 'w') as h5f:
h5f.create_dataset('matrix', data=sparse_matrix, chunks=(100000,),
maxshape=(None,))
maxshape=(None,))
h5f['matrix'].append(to_append)
assert (h5f['matrix'][()] != appended_matrix).size == 0

Expand All @@ -125,10 +123,10 @@ def test_dataset_append(self):
def test_create_dataset_with_format_change(self):
h5_path = mkstemp(suffix=".h5")[1]
sparse_matrix = self.sparse_class([[0, 1, 0, 1],
[0, 0, 1, 0],
[0, 0, 0, 1],
[1, 1, 0, 1]],
dtype=np.float64)
[0, 0, 1, 0],
[0, 0, 0, 1],
[1, 1, 0, 1]],
dtype=np.float64)
with h5sparse.File(h5_path, 'w') as h5f:
h5f.create_dataset('sparse/matrix', data=sparse_matrix, sparse_format='csc')
with h5sparse.File(h5_path, 'r') as h5f:
Expand All @@ -146,8 +144,10 @@ def test_create_dataset_with_format_change(self):

os.remove(h5_path)


class Test5HCSC(unittest.TestCase, AbstractTestH5Sparse):
sparse_class = ss.csc_matrix


class Test5HCOO(unittest.TestCase, AbstractTestH5Sparse):
sparse_class = ss.coo_matrix
sparse_class = ss.coo_matrix

0 comments on commit 1abce2c

Please sign in to comment.