scripts.config_generator.hdf_utils.get_hdf_information() - Code Metrics - DiamondLightSource/Savu - Measure and Improve Code Quality continuously with Scrutinizer

get_hdf_information() F
last analyzed 2022-08-17 15:44 UTC

↳ Parent: scripts.config_generator.hdf_utils

Complexity

Conditions

Size

Total Lines	66
Code Lines	43

Duplication

Lines	0
Ratio	0 %

Importance

Changes

Metric	Value
cc	15
eloc	43
nop	2
dl	0
loc	66
rs	2.9998
c	0
b	0
f	0

How to fix Long Method Complexity

# Copyright 2014 Diamond Light Source Ltd.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

"""
.. module:: hdf_utils
   :platform: Unix
   :synopsis: Utilities for checking hdf/nxs files

.. moduleauthor:: Nghia Vo <[email protected]>

"""

import h5py
import numpy as np
from collections import deque

PIPE = "│"
ELBOW = "└──"
TEE = "├──"
PIPE_PREFIX = "│   "
SPACE_PREFIX = "    "
TOMO_DATA = "entry1/tomo_entry/data/data"
ROTATION_ANGLE = "entry1/tomo_entry/data/rotation_angle"
IMAGE_KEY = "entry1/tomo_entry/instrument/detector/image_key"


def get_hdf_information(file_path, display=False):
    """
    Get information of datasets in a hdf/nxs file.

    Parameters
    ----------
    file_path : str
        Path to the file.
    display : bool
        Print the results onto the screen if True.

    Returns
    -------
    list_key : str
        Keys to the datasets.
    list_shape : tuple of int
        Shapes of the datasets.
    list_type : str
        Types of the datasets.
    """
    hdf_object = h5py.File(file_path, 'r')
    keys = []
    hdf_object.visit(keys.append)
    list_key, list_shape, list_type = [], [], []
    for key in keys:
        try:
            data = hdf_object[key]
            if isinstance(data, h5py.Group):
                list_tmp = list(data.items())
                if list_tmp:
                    for key2, _ in list_tmp:
                        list_key.append(key + "/" + key2)
                else:
                    list_key.append(key)
            else:
                list_key.append(data.name)
        except KeyError:
            list_key.append(key)
            pass
    for i, key in enumerate(list_key):
        try:
            data = hdf_object[list_key[i]]
            if isinstance(data, h5py.Dataset):
                shape, dtype = data.shape, data.dtype
            else:
                shape, dtype = None, None
            if isinstance(data, list):
                if len(data) == 1:
                    if not isinstance(data, np.ndarray):
                        dtype = str(list(data)[0])
                        dtype.replace("b'", "'")
            list_shape.append(shape)
            list_type.append(dtype)
        except KeyError:
            list_shape.append(None)
            list_type.append(None)
            pass
    hdf_object.close()
    if display:
        if list_key:
            for i, key in enumerate(list_key):
                print(key + " : " + str(list_shape[i]) + " : " + str(
                    list_type[i]))
        else:
            print("Empty file !!!")
    return list_key, list_shape, list_type


def find_hdf_key(file_path, pattern, display=False):
    """
    Find datasets matching the name-pattern in a hdf/nxs file.

    Parameters
    ----------
    file_path : str
        Path to the file.
    pattern : str
        Pattern to find the full names of the datasets.
    display : bool
        Print the results onto the screen if True.

    Returns
    -------
    list_key : str
        Keys to the datasets.
    list_shape : tuple of int
        Shapes of the datasets.
    list_type : str
        Types of the datasets.
    """
    hdf_object = h5py.File(file_path, 'r')
    list_key, keys = [], []
    hdf_object.visit(keys.append)
    for key in keys:
        try:
            data = hdf_object[key]
            if isinstance(data, h5py.Group):
                list_tmp = list(data.items())
                if list_tmp:
                    for key2, _ in list_tmp:
                        list_key.append(key + "/" + key2)
                else:
                    list_key.append(key)
            else:
                list_key.append(data.name)
        except KeyError:
            pass
    list_dkey, list_dshape, list_dtype = [], [], []
    for _, key in enumerate(list_key):
        if pattern in key:
            list_dkey.append(key)
            try:
                data = hdf_object[key]
                if isinstance(data, h5py.Dataset):
                    shape, dtype = data.shape, data.dtype
                else:
                    shape, dtype = None, None
                if isinstance(data, list):
                    if len(data) == 1:
                        if not isinstance(data, np.ndarray):
                            dtype = str(list(data)[0])
                            dtype.replace("b'", "'")
                list_dtype.append(dtype)
                list_dshape.append(shape)
            except KeyError:
                list_dtype.append(None)
                list_dshape.append(None)
                pass
    hdf_object.close()
    if display:
        if list_dkey:
            for i, key in enumerate(list_dkey):
                print(key + " : " + str(list_dshape[i]) + " : " + str(
                    list_dtype[i]))
        else:
            print("Can't find datasets with keys matching the "
                  "pattern: {}".format(pattern))
    return list_dkey, list_dshape, list_dtype


def _get_subgroups(hdf_object, key=None):
    """
    Supplementary method for building the tree view of a hdf5 file.
    Return the name of subgroups.
    """
    list_group = []
    if key is None:
        for group in hdf_object.keys():
            list_group.append(group)
        if len(list_group) == 1:
            key = list_group[0]
        else:
            key = ""
    else:
        if key in hdf_object:
            try:
                obj = hdf_object[key]
                if isinstance(obj, h5py.Group):
                    for group in hdf_object[key].keys():
                        list_group.append(group)
            except KeyError:
                pass
    if len(list_group) > 0:
        list_group = sorted(list_group)
    return list_group, key


def _add_branches(tree, hdf_object, key, key1, index, last_index, prefix,
                  connector, level, add_shape):
    """
    Supplementary method for building the tree view of a hdf5 file.
    Add branches to the tree.
    """
    shape = None
    key_comb = key + "/" + key1
    if add_shape is True:
        if key_comb in hdf_object:
            try:
                obj = hdf_object[key_comb]
                if isinstance(obj, h5py.Dataset):
                    shape = str(obj.shape)
            except KeyError:
                shape = str("-> ???External-link???")
    if shape is not None:
        tree.append(f"{prefix}{connector} {key1} {shape}")
    else:
        tree.append(f"{prefix}{connector} {key1}")
    if index != last_index:
        prefix += PIPE_PREFIX
    else:
        prefix += SPACE_PREFIX
    _make_tree_body(tree, hdf_object, prefix=prefix, key=key_comb,
                    level=level, add_shape=add_shape)


def _make_tree_body(tree, hdf_object, prefix="", key=None, level=0,
                    add_shape=True):
    """
    Supplementary method for building the tree view of a hdf5 file.
    Create the tree body.
    """
    entries, key = _get_subgroups(hdf_object, key)
    num_ent = len(entries)
    last_index = num_ent - 1
    level = level + 1
    if num_ent > 0:
        if last_index == 0:
            key = "" if level == 1 else key
            if num_ent > 1:
                connector = PIPE
            else:
                connector = ELBOW if level > 1 else ""
            _add_branches(tree, hdf_object, key, entries[0], 0, 0, prefix,
                          connector, level, add_shape)
        else:
            for index, key1 in enumerate(entries):
                connector = ELBOW if index == last_index else TEE
                if index == 0:
                    tree.append(prefix + PIPE)
                _add_branches(tree, hdf_object, key, key1, index, last_index,
                              prefix, connector, level, add_shape)


def get_hdf_tree(file_path, add_shape=True, display=True):
    """
    Get the tree view of a hdf/nxs file.

    Parameters
    ----------
    file_path : str
        Path to the file.
    add_shape : bool
        Including the shape of a dataset to the tree if True.
    display : bool
        Print the tree onto the screen if True.

    Returns
    -------
    list of string
    """
    hdf_object = h5py.File(file_path, 'r')
    tree = deque()
    _make_tree_body(tree, hdf_object, add_shape=add_shape)
    if display:
        for entry in tree:
            print(entry)
    return tree


def check_tomo_data(file_path):
    """
    To check:
    - If paths to datasets in a hdf/nxs file following the Diamond-tomo data
      convention.
    - Shapes between datasets are consistent.
    """
    path1, shape1, _ = find_hdf_key(file_path, TOMO_DATA)
    path2, shape2, _ = find_hdf_key(file_path, ROTATION_ANGLE)
    path3, shape3, _ = find_hdf_key(file_path, IMAGE_KEY)
    msg = []
    got_it = True
    if not path1:
        msg.append(" -> Can't find the path: '{0}' "
                   "to tomo-data".format(TOMO_DATA))
        got_it = False
    else:
        if not shape1:
            msg.append(" -> Empty data in: '{0}'".format(TOMO_DATA))
            got_it = False
        else:
            shape1 = shape1[0][0]
    if not path2:
        msg.append(" -> Can't find the path: '{0}' to "
                   "rotation angles".format(ROTATION_ANGLE))
        got_it = False
    else:
        if not shape2:
            msg.append(" -> Empty data in: '{0}'".format(ROTATION_ANGLE))
            got_it = False
        else:
            shape2 = list(shape2)[0][0]
    if not path3:
        msg.append(" -> Can't find the path: '{0}' to "
                   "image-keys".format(IMAGE_KEY))
        got_it = False
    else:
        if not shape3:
            msg.append(" -> Empty data in: '{0}'".format(IMAGE_KEY))
            got_it = False
        else:
            shape3 = list(shape3)[0][0]
    if shape1 != shape2:
        msg.append(" -> Number of projections: {0} is not the same as the"
                   " number of rotation-angles: {1}".format(shape1, shape2))
        got_it = False
    if shape1 != shape3:
        msg.append(" -> Number of projections: {0} is not the same as the"
                   " number of image-keys: {1}".format(shape1, shape3))
        got_it = False
    if shape2 != shape3:
        msg.append(" -> Number of rotation-angles: {0} is not the same as the"
                   " number of image-keys: {1}".format(shape2, shape3))
        got_it = False
    if got_it is True:
        print("=============================================================")
        print("Paths to datasets following the default names used by "
              "NxTomoLoader:")
        print("   Path to tomo-data: '{0}'. Shape: {1}".format(
            path1[0], shape1))
        print("   Path to rotation-angles: '{0}'. Shape: {1}".format(
            path2[0], shape2))
        print("   Path to image-keys: '{0}'. Shape: {1}".format(
            path3[0], shape3))
        print("=============================================================")
    else:
        print("=========================!!!WARNING!!!=======================")
        for entry in msg:
            print("  " + entry)
        print("=============================================================")


1			# Copyright 2014 Diamond Light Source Ltd.
2			#
3			# Licensed under the Apache License, Version 2.0 (the "License");
4			# you may not use this file except in compliance with the License.
5			# You may obtain a copy of the License at
6			#
7			# http://www.apache.org/licenses/LICENSE-2.0
8			#
9			# Unless required by applicable law or agreed to in writing, software
10			# distributed under the License is distributed on an "AS IS" BASIS,
11			# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
12			# See the License for the specific language governing permissions and
13			# limitations under the License.
14
15			"""
16			.. module:: hdf_utils
17			:platform: Unix
18			:synopsis: Utilities for checking hdf/nxs files
19
20			.. moduleauthor:: Nghia Vo <[email protected]>
21
22			"""
23
24			import h5py
25			import numpy as np
26			from collections import deque
27
28			PIPE = "│"
29			ELBOW = "└──"
30			TEE = "├──"
31			PIPE_PREFIX = "│ "
32			SPACE_PREFIX = " "
33			TOMO_DATA = "entry1/tomo_entry/data/data"
34			ROTATION_ANGLE = "entry1/tomo_entry/data/rotation_angle"
35			IMAGE_KEY = "entry1/tomo_entry/instrument/detector/image_key"
36
37
38			def get_hdf_information(file_path, display=False):
39			"""
40			Get information of datasets in a hdf/nxs file.
41
42			Parameters
43			----------
44			file_path : str
45			Path to the file.
46			display : bool
47			Print the results onto the screen if True.
48
49			Returns
50			-------
51			list_key : str
52			Keys to the datasets.
53			list_shape : tuple of int
54			Shapes of the datasets.
55			list_type : str
56			Types of the datasets.
57			"""
58			hdf_object = h5py.File(file_path, 'r')
59			keys = []
60			hdf_object.visit(keys.append)
61			list_key, list_shape, list_type = [], [], []
62			for key in keys:
63			try:
64			data = hdf_object[key]
65			if isinstance(data, h5py.Group):
66			list_tmp = list(data.items())
67			if list_tmp:
68			for key2, _ in list_tmp:
69			list_key.append(key + "/" + key2)
70			else:
71			list_key.append(key)
72			else:
73			list_key.append(data.name)
74			except KeyError:
75			list_key.append(key)
76			pass
77			for i, key in enumerate(list_key):
78			try:
79			data = hdf_object[list_key[i]]
80			if isinstance(data, h5py.Dataset):
81			shape, dtype = data.shape, data.dtype
82			else:
83			shape, dtype = None, None
84			if isinstance(data, list):
85			if len(data) == 1:
86			if not isinstance(data, np.ndarray):
87			dtype = str(list(data)[0])
88			dtype.replace("b'", "'")
89			list_shape.append(shape)
90			list_type.append(dtype)
91			except KeyError:
92			list_shape.append(None)
93			list_type.append(None)
94			pass
95			hdf_object.close()
96			if display:
97			if list_key:
98			for i, key in enumerate(list_key):
99			print(key + " : " + str(list_shape[i]) + " : " + str(
100			list_type[i]))
101			else:
102			print("Empty file !!!")
103			return list_key, list_shape, list_type
104
105
106			def find_hdf_key(file_path, pattern, display=False):
107			"""
108			Find datasets matching the name-pattern in a hdf/nxs file.
109
110			Parameters
111			----------
112			file_path : str
113			Path to the file.
114			pattern : str
115			Pattern to find the full names of the datasets.
116			display : bool
117			Print the results onto the screen if True.
118
119			Returns
120			-------
121			list_key : str
122			Keys to the datasets.
123			list_shape : tuple of int
124			Shapes of the datasets.
125			list_type : str
126			Types of the datasets.
127			"""
128			hdf_object = h5py.File(file_path, 'r')
129			list_key, keys = [], []
130			hdf_object.visit(keys.append)
131			for key in keys:
132			try:
133			data = hdf_object[key]
134			if isinstance(data, h5py.Group):
135			list_tmp = list(data.items())
136			if list_tmp:
137			for key2, _ in list_tmp:
138			list_key.append(key + "/" + key2)
139			else:
140			list_key.append(key)
141			else:
142			list_key.append(data.name)
143			except KeyError:
144			pass
145			list_dkey, list_dshape, list_dtype = [], [], []
146			for _, key in enumerate(list_key):
147			if pattern in key:
148			list_dkey.append(key)
149			try:
150			data = hdf_object[key]
151			if isinstance(data, h5py.Dataset):
152			shape, dtype = data.shape, data.dtype
153			else:
154			shape, dtype = None, None
155			if isinstance(data, list):
156			if len(data) == 1:
157			if not isinstance(data, np.ndarray):
158			dtype = str(list(data)[0])
159			dtype.replace("b'", "'")
160			list_dtype.append(dtype)
161			list_dshape.append(shape)
162			except KeyError:
163			list_dtype.append(None)
164			list_dshape.append(None)
165			pass
166			hdf_object.close()
167			if display:
168			if list_dkey:
169			for i, key in enumerate(list_dkey):
170			print(key + " : " + str(list_dshape[i]) + " : " + str(
171			list_dtype[i]))
172			else:
173			print("Can't find datasets with keys matching the "
174			"pattern: {}".format(pattern))
175			return list_dkey, list_dshape, list_dtype
176
177
178			def _get_subgroups(hdf_object, key=None):
179			"""
180			Supplementary method for building the tree view of a hdf5 file.
181			Return the name of subgroups.
182			"""
183			list_group = []
184			if key is None:
185			for group in hdf_object.keys():
186			list_group.append(group)
187			if len(list_group) == 1:
188			key = list_group[0]
189			else:
190			key = ""
191			else:
192			if key in hdf_object:
193			try:
194			obj = hdf_object[key]
195			if isinstance(obj, h5py.Group):
196			for group in hdf_object[key].keys():
197			list_group.append(group)
198			except KeyError:
199			pass
200			if len(list_group) > 0:
201			list_group = sorted(list_group)
202			return list_group, key
203
204
205			def _add_branches(tree, hdf_object, key, key1, index, last_index, prefix,
206			connector, level, add_shape):
207			"""
208			Supplementary method for building the tree view of a hdf5 file.
209			Add branches to the tree.
210			"""
211			shape = None
212			key_comb = key + "/" + key1
213			if add_shape is True:
214			if key_comb in hdf_object:
215			try:
216			obj = hdf_object[key_comb]
217			if isinstance(obj, h5py.Dataset):
218			shape = str(obj.shape)
219			except KeyError:
220			shape = str("-> ???External-link???")
221			if shape is not None:
222			tree.append(f"{prefix}{connector} {key1} {shape}")
223			else:
224			tree.append(f"{prefix}{connector} {key1}")
225			if index != last_index:
226			prefix += PIPE_PREFIX
227			else:
228			prefix += SPACE_PREFIX
229			_make_tree_body(tree, hdf_object, prefix=prefix, key=key_comb,
230			level=level, add_shape=add_shape)
231
232
233			def _make_tree_body(tree, hdf_object, prefix="", key=None, level=0,
234			add_shape=True):
235			"""
236			Supplementary method for building the tree view of a hdf5 file.
237			Create the tree body.
238			"""
239			entries, key = _get_subgroups(hdf_object, key)
240			num_ent = len(entries)
241			last_index = num_ent - 1
242			level = level + 1
243			if num_ent > 0:
244			if last_index == 0:
245			key = "" if level == 1 else key
246			if num_ent > 1:
247			connector = PIPE
248			else:
249			connector = ELBOW if level > 1 else ""
250			_add_branches(tree, hdf_object, key, entries[0], 0, 0, prefix,
251			connector, level, add_shape)
252			else:
253			for index, key1 in enumerate(entries):
254			connector = ELBOW if index == last_index else TEE
255			if index == 0:
256			tree.append(prefix + PIPE)
257			_add_branches(tree, hdf_object, key, key1, index, last_index,
258			prefix, connector, level, add_shape)
259
260
261			def get_hdf_tree(file_path, add_shape=True, display=True):
262			"""
263			Get the tree view of a hdf/nxs file.
264
265			Parameters
266			----------
267			file_path : str
268			Path to the file.
269			add_shape : bool
270			Including the shape of a dataset to the tree if True.
271			display : bool
272			Print the tree onto the screen if True.
273
274			Returns
275			-------
276			list of string
277			"""
278			hdf_object = h5py.File(file_path, 'r')
279			tree = deque()
280			_make_tree_body(tree, hdf_object, add_shape=add_shape)
281			if display:
282			for entry in tree:
283			print(entry)
284			return tree
285
286
287			def check_tomo_data(file_path):
288			"""
289			To check:
290			- If paths to datasets in a hdf/nxs file following the Diamond-tomo data
291			convention.
292			- Shapes between datasets are consistent.
293			"""
294			path1, shape1, _ = find_hdf_key(file_path, TOMO_DATA)
295			path2, shape2, _ = find_hdf_key(file_path, ROTATION_ANGLE)
296			path3, shape3, _ = find_hdf_key(file_path, IMAGE_KEY)
297			msg = []
298			got_it = True
299			if not path1:
300			msg.append(" -> Can't find the path: '{0}' "
301			"to tomo-data".format(TOMO_DATA))
302			got_it = False
303			else:
304			if not shape1:
305			msg.append(" -> Empty data in: '{0}'".format(TOMO_DATA))
306			got_it = False
307			else:
308			shape1 = shape1[0][0]
309			if not path2:
310			msg.append(" -> Can't find the path: '{0}' to "
311			"rotation angles".format(ROTATION_ANGLE))
312			got_it = False
313			else:
314			if not shape2:
315			msg.append(" -> Empty data in: '{0}'".format(ROTATION_ANGLE))
316			got_it = False
317			else:
318			shape2 = list(shape2)[0][0]
319			if not path3:
320			msg.append(" -> Can't find the path: '{0}' to "
321			"image-keys".format(IMAGE_KEY))
322			got_it = False
323			else:
324			if not shape3:
325			msg.append(" -> Empty data in: '{0}'".format(IMAGE_KEY))
326			got_it = False
327			else:
328			shape3 = list(shape3)[0][0]
329			if shape1 != shape2:
330			msg.append(" -> Number of projections: {0} is not the same as the"
331			" number of rotation-angles: {1}".format(shape1, shape2))
332			got_it = False
333			if shape1 != shape3:
334			msg.append(" -> Number of projections: {0} is not the same as the"
335			" number of image-keys: {1}".format(shape1, shape3))
336			got_it = False
337			if shape2 != shape3:
338			msg.append(" -> Number of rotation-angles: {0} is not the same as the"
339			" number of image-keys: {1}".format(shape2, shape3))
340			got_it = False
341			if got_it is True:
342			print("=============================================================")
343			print("Paths to datasets following the default names used by "
344			"NxTomoLoader:")
345			print(" Path to tomo-data: '{0}'. Shape: {1}".format(
346			path1[0], shape1))
347			print(" Path to rotation-angles: '{0}'. Shape: {1}".format(
348			path2[0], shape2))
349			print(" Path to image-keys: '{0}'. Shape: {1}".format(
350			path3[0], shape3))
351			print("=============================================================")
352			else:
353			print("=========================!!!WARNING!!!=======================")
354			for entry in msg:
355			print(" " + entry)
356			print("=============================================================")
357

DiamondLightSource / Savu

get_hdf_information() F last analyzed 2022-08-17 15:44 UTC

Complexity

Size

Duplication

Importance

How to fix Long Method Complexity

Long Method

Complexity

Duplication Side-by-Side

Filter issues like

get_hdf_information() F
last analyzed 2022-08-17 15:44 UTC