File: compression.py

package info (click to toggle)
python-fabio 0.8.0+dfsg-1
  • links: PTS, VCS
  • area: main
  • in suites: buster, sid
  • size: 2,116 kB
  • sloc: python: 15,111; ansic: 1,091; sh: 233; makefile: 200
file content (469 lines) | stat: -rw-r--r-- 15,581 bytes parent folder | download | duplicates (2)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
# coding: utf-8
#
#    Project: X-ray image reader
#             https://github.com/silx-kit/fabio
#
#
#    Copyright (C) European Synchrotron Radiation Facility, Grenoble, France
#
#    Principal author:       Jérôme Kieffer (Jerome.Kieffer@ESRF.eu)
#
#  Permission is hereby granted, free of charge, to any person
#  obtaining a copy of this software and associated documentation files
#  (the "Software"), to deal in the Software without restriction,
#  including without limitation the rights to use, copy, modify, merge,
#  publish, distribute, sublicense, and/or sell copies of the Software,
#  and to permit persons to whom the Software is furnished to do so,
#  subject to the following conditions:
#
#  The above copyright notice and this permission notice shall be
#  included in all copies or substantial portions of the Software.
#
#  THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
#  EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES
#  OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
#  NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
#  HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
#  WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
#  FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
#  OTHER DEALINGS IN THE SOFTWARE.


"""Compression and decompression algorithm for various formats

Authors: Jérôme Kieffer, ESRF
         email:jerome.kieffer@esrf.fr

"""
# get ready for python3
from __future__ import absolute_import, print_function, with_statement, division
__author__ = "Jérôme Kieffer"
__contact__ = "jerome.kieffer@esrf.eu"
__license__ = "MIT"
__date__ = "22/10/2018"
__copyright__ = "European Synchrotron Radiation Facility, Grenoble, France"


import sys
import base64
import hashlib
import logging
import subprocess
import numpy

logger = logging.getLogger(__name__)

from .third_party import six

try:
    from .third_party import gzip
except ImportError:
    logger.error("Unable to import gzip module: disabling gzip compression")
    gzip = None

try:
    import bz2
except ImportError:
    logger.error("Unable to import bz2 module: disabling bz2 compression")
    bz2 = None

try:
    import zlib
except ImportError:
    logger.error("Unable to import zlib module: disabling zlib compression")
    zlib = None

if sys.platform != "win32":
    WindowsError = OSError


def is_incomplete_gz_block_exception(exception):
    """True if the exception looks to be generated when a GZ block is
    incomplete.

    :rtype: bool
    """
    if six.PY2:
        if isinstance(exception, IOError):
            return "CRC check failed" in exception.args[0]
    elif six.PY3:
        version = sys.version_info[0:2]
        if version == (3, 3):
            import struct
            return isinstance(exception, struct.error)
        return isinstance(exception, EOFError)

    return False


def md5sum(blob):
    """
    returns the md5sum of an object...
    """
    return base64.b64encode(hashlib.md5(blob).digest())


def endianness():
    """
    Return the native endianness of the system
    """
    if numpy.little_endian:
        return "LITTLE_ENDIAN"
    else:
        return "BIG_ENDIAN"


class ExternalCompressors(object):
    """Class to handle lazy discovery of external compression programs"""
    COMMANDS = ((".bz2", ["bzip2" "-dcf"]),
                (".gz", ["gzip", "-dcf"]),
                )

    def __init__(self):
        """Empty constructor"""
        self.compressors = {}

    def __getitem__(self, key):
        """Implement the dict-like behavior"""
        if key not in self.compressors:
            for candidate, commandline in self.COMMANDS:
                if key != candidate:
                    continue
                testline = [commandline[0], "-h"]
                try:
                    lines = subprocess.check_output(testline,
                                                    stderr=subprocess.STDOUT,
                                                    universal_newlines=True)
                    if "usage" in lines.lower():
                        self.compressors[key] = commandline
                    else:
                        self.compressors[key] = None
                except (subprocess.CalledProcessError, WindowsError) as err:
                    logger.debug("No %s utility found: %s", commandline[0], err)
                    self.compressors[key] = None
            else:
                self.compressors[key] = None
        return self.compressors[key]


COMPRESSORS = ExternalCompressors()


def decGzip(stream):
    """Decompress a chunk of data using the gzip algorithm from system or from Python

    :param stream: compressed data
    :return: uncompressed stream

    """
    def _python_gzip(stream):
        """Inefficient implementation based on loops in Python"""
        for i in range(1, 513):
            try:
                fileobj = six.BytesIO(stream[:-i])
                uncompessed = gzip.GzipFile(fileobj=fileobj).read()
            except IOError:
                logger.debug("trying with %s bytes less, doesn't work" % i)
            else:
                return uncompessed

    if gzip is None:
        raise ImportError("gzip module is not available")
    fileobj = six.BytesIO(stream)
    try:
        uncompessed = gzip.GzipFile(fileobj=fileobj).read()
    except IOError:
        logger.warning("Encounter the python-gzip bug with trailing garbage, trying subprocess gzip")
        cmd = COMPRESSORS[".gz"]
        if cmd:
            try:
                sub = subprocess.Popen(cmd, stdout=subprocess.PIPE, stdin=subprocess.PIPE, stderr=subprocess.PIPE)
                uncompessed, err = sub.communicate(input=stream)
                logger.debug("Gzip subprocess ended with %s err= %s; I got %s bytes back" % (sub.wait(), err, len(uncompessed)))
            except OSError as error:
                logger.warning("Unable to use the subprocess gzip (%s). Is gzip available? " % error)
                uncompessed = _python_gzip(stream)
        else:
            uncompessed = _python_gzip(stream)
        if uncompessed is None:
            logger.error("I am totally unable to read this gzipped compressed data block, giving up")
    return uncompessed


def decBzip2(stream):
    """
    Decompress a chunk of data using the bzip2 algorithm from Python
    """
    if bz2 is None:
        raise ImportError("bz2 module is not available")
    return bz2.decompress(stream)


def decZlib(stream):
    """
    Decompress a chunk of data using the zlib algorithm from Python
    """
    if zlib is None:
        raise ImportError("zlib module is not available")
    return zlib.decompress(stream)


def decByteOffset_numpy(stream, size=None, dtype="int64"):
    """
    Analyze a stream of char with any length of exception:
                2, 4, or 8 bytes integers

    :param stream: string representing the compressed data
    :param size: the size of the output array (of longInts)
    :return: 1D-ndarray

    """
    logger.debug("CBF decompression using Numpy")
    listnpa = []
    key16 = b"\x80"
    key32 = b"\x00\x80"
    key64 = b"\x00\x00\x00\x80"
    shift = 1
    while True:
        idx = stream.find(key16)
        if idx == -1:
            listnpa.append(numpy.frombuffer(stream, dtype=numpy.int8))
            break
        listnpa.append(numpy.frombuffer(stream[:idx], dtype=numpy.int8))

        if stream[idx + 1:idx + 3] == key32:
            if stream[idx + 3:idx + 7] == key64:
                # 64 bits int
                res = numpy.frombuffer(stream[idx + 7:idx + 15], dtype=numpy.int64)
                listnpa.append(res)
                shift = 15
            else:
                # 32 bits int
                res = numpy.frombuffer(stream[idx + 3:idx + 7], dtype=numpy.int32)
                listnpa.append(res)
                shift = 7
        else:  # int16
            res = numpy.frombuffer(stream[idx + 1:idx + 3], dtype=numpy.int16)
            listnpa.append(res)
            shift = 3
        stream = stream[idx + shift:]
    if not numpy.little_endian:
        for res in listnpa:
            if res.dtype != numpy.int8:
                res.byteswap(True)
    return numpy.ascontiguousarray(numpy.hstack(listnpa), dtype).cumsum()


def decByteOffset_cython(stream, size=None, dtype="int64"):
    """
    Analyze a stream of char with any length of exception:
                2, 4, or 8 bytes integers

    :param stream: string representing the compressed data
    :param size: the size of the output array (of longInts)
    :return: 1D-ndarray

    """
    logger.debug("CBF decompression using cython")
    try:
        from .ext import byte_offset
    except ImportError as error:
        logger.error("Failed to import byte_offset cython module, falling back on numpy method: %s", error)
        return decByteOffset_numpy(stream, size, dtype=dtype)
    else:
        if dtype == "int32":
            return byte_offset.dec_cbf32(stream, size)
        else:
            return byte_offset.dec_cbf(stream, size)


decByteOffset = decByteOffset_cython


def compByteOffset_numpy(data):
    """
    Compress a dataset into a string using the byte_offet algorithm

    :param data: ndarray
    :return: string/bytes with compressed data

    test = numpy.array([0,1,2,127,0,1,2,128,0,1,2,32767,0,1,2,32768,0,1,2,2147483647,0,1,2,2147483648,0,1,2,128,129,130,32767,32768,128,129,130,32768,2147483647,2147483648])

    """
    flat = numpy.ascontiguousarray(data.ravel(), numpy.int64)
    delta = numpy.zeros_like(flat)
    delta[0] = flat[0]
    delta[1:] = flat[1:] - flat[:-1]
    mask = abs(delta) > 127
    exceptions = numpy.nonzero(mask)[0]
    if numpy.little_endian:
        byteswap = False
    else:
        byteswap = True
    start = 0
    binary_blob = b""
    for stop in exceptions:
        if stop - start > 0:
            binary_blob += delta[start:stop].astype(numpy.int8).tostring()
        exc = delta[stop]
        absexc = abs(exc)
        if absexc > 2147483647:  # 2**31-1
            binary_blob += b"\x80\x00\x80\x00\x00\x00\x80"
            if byteswap:
                binary_blob += delta[stop:stop + 1].byteswap().tostring()
            else:
                binary_blob += delta[stop:stop + 1].tostring()
        elif absexc > 32767:  # 2**15-1
            binary_blob += b"\x80\x00\x80"
            if byteswap:
                binary_blob += delta[stop:stop + 1].astype(numpy.int32).byteswap().tostring()
            else:
                binary_blob += delta[stop:stop + 1].astype(numpy.int32).tostring()
        else:  # >127
            binary_blob += b"\x80"
            if byteswap:
                binary_blob += delta[stop:stop + 1].astype(numpy.int16).byteswap().tostring()
            else:
                binary_blob += delta[stop:stop + 1].astype(numpy.int16).tostring()
        start = stop + 1
    if start < delta.size:
        binary_blob += delta[start:].astype(numpy.int8).tostring()
    return binary_blob


def compByteOffset_cython(data):
    """
    Compress a dataset into a string using the byte_offet algorithm

    :param data: ndarray
    :return: string/bytes with compressed data

    test = numpy.array([0,1,2,127,0,1,2,128,0,1,2,32767,0,1,2,32768,0,1,2,2147483647,0,1,2,2147483648,0,1,2,128,129,130,32767,32768,128,129,130,32768,2147483647,2147483648])

    """
    logger.debug("CBF compression using cython")
    try:
        from .ext import byte_offset
    except ImportError as error:
        logger.error("Failed to import byte_offset cython module, falling back on numpy method: %s", error)
        return compByteOffset_numpy(data)
    else:
        if "int32" in str(data.dtype):
            return byte_offset.comp_cbf32(data).tostring()
        else:
            return byte_offset.comp_cbf(data).tostring()


compByteOffset = compByteOffset_cython


def decTY1(raw_8, raw_16=None, raw_32=None):
    """
    Modified byte offset decompressor used in Oxford Diffraction images

    Note: Always expect little endian data on the disk

    :param raw_8:  strings containing raw data with integer 8 bits
    :param raw_16: strings containing raw data with integer 16 bits
    :param raw_32: strings containing raw data with integer 32 bits
    :return: numpy.ndarray

    """
    data = numpy.frombuffer(raw_8, dtype=numpy.uint8).astype(int) - 127
    if raw_32 is not None:
        int32 = numpy.frombuffer(raw_32, dtype=numpy.int32)
        if not numpy.little_endian:
            int32.byteswap(True)
        exception32 = numpy.where(data == 128)
    if raw_16 is not None:
        int16 = numpy.frombuffer(raw_16, dtype="int16")
        if not numpy.little_endian:
            int16.byteswap(True)
        exception16 = numpy.where(data == 127)
        data[exception16] = int16
    if raw_32:
        data[exception32] = int32
    summed = data.cumsum()
    smax = summed.max()
    if (smax > (2 ** 31 - 1)):
        bytecode = "int64"
    elif (smax > (2 ** 15 - 1)):
        bytecode = "int32"
    elif (smax > (2 ** 7 - 1)):
        bytecode = "int16"
    else:
        bytecode = "int8"
    return summed.astype(bytecode)


decKM4CCD = decTY1


def compTY1(data):
    """
    Modified byte offset compressor used in Oxford Diffraction images

    :param data: numpy.ndarray with the input data (integers!)
    :return: 3-tuple of strings: raw_8,raw_16,raw_32 containing raw data with integer of the given size

    """
    fdata = data.ravel()
    diff = numpy.zeros_like(fdata)
    diff[0] = fdata[0]
    diff[1:] = fdata[1:] - fdata[:-1]
    adiff = abs(diff)
    exception32 = (adiff > (1 << 15) - 1)
    exception16 = (adiff >= (1 << 7) - 1) ^ exception32
    we16 = numpy.where(exception16)
    we32 = numpy.where(exception32)
    data_16 = diff[we16].astype(numpy.int16)
    data_32 = diff[we32].astype(numpy.int32)
    if not numpy.little_endian:
        data_16.byteswap(True)
        data_32.byteswap(True)
    diff[we16] = 127
    diff[we32] = 128
    diff += 127
    data_8 = diff.astype(numpy.uint8)
    return data_8.tostring(), data_16.tostring(), data_32.tostring()


def decPCK(stream, dim1=None, dim2=None, overflowPix=None, version=None, normal_start=None, swap_needed=None):
    """
    Modified CCP4  pck decompressor used in MAR345 images

    :param raw: input string (bytes in python3)
    :param dim1,dim2: optional parameters size
    :param overflowPix: optional parameters: number of overflowed pixels
    :param version: PCK version 1 or 2
    :param normal_start: position of the normal value section (can be auto-guessed)
    :param swap_needed: set to True when reading data from a foreign endianness (little on big or big on little)
    :return: ndarray of 2D with the right size

    """
    try:
        from .ext.mar345_IO import uncompress_pck
    except ImportError as error:
        raise RuntimeError("Unable to import mar345_IO to read compressed dataset: %s" % error)
    if "seek" in dir(stream):
        stream.seek(0)
        raw = stream.read()
    else:
        raw = six.binary_type(stream)

    return uncompress_pck(raw, dim1, dim2, overflowPix, version, normal_start, swap_needed)


def compPCK(data):
    """
    Modified CCP4  pck compressor used in MAR345 images

    :param data: numpy.ndarray (square array)
    :return:  compressed stream

    """
    try:
        from .ext.mar345_IO import compress_pck
    except ImportError as error:
        raise RuntimeError("Unable to import mar345_IO to write compressed dataset: %s" % error)
    return compress_pck(data)