File: byte_offset.pyx

package info (click to toggle)
python-fabio 0.14.0%2Bdfsg-2
  • links: PTS, VCS
  • area: main
  • in suites: bookworm
  • size: 10,068 kB
  • sloc: python: 20,053; ansic: 1,085; makefile: 217; sh: 215
file content (334 lines) | stat: -rw-r--r-- 11,519 bytes parent folder | download | duplicates (5)
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
# coding: utf-8
#
#    Project: X-ray image reader
#             https://github.com/silx-kit/fabio
#
#    Copyright (C) 2015 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.

"""
Cif Binary Files images are 2D images written by the Pilatus detector and others.
They use a modified (simplified) byte-offset algorithm.  This file contains the
decompression function from a string to an int64 numpy array.
"""

__author__ = "Jerome Kieffer"
__contact__ = "jerome.kieffer@esrf.eu"
__license__ = "MIT"
__copyright__ = "2010-2016, European Synchrotron Radiation Facility, Grenoble, France"
__date__ = "14/12/2020"


import numpy
import cython

from libc.stdint cimport int8_t, uint8_t, \
                         uint16_t, int16_t,\
                         int32_t, uint32_t,\
                         int64_t, uint64_t


@cython.boundscheck(False)
@cython.wraparound(False)
def comp_cbf32(data not None):
    """Compress a dataset using the byte-offset described for Pilatus

    :param data: array of integers
    :return: numpy array of chars
    """
    cdef:
        int32_t[::1] ary = numpy.ascontiguousarray(data.ravel(), dtype=numpy.int32)
        int size = ary.size, i = 0, j = 0
        int8_t[::1] output = numpy.zeros(size * 7, dtype=numpy.int8)
        int32_t last, current, delta, absdelta
    last = 0
    for i in range(size):
        current = ary[i]
        delta = current - last
        absdelta = delta if delta > 0 else -delta
        if absdelta >= 1 << 15:
            output[j] = -128
            output[j + 1] = 0
            output[j + 2] = -128
            output[j + 3] = (delta & 255)
            output[j + 4] = (delta >> 8) & 255
            output[j + 5] = (delta >> 16) & 255
            output[j + 6] = (delta >> 24)
            j += 7
        elif absdelta >= 1 << 7:
            output[j] = -128
            output[j + 1] = delta & 255
            output[j + 2] = (delta >> 8) & 255
            j += 3
        else:
            output[j] = delta
            j += 1
        last = current
    return numpy.asarray(output)[:j]


@cython.boundscheck(False)
@cython.wraparound(False)
def comp_cbf(data not None):
    """Compress a dataset using the byte-offset described for any int64

    :param data: array of integers
    :return: numpy array of chars
    """
    cdef:
        int64_t[::1] ary = numpy.ascontiguousarray(data.ravel(), dtype=numpy.int64)
        int size = ary.size, i = 0, j = 0
        int8_t[::1] output = numpy.zeros(size * 15, dtype=numpy.int8)
        int64_t last, current, delta, absdelta
    last = 0
    for i in range(size):
        current = ary[i]
        delta = current - last
        absdelta = delta if delta > 0 else -delta
        if absdelta >= 1 << 31:
            output[j] = -128
            output[j + 1] = 0
            output[j + 2] = -128
            output[j + 3] = 0
            output[j + 4] = 0
            output[j + 5] = 0
            output[j + 6] = -128
            output[j + 7] = (delta & 255)
            output[j + 8] = (delta >> 8) & 255
            output[j + 9] = (delta >> 16) & 255
            output[j + 10] = (delta >> 24) & 255
            output[j + 11] = (delta >> 32) & 255
            output[j + 12] = (delta >> 40) & 255
            output[j + 13] = (delta >> 48) & 255
            output[j + 14] = (delta >> 56) & 255
            j += 15
        elif absdelta >= 1 << 15:
            output[j] = -128
            output[j + 1] = 0
            output[j + 2] = -128
            output[j + 3] = (delta & 255)
            output[j + 4] = (delta >> 8) & 255
            output[j + 5] = (delta >> 16) & 255
            output[j + 6] = (delta >> 24)
            j += 7
        elif absdelta >= 1 << 7:
            output[j] = -128
            output[j + 1] = delta & 255
            output[j + 2] = (delta >> 8) & 255
            j += 3
        else:
            output[j] = delta
            j += 1
        last = current
    return numpy.asarray(output)[:j]


@cython.boundscheck(False)
@cython.wraparound(False)
def dec_cbf(bytes stream not None, size=None):
    """
    Analyze a stream of char with any length of exception (2,4, or 8 bytes integers)
    :param stream: bytes (string) representing the compressed data
    :param size: the size of the output array (of longInts)
    :return: int64 ndArrays
    """
    cdef:
        int               i = 0
        int               j = 0
        uint8_t     tmp8 = 0

        int64_t    last = 0
        int64_t    current = 0
        int64_t    tmp64 = 0
        int64_t    tmp64a = 0
        int64_t    tmp64b = 0
        int64_t    tmp64c = 0
        int64_t    tmp64d = 0
        int64_t    tmp64e = 0
        int64_t    tmp64f = 0
        int64_t    tmp64g = 0

        uint8_t    key8 = 0x80
        uint8_t    key0 = 0x00

        int csize
        int lenStream = < int > len(stream)
        uint8_t[::1] cstream = bytearray(stream)
        int64_t[::1] data_out
        
    if size is None:
        csize = lenStream
    else:
        csize = < int > size
    
    data_out = numpy.empty(csize, dtype=numpy.int64)
    
    with nogil:
        while (i < lenStream) and (j < csize):
            if (cstream[i] == key8):
                if ((cstream[i + 1] == key0) and (cstream[i + 2] == key8)):
                    if (cstream[i + 3] == key0) and (cstream[i + 4] == key0) and (cstream[i + 5] == key0) and (cstream[i + 6] == key8):
                        # Retrieve the interesting Bytes of data
                        tmp64g = cstream[i + 7]
                        tmp64f = cstream[i + 8]
                        tmp64e = cstream[i + 9]
                        tmp64d = cstream[i + 10]
                        tmp64c = cstream[i + 11]
                        tmp64b = cstream[i + 12]
                        tmp64a = cstream[i + 13]
                        tmp64  = <int8_t> cstream[i + 14]
                        # Assemble data into a 64 bits integer
                        current = (tmp64 << 56) | (tmp64a << 48) | (tmp64b << 40) | (tmp64c << 32) | (tmp64d << 24) | (tmp64e << 16) | (tmp64f << 8) | (tmp64g)
                        i += 15
                    else:
                        # Retrieve the interesting Bytes of data
                        tmp64c = cstream[i + 3]
                        tmp64b = cstream[i + 4]
                        tmp64a = cstream[i + 5]
                        tmp64  = <int8_t> cstream[i + 6]
                        # Assemble data into a 64 bits integer
                        current = (tmp64 << 24) | (tmp64a << 16) | (tmp64b << 8) | (tmp64c)
                        i += 7
                else:
                    tmp64a = cstream[i + 1]
                    tmp64 = <int8_t> cstream[i + 2]

                    current = (tmp64 << 8) | (tmp64a)
                    i += 3
            else:
                current = (<int8_t> cstream[i])
                i += 1
            last += current
            data_out[j] = last
            j += 1

    return data_out[:j]


@cython.boundscheck(False)
def dec_cbf32(bytes stream not None, size=None):
    """
    Analyze a stream of char with any length of exception (2 or 4 bytes integers)
    Optimized for int32 decompression

    :param stream: bytes (string) representing the compressed data
    :param size: the size of the output array (of longInts)
    :return: int64 ndArrays
    """
    cdef:
        int               i = 0
        int               j = 0
        uint8_t     tmp8 = 0

        int32_t    last = 0
        int32_t    current = 0
        int32_t    tmp64 = 0
        int32_t    tmp64a = 0
        int32_t    tmp64b = 0
        int32_t    tmp64c = 0

        uint8_t    key8 = 0x80
        uint8_t    key0 = 0x00

        int csize
        int lenStream = < int > len(stream)
        uint8_t[:] cstream = bytearray(stream)
        int32_t[::1] data_out
    if size is None:
        csize = lenStream
    else:
        csize = < int > size
    data_out = numpy.empty(csize, dtype=numpy.int32)
    with nogil:
        while (i < lenStream) and (j < csize):
            if (cstream[i] == key8):
                if ((cstream[i + 1] == key0) and (cstream[i + 2] == key8)):
                    # Retrieve the interesting Bytes of data
                    tmp64c = cstream[i + 3]
                    tmp64b = cstream[i + 4]
                    tmp64a = cstream[i + 5]
                    tmp64  = <int8_t> cstream[i + 6]
                    # Assemble data into a 32 bits integer
                    current = (tmp64 << 24) | (tmp64a << 16) | (tmp64b << 8) | (tmp64c)
                    i += 7
                else:
                    tmp64a = cstream[i + 1]
                    tmp64  = <int8_t> cstream[i + 2]

                    current = (tmp64 << 8) | (tmp64a)
                    i += 3
            else:
                current = (<int8_t> cstream[i])
                i += 1
            last += current
            data_out[j] = last
            j += 1

    return numpy.asarray(data_out[:j])


@cython.boundscheck(False)
def dec_TY5(bytes stream not None, size=None):
    """
    Analyze a stream of char with a TY5 compression scheme and exception (2 or 4 bytes integers)

    TODO: known broken, FIXME

    :param stream: bytes (string) representing the compressed data
    :param size: the size of the output array (of longInts)
    :return: int32 ndArrays
    """

    cdef:
        int               i = 0
        int               j = 0
        int32_t    last = 0
        int32_t    current = 0
        uint8_t    key8 = 0xfe  # 127+127
        int32_t    tmp32a = 0
        int32_t    tmp32b = 0

        int csize
        int lenStream = len(stream)
        uint8_t[::1] cstream = bytearray(stream)
        int32_t[::1] data_out

    if size is None:
        csize = lenStream
    else:
        csize = < int > size

    data_out = numpy.zeros(csize, dtype=numpy.int32)
    if True:
        while (i < lenStream) and (j < csize):
            if (cstream[i] == key8):
                    tmp32a = cstream[i + 1] - 127
                    tmp32b = <int16_t>( <int8_t> cstream[i + 2] << 8 )
                    current = (tmp32b) | (tmp32a)
                    i += 3
            else:
                current = <int32_t>(<uint8_t> cstream[i]) - 127
                i += 1
            last += current
            data_out[j] = last
            j += 1
    return numpy.asarray(data_out[:j])