This file is indexed.

/usr/share/pyshared/PyMca/EdfFile.py is in pymca 4.5.0-4.

This file is owned by root:root, with mode 0o644.

The actual contents of the file can be viewed below.

   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
 470
 471
 472
 473
 474
 475
 476
 477
 478
 479
 480
 481
 482
 483
 484
 485
 486
 487
 488
 489
 490
 491
 492
 493
 494
 495
 496
 497
 498
 499
 500
 501
 502
 503
 504
 505
 506
 507
 508
 509
 510
 511
 512
 513
 514
 515
 516
 517
 518
 519
 520
 521
 522
 523
 524
 525
 526
 527
 528
 529
 530
 531
 532
 533
 534
 535
 536
 537
 538
 539
 540
 541
 542
 543
 544
 545
 546
 547
 548
 549
 550
 551
 552
 553
 554
 555
 556
 557
 558
 559
 560
 561
 562
 563
 564
 565
 566
 567
 568
 569
 570
 571
 572
 573
 574
 575
 576
 577
 578
 579
 580
 581
 582
 583
 584
 585
 586
 587
 588
 589
 590
 591
 592
 593
 594
 595
 596
 597
 598
 599
 600
 601
 602
 603
 604
 605
 606
 607
 608
 609
 610
 611
 612
 613
 614
 615
 616
 617
 618
 619
 620
 621
 622
 623
 624
 625
 626
 627
 628
 629
 630
 631
 632
 633
 634
 635
 636
 637
 638
 639
 640
 641
 642
 643
 644
 645
 646
 647
 648
 649
 650
 651
 652
 653
 654
 655
 656
 657
 658
 659
 660
 661
 662
 663
 664
 665
 666
 667
 668
 669
 670
 671
 672
 673
 674
 675
 676
 677
 678
 679
 680
 681
 682
 683
 684
 685
 686
 687
 688
 689
 690
 691
 692
 693
 694
 695
 696
 697
 698
 699
 700
 701
 702
 703
 704
 705
 706
 707
 708
 709
 710
 711
 712
 713
 714
 715
 716
 717
 718
 719
 720
 721
 722
 723
 724
 725
 726
 727
 728
 729
 730
 731
 732
 733
 734
 735
 736
 737
 738
 739
 740
 741
 742
 743
 744
 745
 746
 747
 748
 749
 750
 751
 752
 753
 754
 755
 756
 757
 758
 759
 760
 761
 762
 763
 764
 765
 766
 767
 768
 769
 770
 771
 772
 773
 774
 775
 776
 777
 778
 779
 780
 781
 782
 783
 784
 785
 786
 787
 788
 789
 790
 791
 792
 793
 794
 795
 796
 797
 798
 799
 800
 801
 802
 803
 804
 805
 806
 807
 808
 809
 810
 811
 812
 813
 814
 815
 816
 817
 818
 819
 820
 821
 822
 823
 824
 825
 826
 827
 828
 829
 830
 831
 832
 833
 834
 835
 836
 837
 838
 839
 840
 841
 842
 843
 844
 845
 846
 847
 848
 849
 850
 851
 852
 853
 854
 855
 856
 857
 858
 859
 860
 861
 862
 863
 864
 865
 866
 867
 868
 869
 870
 871
 872
 873
 874
 875
 876
 877
 878
 879
 880
 881
 882
 883
 884
 885
 886
 887
 888
 889
 890
 891
 892
 893
 894
 895
 896
 897
 898
 899
 900
 901
 902
 903
 904
 905
 906
 907
 908
 909
 910
 911
 912
 913
 914
 915
 916
 917
 918
 919
 920
 921
 922
 923
 924
 925
 926
 927
 928
 929
 930
 931
 932
 933
 934
 935
 936
 937
 938
 939
 940
 941
 942
 943
 944
 945
 946
 947
 948
 949
 950
 951
 952
 953
 954
 955
 956
 957
 958
 959
 960
 961
 962
 963
 964
 965
 966
 967
 968
 969
 970
 971
 972
 973
 974
 975
 976
 977
 978
 979
 980
 981
 982
 983
 984
 985
 986
 987
 988
 989
 990
 991
 992
 993
 994
 995
 996
 997
 998
 999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
#/*##########################################################################
# Copyright (C) 2004-2012 European Synchrotron Radiation Facility
#
# This file is part of the PyMCA X-ray Fluorescence Toolkit developed at
# the ESRF by the Beamline Instrumentation Software Support (BLISS) group.
#
# This toolkit is free software; you can redistribute it and/or modify it 
# under the terms of the GNU General Public License as published by the Free
# Software Foundation; either version 2 of the License, or (at your option) 
# any later version.
#
# PyMCA is distributed in the hope that it will be useful, but WITHOUT ANY
# WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
# FOR A PARTICULAR PURPOSE.  See the GNU General Public License for more
# details.
#
# You should have received a copy of the GNU General Public License along with
# PyMCA; if not, write to the Free Software Foundation, Inc., 59 Temple Place,
# Suite 330, Boston, MA 02111-1307, USA.
#
# PyMCA follows the dual licensing model of Trolltech's Qt and Riverbank's PyQt
# and cannot be used as a free plugin for a non-free program. 
#
# Please contact the ESRF industrial unit (industry@esrf.fr) if this license 
# is a problem for you.
#############################################################################*/
"""
    EdfFile.py
    Generic class for Edf files manipulation.    

    Interface:
    ===========================
    class EdfFile:          
        __init__(self,FileName)	
        GetNumImages(self)
        def GetData(self,Index, DataType="",Pos=None,Size=None):
        GetPixel(self,Index,Position)
        GetHeader(self,Index)
        GetStaticHeader(self,Index)
        WriteImage (self,Header,Data,Append=1,DataType="",WriteAsUnsigened=0,ByteOrder="")


    Edf format assumptions:
    ===========================
    The following details were assumed for this implementation:
    - Each Edf file contains a certain number of data blocks.
    - Each data block represents data stored in an one, two or three-dimensional array.
    - Each data block contains a header section, written in ASCII, and a data section of
      binary information.
    - The size of the header section in bytes is a multiple of 1024. The header is
      padded with spaces (0x20). If the header is not padded to a multiple of 1024,
      the file is recognized, but the output is always made in this format.
    - The header section starts by '{' and finishes by '}'. It is composed by several
      pairs 'keyword = value;'. The keywords are case insensitive, but the values are case
      sensitive. Each pair is put in a new line (they are separeted by 0x0A). In the
      end of each line, a semicolon (;) separes the pair of a comment, not interpreted.
      Exemple:
        {
        ; Exemple Header
        HeaderID = EH:000001:000000:000000    ; automatically generated
        ByteOrder = LowByteFirst              ; 
        DataType = FloatValue                 ; 4 bytes per pixel
        Size = 4000000                        ; size of data section
        Dim_1= 1000                           ; x coordinates
        Dim_2 = 1000                          ; y coordinates
        
        (padded with spaces to complete 1024 bytes)
        }
    - There are some fields in the header that are required for this implementation. If any of
      these is missing, or inconsistent, it will be generated an error:
        Size: Represents size of data block
        Dim_1: size of x coordinates (Dim_2 for 2-dimentional images, and also Dim_3 for 3d)
        DataType
        ByteOrder
    - For the written images, these fields are automatically genereted:
        Size,Dim_1 (Dim_2 and Dim_3, if necessary), Byte Order, DataType, HeaderID and Image
      These fields are called here "static header", and can be retrieved by the method
      GetStaticHeader. Other header components are taken by GetHeader. Both methods returns
      a dictionary in which the key is the keyword of the pair. When writting an image through
      WriteImage method, the Header parameter should not contain the static header information,
      which is automatically generated.
    - The indexing of images through these functions is based just on the 0-based position in
      the file, the header items HeaderID and Image are not considered for referencing the
      images.
    - The data section contais a number of bytes equal to the value of Size keyword. Data
      section is going to be translated into an 1D, 2D or 3D Numpy Array, and accessed
      through GetData method call.      
"""
__author__ = 'Alexandre Gobbo (gobbo@esrf.fr)'
__version__ = '$Revision: 1.6 $'
DEBUG = 0
################################################################################  
import sys
import numpy
import os.path #, tempfile, shutil
try:
    from PyMca import MarCCD
    MARCCD_SUPPORT = True
except ImportError:
    #MarCCD
    MARCCD_SUPPORT = False
try:
    from PyMca import TiffIO
    TIFF_SUPPORT = True
except ImportError:
    #MarCCD
    TIFF_SUPPORT = False
try:
    from PyMca import PilatusCBF
    PILATUS_CBF_SUPPORT = True
except ImportError:
    PILATUS_CBF_SUPPORT = False
try:
    from PyMca.FastEdf import extended_fread
    CAN_USE_FASTEDF = 1
except:
    CAN_USE_FASTEDF = 0

################################################################################
# constants
HEADER_BLOCK_SIZE = 1024
STATIC_HEADER_ELEMENTS = ("HeaderID", "Image", "ByteOrder", "DataType",
                        "Dim_1", "Dim_2", "Dim_3",
                        "Offset_1", "Offset_2", "Offset_3",
                        "Size")
STATIC_HEADER_ELEMENTS_CAPS = ("HEADERID", "IMAGE", "BYTEORDER", "DATATYPE",
                             "DIM_1", "DIM_2", "DIM_3",
                             "OFFSET_1", "OFFSET_2", "OFFSET_3",
                             "SIZE")

LOWER_CASE = 0
UPPER_CASE = 1

KEYS = 1
VALUES = 2

###############################################################################
class Image(object):
    """
    """
    def __init__(self):
        """ Constructor
        """
        self.Header = {}
        self.StaticHeader = {}
        self.HeaderPosition = 0
        self.DataPosition = 0
        self.Size = 0
        self.NumDim = 1
        self.Dim1 = 0
        self.Dim2 = 0
        self.Dim3 = 0
        self.DataType = ""
        #for i in STATIC_HEADER_ELEMENTS: self.StaticHeader[i]=""

################################################################################

class  EdfFile(object):
    """
    """
    ############################################################################
    #Interface
    def __init__(self, FileName, access=None, fastedf=None):
        """ Constructor
        
        @param  FileName:   Name of the file (either existing or to be created)
        @type FileName: string
        @param access: access mode "r" for reading (the file should exist) or 
                                   "w" for writing (if the file does not exist, it does not matter).
        @type access: string
        @type fastedf= True to use the fastedf module
        @param fastedf= boolean
        """
        self.Images = []
        self.NumImages = 0
        self.FileName = FileName
        self.File = 0
        if fastedf is None:fastedf = 0
        self.fastedf = fastedf
        self.ADSC = False
        self.MARCCD = False
        self.TIFF = False
        self.PILATUS_CBF = False
        self.SPE = False
        if sys.byteorder == "big": self.SysByteOrder = "HighByteFirst"
        else: self.SysByteOrder = "LowByteFirst"
        if access is not None:
            if access[0].upper() == "R":
                if not os.path.isfile(self.FileName):
                    raise IOError("File %s not found" % FileName)
            if 'b' not in access:
                access += 'b'
        try:
            if not os.path.isfile(self.FileName):
                #write access
                if access is None:
                    #allow writing and reading
                    access = "ab+"
                    self.File = open(self.FileName, access)
                    self.File.seek(0, 0)
                    return
                if 'b' not in access:
                    access += 'b'
                self.File = open(self.FileName, access)
                return
            else:
                if access is None:
                    if (os.access(self.FileName, os.W_OK)):
                        access = "r+b"
                    else:
                        access = "rb"
                self.File = open(self.FileName, access)
                self.File.seek(0, 0)
                twoChars = self.File.read(2)
                tiff = False
                if sys.version < '3.0':
                    if twoChars in ["II", "MM"]:
                        tiff = True
                elif twoChars in [eval('b"II"'), eval('b"MM"')]:
                        tiff = True
                if tiff:
                    fileExtension = os.path.splitext(self.FileName)[-1]
                    if fileExtension.lower() in [".tif", ".tiff"] or\
                       sys.version > '2.9':
                        if not TIFF_SUPPORT:
                            raise IOError("TIFF support not implemented")
                        else:
                            self.TIFF = True
                    elif not MARCCD_SUPPORT:
                        if not TIFF_SUPPORT:
                            raise IOError("MarCCD support not implemented")
                        else:
                            self.TIFF = True
                    else:
                        self.MARCCD = True
                if os.path.basename(FileName).upper().endswith('.CBF'):
                    if not PILATUS_CBF_SUPPORT:
                        raise IOError("CBF support not implemented")
                    if twoChars[0] != "{":
                        self.PILATUS_CBF = True
                elif os.path.basename(FileName).upper().endswith('.SPE'):
                    if twoChars[0] != "$":
                        self.SPE = True
        except:
            try:
                self.File.close()
            except:
                pass
            raise IOError("EdfFile: Error opening file")

        self.File.seek(0, 0)
        if self.TIFF:
            self._wrapTIFF()
            self.File.close()
            return
        if self.MARCCD:
            self._wrapMarCCD()
            self.File.close()
            return
        if self.PILATUS_CBF:
            self._wrapPilatusCBF()
            self.File.close()
            return
        if self.SPE:
            self._wrapSPE()
            self.File.close()
            return

        Index = 0
        line = self.File.readline()
        selectedLines = [""]
        if sys.version > '2.6':
            selectedLines.append(eval('b""'))
        while line not in selectedLines:
            #decode to make sure I have character string
            #str to make sure python 2.x sees it as string and not unicode
            if sys.version < '3.0':
                if type(line) != type(str("")):
                    line = "%s" % line
            else:
                try:
                    line = str(line.decode())
                except UnicodeDecodeError:
                    try:
                        line = str(line.decode('utf-8'))
                    except UnicodeDecodeError:
                        try:
                            line = str(line.decode('latin-1'))
                        except UnicodeDecodeError:
                            line = "%s" % line
            if (line.count("{\n") >= 1) or (line.count("{\r\n") >= 1):
                Index = self.NumImages
                self.NumImages = self.NumImages + 1
                self.Images.append(Image())
#            Position = self.File.tell()

            if line.count("=") >= 1:
                listItems = line.split("=", 1)
                typeItem = listItems[0].strip()
                listItems = listItems[1].split(";", 1)
                valueItem = listItems[0].strip()
                if (typeItem == "HEADER_BYTES") and (Index == 0):
                    self.ADSC = True
                    break

                #if typeItem in self.Images[Index].StaticHeader.keys():          
                if typeItem.upper() in STATIC_HEADER_ELEMENTS_CAPS:
                    self.Images[Index].StaticHeader[typeItem] = valueItem
                else:
                    self.Images[Index].Header[typeItem] = valueItem
            if (line.count("}\n") >= 1) or (line.count("}\r") >= 1):
                #for i in STATIC_HEADER_ELEMENTS_CAPS:
                #    if self.Images[Index].StaticHeader[i]=="":
                #        raise "Bad File Format"
                self.Images[Index].DataPosition = self.File.tell()
                #self.File.seek(int(self.Images[Index].StaticHeader["Size"]), 1)
                StaticPar = SetDictCase(self.Images[Index].StaticHeader, UPPER_CASE, KEYS)
                if "SIZE" in StaticPar.keys():
                    self.Images[Index].Size = int(StaticPar["SIZE"])
                    if self.Images[Index].Size <= 0:
                        self.NumImages = Index
                        line = self.File.readline()
                        continue
                else:
                    raise TypeError("EdfFile: Image doesn't have size information")
                if "DIM_1" in StaticPar.keys():
                    self.Images[Index].Dim1 = int(StaticPar["DIM_1"])
                    self.Images[Index].Offset1 = int(\
                                            StaticPar.get("Offset_1", "0"))
                else:
                    raise TypeError("EdfFile: Image doesn't have dimension information")
                if "DIM_2" in StaticPar.keys():
                    self.Images[Index].NumDim = 2
                    self.Images[Index].Dim2 = int(StaticPar["DIM_2"])
                    self.Images[Index].Offset2 = int(\
                                            StaticPar.get("Offset_2", "0"))
                if "DIM_3" in StaticPar.keys():
                    self.Images[Index].NumDim = 3
                    self.Images[Index].Dim3 = int(StaticPar["DIM_3"])
                    self.Images[Index].Offset3 = int(\
                                            StaticPar.get("Offset_3", "0"))
                if "DATATYPE" in StaticPar.keys():
                    self.Images[Index].DataType = StaticPar["DATATYPE"]
                else:
                    raise TypeError("EdfFile: Image doesn't have datatype information")
                if "BYTEORDER" in StaticPar.keys():
                    self.Images[Index].ByteOrder = StaticPar["BYTEORDER"]
                else:
                    raise TypeError("EdfFile: Image doesn't have byteorder information")



                self.File.seek(self.Images[Index].Size, 1)

            line = self.File.readline()

        if self.ADSC:
            self.File.seek(0, 0)
            self.NumImages = 1
            #this is a bad implementation of fabio adscimage
            #please take a look at the fabio module of fable at sourceforge
            infile = self.File
            header_keys = []
            header = {}
            try:
                """ read an adsc header """
                line = infile.readline()
                bytesread = len(line)
                while '}' not in line:
                    if '=' in line:
                        (key, val) = line.split('=')
                        header_keys.append(key.strip())
                        header[key.strip()] = val.strip(' ;\n')
                    line = infile.readline()
                    bytesread = bytesread + len(line)
            except:
                raise Exception("Error processing adsc header")
            # banned by bzip/gzip???
            try:
                infile.seek(int(header['HEADER_BYTES']), 0)
            except TypeError:
                # Gzipped does not allow a seek and read header is not
                # promising to stop in the right place
                infile.close()
                infile = self._open(fname, "rb")
                infile.read(int(header['HEADER_BYTES']))
            binary = infile.read()
            infile.close()

            #now read the data into the array
            self.Images[Index].Dim1 = int(header['SIZE1'])
            self.Images[Index].Dim2 = int(header['SIZE2'])
            self.Images[Index].NumDim = 2
            self.Images[Index].DataType = 'UnsignedShort'
            try:
                self.__data = numpy.reshape(
                    numpy.fromstring(binary, numpy.uint16),
                    (self.Images[Index].Dim2, self.Images[Index].Dim1))
            except ValueError:
                raise IOError('Size spec in ADSC-header does not match ' + \
                    'size of image data field')
            if 'little' in header['BYTE_ORDER']:
                self.Images[Index].ByteOrder = 'LowByteFirst'
            else:
                self.Images[Index].ByteOrder = 'HighByteFirst'
            if self.SysByteOrder.upper() != self.Images[Index].ByteOrder.upper():
                self.__data = self.__data.byteswap()
                self.Images[Index].ByteOrder = self.SysByteOrder

            self.Images[Index].StaticHeader['Dim_1'] = self.Images[Index].Dim1
            self.Images[Index].StaticHeader['Dim_2'] = self.Images[Index].Dim2
            self.Images[Index].StaticHeader['Offset_1'] = 0
            self.Images[Index].StaticHeader['Offset_2'] = 0
            self.Images[Index].StaticHeader['DataType'] = self.Images[Index].DataType

        self.__makeSureFileIsClosed()

    def _wrapTIFF(self):
        self._wrappedInstance = TiffIO.TiffIO(self.File, cache_length = 0, mono_output=True)
        self.NumImages = self._wrappedInstance.getNumberOfImages()
        if self.NumImages < 1:
            return
        info0 = self._wrappedInstance.getInfo(0)
        #for the time being I am going to assume all the images have the same shape
        data = self._wrappedInstance.getData(0)

        for Index in range(self.NumImages):
            info = self._wrappedInstance.getInfo(Index)
            self.Images.append(Image())
            self.Images[Index].Dim1 = info['nRows']
            self.Images[Index].Dim2 = info['nColumns']
            self.Images[Index].NumDim = 2
            self.Images[Index].DataType = self.__GetDefaultEdfType__(data.dtype)
            self.Images[Index].StaticHeader['Dim_1'] = self.Images[Index].Dim1
            self.Images[Index].StaticHeader['Dim_2'] = self.Images[Index].Dim2
            self.Images[Index].StaticHeader['Offset_1'] = 0
            self.Images[Index].StaticHeader['Offset_2'] = 0
            self.Images[Index].StaticHeader['DataType'] = self.Images[Index].DataType
            self.Images[Index].Header.update(info)

    def _wrapMarCCD(self):
        mccd = MarCCD.MarCCD(self.File)
        self.NumImages = 1
        self.__data = mccd.getData()
        self.__info = mccd.getInfo()
        self.Images.append(Image())
        Index = 0
        self.Images[Index].Dim1 = self.__data.shape[0]
        self.Images[Index].Dim2 = self.__data.shape[1]
        self.Images[Index].NumDim = 2
        if self.__data.dtype == numpy.uint8:
            self.Images[Index].DataType = 'UnsignedByte'
        elif self.__data.dtype == numpy.uint16:
            self.Images[Index].DataType = 'UnsignedShort'
        else:
            self.Images[Index].DataType = 'UnsignedInteger'
        self.Images[Index].StaticHeader['Dim_1'] = self.Images[Index].Dim1
        self.Images[Index].StaticHeader['Dim_2'] = self.Images[Index].Dim2
        self.Images[Index].StaticHeader['Offset_1'] = 0
        self.Images[Index].StaticHeader['Offset_2'] = 0
        self.Images[Index].StaticHeader['DataType'] = self.Images[Index].DataType
        self.Images[Index].Header.update(self.__info)

    def _wrapPilatusCBF(self):
        mccd = PilatusCBF.PilatusCBF(self.File)
        self.NumImages = 1
        self.__data = mccd.getData()
        self.__info = mccd.getInfo()
        self.Images.append(Image())
        Index = 0
        self.Images[Index].Dim1 = self.__data.shape[0]
        self.Images[Index].Dim2 = self.__data.shape[1]
        self.Images[Index].NumDim = 2
        if self.__data.dtype == numpy.uint8:
            self.Images[Index].DataType = 'UnsignedByte'
        elif self.__data.dtype == numpy.uint16:
            self.Images[Index].DataType = 'UnsignedShort'
        else:
            self.Images[Index].DataType = 'UnsignedInteger'
        self.Images[Index].StaticHeader['Dim_1'] = self.Images[Index].Dim1
        self.Images[Index].StaticHeader['Dim_2'] = self.Images[Index].Dim2
        self.Images[Index].StaticHeader['Offset_1'] = 0
        self.Images[Index].StaticHeader['Offset_2'] = 0
        self.Images[Index].StaticHeader['DataType'] = self.Images[Index].DataType
        self.Images[Index].Header.update(self.__info)

    def _wrapSPE(self):
        if 0 and sys.version < '3.0':
            self.File.seek(42)
            xdim = numpy.int64(numpy.fromfile(self.File, numpy.int16, 1)[0])
            self.File.seek(656)
            ydim = numpy.int64(numpy.fromfile(self.File, numpy.int16, 1))
            self.File.seek(4100)
            self.__data = numpy.fromfile(self.File, numpy.uint16, int(xdim * ydim))
        else:
            import struct
            self.File.seek(0)
            a = self.File.read()
            xdim = numpy.int64(struct.unpack('<h', a[42:44])[0])
            ydim = numpy.int64(struct.unpack('<h', a[656:658])[0])
            fmt = '<%dH' % int(xdim * ydim)
            self.__data = numpy.array(struct.unpack(fmt, a[4100:int(4100+ int(2 * xdim * ydim))])).astype(numpy.uint16)
        self.__data.shape = ydim, xdim
        Index = 0
        self.Images.append(Image())
        self.NumImages = 1
        self.Images[Index].Dim1 = ydim
        self.Images[Index].Dim2 = xdim
        self.Images[Index].NumDim = 2
        self.Images[Index].DataType = 'UnsignedShort'
        self.Images[Index].ByteOrder = 'LowByteFirst'
        if self.SysByteOrder.upper() != self.Images[Index].ByteOrder.upper():
            self.__data = self.__data.byteswap()
        self.Images[Index].StaticHeader['Dim_1'] = self.Images[Index].Dim1
        self.Images[Index].StaticHeader['Dim_2'] = self.Images[Index].Dim2
        self.Images[Index].StaticHeader['Offset_1'] = 0
        self.Images[Index].StaticHeader['Offset_2'] = 0
        self.Images[Index].StaticHeader['DataType'] = self.Images[Index].DataType

    def GetNumImages(self):
        """ Returns number of images of the object (and associated file)
        """
        return self.NumImages

    def GetData(self, *var, **kw):
        try:
            self.__makeSureFileIsOpen()
            return self._GetData(*var, **kw)
        finally:
            self.__makeSureFileIsClosed()
            
    def _GetData(self, Index, DataType="", Pos=None, Size=None):
        """ Returns numpy array with image data
            Index:          The zero-based index of the image in the file
            DataType:       The edf type of the array to be returnd
                            If ommited, it is used the default one for the type
                            indicated in the image header
                            Attention to the absence of UnsignedShort,
                            UnsignedInteger and UnsignedLong types in
                            Numpy Python
                            Default relation between Edf types and NumPy's typecodes:
                                SignedByte          int8   b
                                UnsignedByte        uint8  B       
                                SignedShort         int16  h
                                UnsignedShort       uint16 H
                                SignedInteger       int32  i
                                UnsignedInteger     uint32 I
                                SignedLong          int32  i
                                UnsignedLong        uint32 I
                                Signed64            int64  (l in 64bit, q in 32 bit)
                                Unsigned64          uint64 (L in 64bit, Q in 32 bit) 
                                FloatValue          float32 f
                                DoubleValue         float64 d
            Pos:            Tuple (x) or (x,y) or (x,y,z) that indicates the begining
                            of data to be read. If ommited, set to the origin (0),
                            (0,0) or (0,0,0)
            Size:           Tuple, size of the data to be returned as x) or (x,y) or
                            (x,y,z) if ommited, is the distance from Pos to the end.

            If Pos and Size not mentioned, returns the whole data.                         
        """
        fastedf = self.fastedf
        if Index < 0 or Index >= self.NumImages:
            raise ValueError("EdfFile: Index out of limit")
        if fastedf is None:fastedf = 0
        if Pos is None and Size is None:
            if self.ADSC or self.MARCCD or self.PILATUS_CBF or self.SPE:
                return self.__data
            elif self.TIFF:
                data = self._wrappedInstance.getData(Index)
                return data
            else:
                self.File.seek(self.Images[Index].DataPosition, 0)
                datatype = self.__GetDefaultNumpyType__(self.Images[Index].DataType, index=Index)
                try:
                    datasize = self.__GetSizeNumpyType__(datatype)
                except TypeError:
                    print("What is the meaning of this error?")
                    datasize = 8
                if self.Images[Index].NumDim == 3:
                    sizeToRead = self.Images[Index].Dim1 * \
                                 self.Images[Index].Dim2 * \
                                 self.Images[Index].Dim3 * datasize
                    Data = numpy.fromstring(self.File.read(sizeToRead),
                                datatype)
                    Data = numpy.reshape(Data, (self.Images[Index].Dim3, self.Images[Index].Dim2, self.Images[Index].Dim1))
                elif self.Images[Index].NumDim == 2:
                    sizeToRead = self.Images[Index].Dim1 * \
                                 self.Images[Index].Dim2 * datasize
                    Data = numpy.fromstring(self.File.read(sizeToRead),
                                datatype)
                    #print "datatype = ",datatype
                    #print "Data.type = ", Data.dtype.char
                    #print "self.Images[Index].DataType ", self.Images[Index].DataType
                    #print "Data.shape",Data.shape
                    #print "datasize = ",datasize
                    #print "sizeToRead ",sizeToRead 
                    #print "lenData = ", len(Data)
                    Data = numpy.reshape(Data, (self.Images[Index].Dim2, self.Images[Index].Dim1))
                elif self.Images[Index].NumDim == 1:
                    sizeToRead = self.Images[Index].Dim1 * datasize
                    Data = numpy.fromstring(self.File.read(sizeToRead),
                                datatype)
        elif self.ADSC or self.MARCCD or self.PILATUS_CBF or self.SPE:
            return self.__data[Pos[1]:(Pos[1] + Size[1]),
                               Pos[0]:(Pos[0] + Size[0])]
        elif self.TIFF:
            data = self._wrappedInstance.getData(Index)
            return data[Pos[1]:(Pos[1] + Size[1]),
                               Pos[0]:(Pos[0] + Size[0])]
        elif fastedf and CAN_USE_FASTEDF:
            type = self.__GetDefaultNumpyType__(self.Images[Index].DataType, index=Index)
            size_pixel = self.__GetSizeNumpyType__(type)
            Data = numpy.array([], type)
            if self.Images[Index].NumDim == 1:
                if Pos == None: Pos = (0,)
                if Size == None: Size = (0,)
                sizex = self.Images[Index].Dim1
                Size = list(Size)
                if Size[0] == 0:Size[0] = sizex - Pos[0]
                self.File.seek((Pos[0] * size_pixel) + self.Images[Index].DataPosition, 0)
                Data = numpy.fromstring(self.File.read(Size[0] * size_pixel), type)
            elif self.Images[Index].NumDim == 2:
                if Pos == None: Pos = (0, 0)
                if Size == None: Size = (0, 0)
                Size = list(Size)
                sizex, sizey = self.Images[Index].Dim1, self.Images[Index].Dim2
                if Size[0] == 0:Size[0] = sizex - Pos[0]
                if Size[1] == 0:Size[1] = sizey - Pos[1]
                Data = numpy.zeros([Size[1], Size[0]], type)
                self.File.seek((((Pos[1] * sizex) + Pos[0]) * size_pixel) + self.Images[Index].DataPosition, 0)
                extended_fread(Data, Size[0] * size_pixel , numpy.array([Size[1]]),
                               numpy.array([sizex * size_pixel]) , self.File)

            elif self.Images[Index].NumDim == 3:
                if Pos == None: Pos = (0, 0, 0)
                if Size == None: Size = (0, 0, 0)
                Size = list(Size)
                sizex, sizey, sizez = self.Images[Index].Dim1, self.Images[Index].Dim2, self.Images[Index].Dim3
                if Size[0] == 0:Size[0] = sizex - Pos[0]
                if Size[1] == 0:Size[1] = sizey - Pos[1]
                if Size[2] == 0:Size[2] = sizez - Pos[2]
                Data = numpy.zeros([Size[2], Size[1], Size[0]], type)
                self.File.seek(((((Pos[2] * sizey + Pos[1]) * sizex) + Pos[0]) * size_pixel) + self.Images[Index].DataPosition, 0)
                extended_fread(Data, Size[0] * size_pixel , numpy.array([Size[2], Size[1]]),
                        numpy.array([ sizey * sizex * size_pixel , sizex * size_pixel]) , self.File)

        else:
            if fastedf:
                print("I could not use fast routines")
            type = self.__GetDefaultNumpyType__(self.Images[Index].DataType, index=Index)
            size_pixel = self.__GetSizeNumpyType__(type)
            Data = numpy.array([], type)
            if self.Images[Index].NumDim == 1:
                if Pos == None: Pos = (0,)
                if Size == None: Size = (0,)
                sizex = self.Images[Index].Dim1
                Size = list(Size)
                if Size[0] == 0:Size[0] = sizex - Pos[0]
                self.File.seek((Pos[0] * size_pixel) + self.Images[Index].DataPosition, 0)
                Data = numpy.fromstring(self.File.read(Size[0] * size_pixel), type)
            elif self.Images[Index].NumDim == 2:
                if Pos == None: Pos = (0, 0)
                if Size == None: Size = (0, 0)
                Size = list(Size)
                sizex, sizey = self.Images[Index].Dim1, self.Images[Index].Dim2
                if Size[0] == 0:Size[0] = sizex - Pos[0]
                if Size[1] == 0:Size[1] = sizey - Pos[1]
                #print len(range(Pos[1],Pos[1]+Size[1])), "LECTURES OF ", Size[0], "POINTS"
                #print "sizex = ", sizex, "sizey = ", sizey
                Data = numpy.zeros((Size[1], Size[0]), type)
                dataindex = 0
                for y in range(Pos[1], Pos[1] + Size[1]):
                    self.File.seek((((y * sizex) + Pos[0]) * size_pixel) + self.Images[Index].DataPosition, 0)
                    line = numpy.fromstring(self.File.read(Size[0] * size_pixel), type)
                    Data[dataindex, :] = line[:]
                    #Data=numpy.concatenate((Data,line))
                    dataindex += 1
                #print "DataSize = ",Data.shape
                #print "Requested reshape = ",Size[1],'x',Size[0]
                #Data = numpy.reshape(Data, (Size[1],Size[0]))                            
            elif self.Images[Index].NumDim == 3:
                if Pos == None: Pos = (0, 0, 0)
                if Size == None: Size = (0, 0, 0)
                Size = list(Size)
                sizex, sizey, sizez = self.Images[Index].Dim1, self.Images[Index].Dim2, self.Images[Index].Dim3
                if Size[0] == 0:Size[0] = sizex - Pos[0]
                if Size[1] == 0:Size[1] = sizey - Pos[1]
                if Size[2] == 0:Size[2] = sizez - Pos[2]
                for z in range(Pos[2], Pos[2] + Size[2]):
                    for y in range(Pos[1], Pos[1] + Size[1]):
                        self.File.seek(((((z * sizey + y) * sizex) + Pos[0]) * size_pixel) + self.Images[Index].DataPosition, 0)
                        line = numpy.fromstring(self.File.read(Size[0] * size_pixel), type)
                        Data = numpy.concatenate((Data, line))
                Data = numpy.reshape(Data, (Size[2], Size[1], Size[0]))

        if self.SysByteOrder.upper() != self.Images[Index].ByteOrder.upper():
            Data = Data.byteswap()
        if DataType != "":
            Data = self.__SetDataType__ (Data, DataType)
        return Data



    def GetPixel(self, Index, Position):
        """ Returns double value of the pixel, regardless the format of the array
            Index:      The zero-based index of the image in the file
            Position:   Tuple with the coordinete (x), (x,y) or (x,y,z)
        """
        if Index < 0 or Index >= self.NumImages:
            raise ValueError("EdfFile: Index out of limit")
        if len(Position) != self.Images[Index].NumDim:
            raise ValueError("EdfFile: coordinate with wrong dimension ")

        size_pixel = self.__GetSizeNumpyType__(self.__GetDefaultNumpyType__(self.Images[Index].DataType), index=Index)
        offset = Position[0] * size_pixel
        if self.Images[Index].NumDim > 1:
            size_row = size_pixel * self.Images[Index].Dim1
            offset = offset + (Position[1] * size_row)
            if self.Images[Index].NumDim == 3:
                size_img = size_row * self.Images[Index].Dim2
                offset = offset + (Position[2] * size_img)
        self.File.seek(self.Images[Index].DataPosition + offset, 0)
        Data = numpy.fromstring(self.File.read(size_pixel), self.__GetDefaultNumpyType__(self.Images[Index].DataType, index=Index))
        if self.SysByteOrder.upper() != self.Images[Index].ByteOrder.upper():
            Data = Data.byteswap()
        Data = self.__SetDataType__ (Data, "DoubleValue")
        return Data[0]


    def GetHeader(self, Index):
        """ Returns dictionary with image header fields.
            Does not include the basic fields (static) defined by data shape, 
            type and file position. These are get with GetStaticHeader
            method.
            Index:          The zero-based index of the image in the file
        """
        if Index < 0 or Index >= self.NumImages:
            raise ValueError("Index out of limit")
        #return self.Images[Index].Header
        ret = {}
        for i in self.Images[Index].Header.keys():
            ret[i] = self.Images[Index].Header[i]
        return ret


    def GetStaticHeader(self, Index):
        """ Returns dictionary with static parameters
            Data format and file position dependent information
            (dim1,dim2,size,datatype,byteorder,headerId,Image)
            Index:          The zero-based index of the image in the file
        """
        if Index < 0 or Index >= self.NumImages:
            raise ValueError("Index out of limit")
        #return self.Images[Index].StaticHeader
        ret = {}
        for i in self.Images[Index].StaticHeader.keys():
            ret[i] = self.Images[Index].StaticHeader[i]
        return ret

    def WriteImage(self, *var, **kw):
        try:
            self.__makeSureFileIsOpen()
            return self._WriteImage(*var, **kw)
        finally:
            self.__makeSureFileIsClosed()
    
    def _WriteImage (self, Header, Data, Append=1, DataType="", ByteOrder=""):
        """ Writes image to the file. 
            Header:         Dictionary containing the non-static header
                            information (static information is generated
                            according to position of image and data format
            Append:         If equals to 0, overwrites the file. Otherwise, appends
                            to the end of the file
            DataType:       The data type to be saved to the file:
                                SignedByte          
                                UnsignedByte               
                                SignedShort         
                                UnsignedShort       
                                SignedInteger       
                                UnsignedInteger     
                                SignedLong          
                                UnsignedLong        
                                FloatValue          
                                DoubleValue         
                            Default: according to Data array typecode:
                                    1:  SignedByte
                                    b:  UnsignedByte
                                    s:  SignedShort       
				    w:  UnsignedShort
                                    i:  SignedInteger
                                    l:  SignedLong          
				    u:  UnsignedLong
                                    f:  FloatValue       
                                    d:  DoubleValue
            ByteOrder:      Byte order of the data in file:
                                HighByteFirst
                                LowByteFirst
                            Default: system's byte order
        """
        if Append == 0:
            self.File.truncate(0)
            self.Images = []
            self.NumImages = 0
        Index = self.NumImages
        self.NumImages = self.NumImages + 1
        self.Images.append(Image())

        #self.Images[Index].StaticHeader["Dim_1"] = "%d" % Data.shape[1]
        #self.Images[Index].StaticHeader["Dim_2"] = "%d" % Data.shape[0]
        if len(Data.shape) == 1:
            self.Images[Index].Dim1 = Data.shape[0]
            self.Images[Index].StaticHeader["Dim_1"] = "%d" % self.Images[Index].Dim1
            self.Images[Index].Size = (Data.shape[0] * \
                                     self.__GetSizeNumpyType__(Data.dtype))
        elif len(Data.shape) == 2:
            self.Images[Index].Dim1 = Data.shape[1]
            self.Images[Index].Dim2 = Data.shape[0]
            self.Images[Index].StaticHeader["Dim_1"] = "%d" % self.Images[Index].Dim1
            self.Images[Index].StaticHeader["Dim_2"] = "%d" % self.Images[Index].Dim2
            self.Images[Index].Size = (Data.shape[0] * Data.shape[1] * \
                                     self.__GetSizeNumpyType__(Data.dtype))
            self.Images[Index].NumDim = 2
        elif len(Data.shape) == 3:
            self.Images[Index].Dim1 = Data.shape[2]
            self.Images[Index].Dim2 = Data.shape[1]
            self.Images[Index].Dim3 = Data.shape[0]
            self.Images[Index].StaticHeader["Dim_1"] = "%d" % self.Images[Index].Dim1
            self.Images[Index].StaticHeader["Dim_2"] = "%d" % self.Images[Index].Dim2
            self.Images[Index].StaticHeader["Dim_3"] = "%d" % self.Images[Index].Dim3
            self.Images[Index].Size = (Data.shape[0] * Data.shape[1] * Data.shape[2] * \
                                     self.__GetSizeNumpyType__(Data.dtype))
            self.Images[Index].NumDim = 3
        elif len(Data.shape) > 3:
            raise TypeError("EdfFile: Data dimension not suported")


        if DataType == "":
            self.Images[Index].DataType = self.__GetDefaultEdfType__(Data.dtype)
        else:
            self.Images[Index].DataType = DataType
            Data = self.__SetDataType__ (Data, DataType)

        if ByteOrder == "":
            self.Images[Index].ByteOrder = self.SysByteOrder
        else:
            self.Images[Index].ByteOrder = ByteOrder

        self.Images[Index].StaticHeader["Size"] = "%d" % self.Images[Index].Size
        self.Images[Index].StaticHeader["Image"] = Index + 1
        self.Images[Index].StaticHeader["HeaderID"] = "EH:%06d:000000:000000" % self.Images[Index].StaticHeader["Image"]
        self.Images[Index].StaticHeader["ByteOrder"] = self.Images[Index].ByteOrder
        self.Images[Index].StaticHeader["DataType"] = self.Images[Index].DataType


        self.Images[Index].Header = {}
        self.File.seek(0, 2)
        StrHeader = "{\n"
        for i in STATIC_HEADER_ELEMENTS:
            if i in self.Images[Index].StaticHeader.keys():
                StrHeader = StrHeader + ("%s = %s ;\n" % (i , self.Images[Index].StaticHeader[i]))
        for i in Header.keys():
            StrHeader = StrHeader + ("%s = %s ;\n" % (i, Header[i]))
            self.Images[Index].Header[i] = Header[i]
        newsize = (((len(StrHeader) + 1) / HEADER_BLOCK_SIZE) + 1) * HEADER_BLOCK_SIZE - 2
        newsize = int(newsize)
        StrHeader = StrHeader.ljust(newsize)
        StrHeader = StrHeader + "}\n"

        self.Images[Index].HeaderPosition = self.File.tell()
        self.File.write(StrHeader.encode())
        self.Images[Index].DataPosition = self.File.tell()

        #if self.Images[Index].StaticHeader["ByteOrder"] != self.SysByteOrder:
        if self.Images[Index].ByteOrder.upper() != self.SysByteOrder.upper():
            self.File.write((Data.byteswap()).tostring())
        else:
            self.File.write(Data.tostring())



    ############################################################################
    #Internal Methods

    def __makeSureFileIsOpen(self):
        if DEBUG:
            print("Making sure file is open")
        if self.ADSC or self.MARCCD or self.PILATUS_CBF or self.SPE:
            if DEBUG:
                print("Special case. Image is buffered")
            return
        if self.File in [0, None]:
            if DEBUG:
                print("File is None")
        elif self.File.closed:
            if DEBUG:
                print("Reopening closed file")
            accessMode = self.File.mode
            fileName = self.File.name
            newFile = open(fileName, accessMode)
            self.File  = newFile
        return

    def __makeSureFileIsClosed(self):
        if DEBUG:
            print("Making sure file is closed")
        if self.ADSC or self.MARCCD or self.PILATUS_CBF or self.SPE:
            if DEBUG:
                print("Special case. Image is buffered")
            return
        if self.File in [0, None]:
            if DEBUG:
                print("File is None")
        elif not self.File.closed:
            if DEBUG:
                print("Closing file")
            self.File.close()
        return

        
    def __GetDefaultNumpyType__(self, EdfType, index=None):
        """ Internal method: returns NumPy type according to Edf type
        """
        return self.GetDefaultNumpyType(EdfType, index)

    def __GetDefaultEdfType__(self, NumpyType):
        """ Internal method: returns Edf type according Numpy type
        """
        if   NumpyType in ["b", numpy.int8]:            return "SignedByte"
        elif NumpyType in ["B", numpy.uint8]:            return "UnsignedByte"
        elif NumpyType in ["h", numpy.int16]:            return "SignedShort"
        elif NumpyType in ["H", numpy.uint16]:            return "UnsignedShort"
        elif NumpyType in ["i", numpy.int32]:            return "SignedInteger"
        elif NumpyType in ["I", numpy.uint32]:            return "UnsignedInteger"
        elif NumpyType == "l":
            if sys.platform == 'linux2':
                return "Signed64"
            else:
                return "SignedLong"
        elif NumpyType == "L":
            if sys.platform == 'linux2':
                return "Unsigned64"
            else:
                return "UnsignedLong"
        elif NumpyType == numpy.int64:
            return "Signed64"
        elif NumpyType == numpy.uint64:
            return "Unsigned64"
        elif NumpyType in ["f", numpy.float32]:
            return "FloatValue"
        elif NumpyType in ["d", numpy.float64]:
            return "DoubleValue"
        else:
            raise TypeError("unknown NumpyType %s" % NumpyType)


    def __GetSizeNumpyType__(self, NumpyType):
        """ Internal method: returns size of NumPy's Array Types
        """
        if   NumpyType in ["b", numpy.int8]:  return 1
        elif NumpyType in ["B", numpy.uint8]:  return 1
        elif NumpyType in ["h", numpy.int16]:  return 2
        elif NumpyType in ["H", numpy.uint16]:  return 2
        elif NumpyType in ["i", numpy.int32]:  return 4
        elif NumpyType in ["I", numpy.uint32]:  return 4
        elif NumpyType == "l":
            if sys.platform == 'linux2':
                return 8    #64 bit
            else:
                return 4    #32 bit
        elif NumpyType == "L":
            if sys.platform == 'linux2':
                return 8    #64 bit
            else:
                return 4    #32 bit
        elif NumpyType in ["f", numpy.float32]: return 4
        elif NumpyType in ["d", numpy.float64]: return 8
        elif NumpyType == "Q":            return 8 #unsigned 64 in 32 bit
        elif NumpyType == "q":            return 8 #signed 64 in 32 bit
        elif NumpyType == numpy.uint64:   return 8
        elif NumpyType == numpy.int64:    return 8
        else:
            raise TypeError("unknown NumpyType %s" % NumpyType)


    def __SetDataType__ (self, Array, DataType):
        """ Internal method: array type convertion
        """
        # AVOID problems not using FromEdfType= Array.dtype.char
        FromEdfType = Array.dtype
        ToEdfType = self.__GetDefaultNumpyType__(DataType)
        if ToEdfType != FromEdfType:
            aux = Array.astype(self.__GetDefaultNumpyType__(DataType))
            return aux
        return Array

    def __del__(self):
        try:
            self.__makeSureFileIsClosed()
        except:
            pass

    def GetDefaultNumpyType(self, EdfType, index=None):
        """ Returns NumPy type according Edf type
        """
        if index is None:return GetDefaultNumpyType(EdfType)
        EdfType = EdfType.upper()
        if EdfType in ['SIGNED64']  :return numpy.int64
        if EdfType in ['UNSIGNED64']:return numpy.uint64
        if EdfType in ["SIGNEDLONG", "UNSIGNEDLONG"]:
            dim1 = 1
            dim2 = 1
            dim3 = 1
            if hasattr(self.Images[index], "Dim1"):
                dim1 = self.Images[index].Dim1
                if hasattr(self.Images[index], "Dim2"):
                    dim2 = self.Images[index].Dim2
                    if dim2 <= 0: dim2 = 1
                    if hasattr(self.Images[index], "Dim3"):
                        dim3 = self.Images[index].Dim3
                        if dim3 <= 0: dim3 = 1
                if hasattr(self.Images[index], "Size"):
                    size = self.Images[index].Size
                    if size / (dim1 * dim2 * dim3) == 8:
                        if EdfType == "UNSIGNEDLONG":
                            return numpy.uint64
                        else:
                            return numpy.int64
            if EdfType == "UNSIGNEDLONG":
                return numpy.uint32
            else:
                return numpy.int32

        return GetDefaultNumpyType(EdfType)


def GetDefaultNumpyType(EdfType):
    """ Returns NumPy type according Edf type
    """
    EdfType = EdfType.upper()
    if   EdfType == "SIGNEDBYTE":       return numpy.int8   # "b"
    elif EdfType == "UNSIGNEDBYTE":     return numpy.uint8  # "B"       
    elif EdfType == "SIGNEDSHORT":      return numpy.int16  # "h"
    elif EdfType == "UNSIGNEDSHORT":    return numpy.uint16 # "H"
    elif EdfType == "SIGNEDINTEGER":    return numpy.int32  # "i"
    elif EdfType == "UNSIGNEDINTEGER":  return numpy.uint32 # "I"
    elif EdfType == "SIGNEDLONG":       return numpy.int32  # "i" #ESRF acquisition is made in 32bit
    elif EdfType == "UNSIGNEDLONG":     return numpy.uint32 # "I" #ESRF acquisition is made in 32bit
    elif EdfType == "SIGNED64":         return numpy.int64  # "l"
    elif EdfType == "UNSIGNED64":       return numpy.uint64 # "L"
    elif EdfType == "FLOATVALUE":       return numpy.float32 # "f"
    elif EdfType == "FLOAT":            return numpy.float32 # "f"
    elif EdfType == "DOUBLEVALUE":      return numpy.float64 # "d"
    else: raise TypeError("unknown EdfType %s" % EdfType)


def SetDictCase(Dict, Case, Flag):
    """ Returns dictionary with keys and/or values converted into upper or lowercase
        Dict:   input dictionary
        Case:   LOWER_CASE, UPPER_CASE
        Flag:   KEYS, VALUES or KEYS | VALUES        
    """
    newdict = {}
    for i in Dict.keys():
        newkey = i
        newvalue = Dict[i]
        if Flag & KEYS:
            if Case == LOWER_CASE:  newkey = newkey.lower()
            else:                   newkey = newkey.upper()
        if Flag & VALUES:
            if Case == LOWER_CASE:  newvalue = newvalue.lower()
            else:                   newvalue = newvalue.upper()
        newdict[newkey] = newvalue
    return newdict


def GetRegion(Arr, Pos, Size):
    """Returns array with refion of Arr.
       Arr must be 1d, 2d or 3d
       Pos and Size are tuples in the format (x) or (x,y) or (x,y,z)
       Both parameters must have the same size as the dimention of Arr
    """
    Dim = len(Arr.shape)
    if len(Pos) != Dim:  return None
    if len(Size) != Dim: return None

    if (Dim == 1):
        SizeX = Size[0]
        if SizeX == 0: SizeX = Arr.shape[0] - Pos[0]
        ArrRet = numpy.take(Arr, range(Pos[0], Pos[0] + SizeX))
    elif (Dim == 2):
        SizeX = Size[0]
        SizeY = Size[1]
        if SizeX == 0: SizeX = Arr.shape[1] - Pos[0]
        if SizeY == 0: SizeY = Arr.shape[0] - Pos[1]
        ArrRet = numpy.take(Arr, range(Pos[1], Pos[1] + SizeY))
        ArrRet = numpy.take(ArrRet, range(Pos[0], Pos[0] + SizeX), 1)
    elif (Dim == 3):
        SizeX = Size[0]
        SizeY = Size[1]
        SizeZ = Size[2]
        if SizeX == 0: SizeX = Arr.shape[2] - Pos[0]
        if SizeY == 0: SizeX = Arr.shape[1] - Pos[1]
        if SizeZ == 0: SizeZ = Arr.shape[0] - Pos[2]
        ArrRet = numpy.take(Arr, range(Pos[2], Pos[2] + SizeZ))
        ArrRet = numpy.take(ArrRet, range(Pos[1], Pos[1] + SizeY), 1)
        ArrRet = numpy.take(ArrRet, range(Pos[0], Pos[0] + SizeX), 2)
    else:
        ArrRet = None
    return ArrRet

#EXEMPLE CODE:        
if __name__ == "__main__":
    if 1:
#        import os
        a = numpy.zeros((5, 10))
        for i in range(5):
            for j in range(10):
                a[i, j] = 10 * i + j
        edf = EdfFile("armando.edf", access="ab+")
        edf.WriteImage({}, a)
        del edf #force to close the file
        inp = EdfFile("armando.edf")
        b = inp.GetData(0)
        out = EdfFile("armando2.edf")
        out.WriteImage({}, b)
        del out #force to close the file
        inp2 = EdfFile("armando2.edf")
        c = inp2.GetData(0)
        print("A SHAPE = ", a.shape)
        print("B SHAPE = ", b.shape)
        print("C SHAPE = ", c.shape)
        for i in range(5):
            print("A", a[i, :])
            print("B", b[i, :])
            print("C", c[i, :])

        x = numpy.arange(100)
        x.shape = 5, 20
        for item in ["SignedByte", "UnsignedByte",
                     "SignedShort", "UnsignedShort",
                     "SignedLong", "UnsignedLong",
                     "Signed64", "Unsigned64",
                     "FloatValue", "DoubleValue"]:
            fname = item + ".edf"
            if os.path.exists(fname):
                os.remove(fname)
            towrite = EdfFile(fname)
            towrite.WriteImage({}, x, DataType=item, Append=0)
        sys.exit(0)

    #Creates object based on file exe.edf
    exe = EdfFile("images/test_image.edf")
    x = EdfFile("images/test_getdata.edf")
    #Gets unsigned short data, storing in an signed long
    arr = exe.GetData(0, Pos=(100, 200), Size=(200, 400))
    x.WriteImage({}, arr, 0)

    arr = exe.GetData(0, Pos=(100, 200))
    x.WriteImage({}, arr)

    arr = exe.GetData(0, Size=(200, 400))
    x.WriteImage({}, arr)

    arr = exe.GetData(0)
    x.WriteImage({}, arr)

    sys.exit()

    #Creates object based on file exe.edf
    exe = EdfFile("images/.edf")

    #Creates long array , filled with 0xFFFFFFFF(-1)
    la = numpy.zeros((100, 100))
    la = la - 1

    #Creates a short array, filled with 0xFFFF
    sa = numpy.zeros((100, 100))
    sa = sa + 0xFFFF
    sa = sa.astype("s")

    #Writes long array, initializing file (append=0)
    exe.WriteImage({}, la, 0, "")

    #Appends short array with new header items
    exe.WriteImage({'Name': 'Alexandre', 'Date': '16/07/2001'}, sa)

    #Appends short array, in Edf type unsigned
    exe.WriteImage({}, sa, DataType="UnsignedShort")

    #Appends short array, in Edf type unsigned
    exe.WriteImage({}, sa, DataType="UnsignedLong")

    #Appends long array as a double, considering unsigned
    exe.WriteImage({}, la, DataType="DoubleValue", WriteAsUnsigened=1)

    #Gets unsigned short data, storing in an signed long
    ushort = exe.GetData(2, "SignedLong")

    #Makes an operation
    ushort = ushort - 0x10

    #Saves Result as signed long
    exe.WriteImage({}, ushort)

    #Saves in the original format (unsigned short)
    OldHeader = exe.GetStaticHeader(2)
    exe.WriteImage({}, ushort, 1, OldHeader["DataType"])