Source

JsonpDataProxy / example / dataproxy / xlrd / sheet.py

Full commit
   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
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
# -*- coding: cp1252 -*-

##
# <p> Portions copyright Š 2005-2009 Stephen John Machin, Lingfo Pty Ltd</p>
# <p>This module is part of the xlrd package, which is released under a BSD-style licence.</p>
##

# 2009-05-31 SJM Fixed problem with no CODEPAGE record on extremely minimal BIFF2.x 3rd-party file
# 2009-04-27 SJM Integrated on_demand patch by Armando Serrano Lombillo
# 2008-02-09 SJM Excel 2.0: build XFs on the fly from cell attributes
# 2007-12-04 SJM Added support for Excel 2.x (BIFF2) files.
# 2007-10-11 SJM Added missing entry for blank cell type to ctype_text
# 2007-07-11 SJM Allow for BIFF2/3-style FORMAT record in BIFF4/8 file
# 2007-04-22 SJM Remove experimental "trimming" facility.

from biffh import *
from timemachine import *
from struct import unpack
from formula import dump_formula, decompile_formula, rangename2d
from formatting import nearest_colour_index, Format
import time

DEBUG = 0
OBJ_MSO_DEBUG = 0

_WINDOW2_options = (
    # Attribute names and initial values to use in case
    # a WINDOW2 record is not written.
    ("show_formulas", 0),
    ("show_grid_lines", 1),
    ("show_sheet_headers", 1),
    ("panes_are_frozen", 0),
    ("show_zero_values", 1),
    ("automatic_grid_line_colour", 1),
    ("columns_from_right_to_left", 0),
    ("show_outline_symbols", 1),
    ("remove_splits_if_pane_freeze_is_removed", 0),
    ("sheet_selected", 0),
    # "sheet_visible" appears to be merely a clone of "sheet_selected".
    # The real thing is the visibility attribute from the BOUNDSHEET record.
    ("sheet_visible", 0),
    ("show_in_page_break_preview", 0),
    )

##
# <p>Contains the data for one worksheet.</p>
#
# <p>In the cell access functions, "rowx" is a row index, counting from zero, and "colx" is a
# column index, counting from zero.
# Negative values for row/column indexes and slice positions are supported in the expected fashion.</p>
#
# <p>For information about cell types and cell values, refer to the documentation of the Cell class.</p>
#
# <p>WARNING: You don't call this class yourself. You access Sheet objects via the Book object that
# was returned when you called xlrd.open_workbook("myfile.xls").</p>


class Sheet(BaseObject):
    ##
    # Name of sheet.
    name = ''

    ##
    # Number of rows in sheet. A row index is in range(thesheet.nrows).
    nrows = 0

    ##
    # Number of columns in sheet. A column index is in range(thesheet.ncols).
    ncols = 0

    ##
    # The map from a column index to a Colinfo object. Often there is an entry
    # in COLINFO records for all column indexes in range(257).
    # Note that xlrd ignores the entry for the non-existent
    # 257th column. On the other hand, there may be no entry for unused columns.
    # <br /> -- New in version 0.6.1
    colinfo_map = {}

    ##
    # The map from a row index to a Rowinfo object. Note that it is possible
    # to have missing entries -- at least one source of XLS files doesn't
    # bother writing ROW records.
    # <br /> -- New in version 0.6.1
    rowinfo_map = {}

    ##
    # List of address ranges of cells containing column labels.
    # These are set up in Excel by Insert > Name > Labels > Columns.
    # <br> -- New in version 0.6.0
    # <br>How to deconstruct the list:
    # <pre>
    # for crange in thesheet.col_label_ranges:
    #     rlo, rhi, clo, chi = crange
    #     for rx in xrange(rlo, rhi):
    #         for cx in xrange(clo, chi):
    #             print "Column label at (rowx=%d, colx=%d) is %r" \
    #                 (rx, cx, thesheet.cell_value(rx, cx))
    # </pre>
    col_label_ranges = []

    ##
    # List of address ranges of cells containing row labels.
    # For more details, see <i>col_label_ranges</i> above.
    # <br> -- New in version 0.6.0
    row_label_ranges = []

    ##
    # List of address ranges of cells which have been merged.
    # These are set up in Excel by Format > Cells > Alignment, then ticking
    # the "Merge cells" box.
    # <br> -- New in version 0.6.1. Extracted only if open_workbook(..., formatting_info=True)
    # <br>How to deconstruct the list:
    # <pre>
    # for crange in thesheet.merged_cells:
    #     rlo, rhi, clo, chi = crange
    #     for rowx in xrange(rlo, rhi):
    #         for colx in xrange(clo, chi):
    #             # cell (rlo, clo) (the top left one) will carry the data
    #             # and formatting info; the remainder will be recorded as
    #             # blank cells, but a renderer will apply the formatting info
    #             # for the top left cell (e.g. border, pattern) to all cells in
    #             # the range.
    # </pre>
    merged_cells = []

    ##
    # Default column width from DEFCOLWIDTH record, else None.
    # From the OOo docs:<br />
    # """Column width in characters, using the width of the zero character
    # from default font (first FONT record in the file). Excel adds some
    # extra space to the default width, depending on the default font and
    # default font size. The algorithm how to exactly calculate the resulting
    # column width is not known.<br />
    # Example: The default width of 8 set in this record results in a column
    # width of 8.43 using Arial font with a size of 10 points."""<br />
    # For the default hierarchy, refer to the Colinfo class above.
    # <br /> -- New in version 0.6.1
    defcolwidth = None

    ##
    # Default column width from STANDARDWIDTH record, else None.
    # From the OOo docs:<br />
    # """Default width of the columns in 1/256 of the width of the zero
    # character, using default font (first FONT record in the file)."""<br />
    # For the default hierarchy, refer to the Colinfo class above.
    # <br /> -- New in version 0.6.1
    standardwidth = None

    ##
    # Default value to be used for a row if there is
    # no ROW record for that row.
    # From the <i>optional</i> DEFAULTROWHEIGHT record.
    default_row_height = None

    ##
    # Default value to be used for a row if there is
    # no ROW record for that row.
    # From the <i>optional</i> DEFAULTROWHEIGHT record.
    default_row_height_mismatch = None

    ##
    # Default value to be used for a row if there is
    # no ROW record for that row.
    # From the <i>optional</i> DEFAULTROWHEIGHT record.
    default_row_hidden = None

    ##
    # Default value to be used for a row if there is
    # no ROW record for that row.
    # From the <i>optional</i> DEFAULTROWHEIGHT record.
    default_additional_space_above = None

    ##
    # Default value to be used for a row if there is
    # no ROW record for that row.
    # From the <i>optional</i> DEFAULTROWHEIGHT record.
    default_additional_space_below = None

    ##
    # Visibility of the sheet. 0 = visible, 1 = hidden (can be unhidden
    # by user -- Format/Sheet/Unhide), 2 = "very hidden" (can be unhidden
    # only by VBA macro).
    visibility = 0

    ##
    # A 256-element tuple corresponding to the contents of the GCW record for this sheet.
    # If no such record, treat as all bits zero.
    # Applies to BIFF4-7 only. See docs of Colinfo class for discussion.
    gcw = (0, ) * 256

    def __init__(self, book, position, name, number):
        self.book = book
        self.biff_version = book.biff_version
        self._position = position
        self.logfile = book.logfile
        self.pickleable = book.pickleable
        self.dont_use_array = not(array_array and (CAN_PICKLE_ARRAY or not book.pickleable))
        self.name = name
        self.number = number
        self.verbosity = book.verbosity
        self.formatting_info = book.formatting_info
        self._xf_index_to_xl_type_map = book._xf_index_to_xl_type_map
        self.nrows = 0 # actual, including possibly empty cells
        self.ncols = 0
        self._maxdatarowx = -1 # highest rowx containing a non-empty cell
        self._maxdatacolx = -1 # highest colx containing a non-empty cell
        self._dimnrows = 0 # as per DIMENSIONS record
        self._dimncols = 0
        self._cell_values = []
        self._cell_types = []
        self._cell_xf_indexes = []
        self._need_fix_ragged_rows = 0
        self.defcolwidth = None
        self.standardwidth = None
        self.default_row_height = None
        self.default_row_height_mismatch = 0
        self.default_row_hidden = 0
        self.default_additional_space_above = 0
        self.default_additional_space_below = 0
        self.colinfo_map = {}
        self.rowinfo_map = {}
        self.col_label_ranges = []
        self.row_label_ranges = []
        self.merged_cells = []
        self._xf_index_stats = [0, 0, 0, 0]
        self.visibility = book._sheet_visibility[number] # from BOUNDSHEET record
        for attr, defval in _WINDOW2_options:
            setattr(self, attr, defval)
        self.first_visible_rowx = 0
        self.first_visible_colx = 0
        self.gridline_colour_index = 0x40
        self.gridline_colour_rgb = None # pre-BIFF8
        self.cached_page_break_preview_mag_factor = 0
        self.cached_normal_view_mag_factor = 0
        self._ixfe = None # BIFF2 only
        self._cell_attr_to_xfx = {} # BIFF2.0 only

        #### Don't initialise this here, use class attribute initialisation.
        #### self.gcw = (0, ) * 256 ####

        if self.biff_version >= 80:
            self.utter_max_rows = 65536
        else:
            self.utter_max_rows = 16384
        self.utter_max_cols = 256

    ##
    # Cell object in the given row and column.
    def cell(self, rowx, colx):
        if self.formatting_info:
            xfx = self.cell_xf_index(rowx, colx)
        else:
            xfx = None
        return Cell(
            self._cell_types[rowx][colx],
            self._cell_values[rowx][colx],
            xfx,
            )

    ##
    # Value of the cell in the given row and column.
    def cell_value(self, rowx, colx):
        return self._cell_values[rowx][colx]

    ##
    # Type of the cell in the given row and column.
    # Refer to the documentation of the Cell class.
    def cell_type(self, rowx, colx):
        return self._cell_types[rowx][colx]

    ##
    # XF index of the cell in the given row and column.
    # This is an index into Book.xf_list.
    # <br /> -- New in version 0.6.1
    def cell_xf_index(self, rowx, colx):
        self.req_fmt_info()
        xfx = self._cell_xf_indexes[rowx][colx]
        if xfx > -1:
            self._xf_index_stats[0] += 1
            return xfx
        # Check for a row xf_index
        try:
            xfx = self.rowinfo_map[rowx].xf_index
            if xfx > -1:
                self._xf_index_stats[1] += 1
                return xfx
        except KeyError:
            pass
        # Check for a column xf_index
        try:
            xfx = self.colinfo_map[colx].xf_index
            assert xfx > -1
            self._xf_index_stats[2] += 1
            return xfx
        except KeyError:
            # If all else fails, 15 is used as hardwired global default xf_index.
            self._xf_index_stats[3] += 1
            return 15

    ##
    # Returns a sequence of the Cell objects in the given row.
    def row(self, rowx):
        return [
            self.cell(rowx, colx)
            for colx in xrange(self.ncols)
            ]

    ##
    # Returns a slice of the types
    # of the cells in the given row.
    def row_types(self, rowx, start_colx=0, end_colx=None):
        if end_colx is None:
            return self._cell_types[rowx][start_colx:]
        return self._cell_types[rowx][start_colx:end_colx]

    ##
    # Returns a slice of the values
    # of the cells in the given row.
    def row_values(self, rowx, start_colx=0, end_colx=None):
        if end_colx is None:
            return self._cell_values[rowx][start_colx:]
        return self._cell_values[rowx][start_colx:end_colx]

    ##
    # Returns a slice of the Cell objects in the given row.
    def row_slice(self, rowx, start_colx=0, end_colx=None):
        nc = self.ncols
        if start_colx < 0:
            start_colx += nc
            if start_colx < 0:
                start_colx = 0
        if end_colx is None or end_colx > nc:
            end_colx = nc
        elif end_colx < 0:
            end_colx += nc
        return [
            self.cell(rowx, colx)
            for colx in xrange(start_colx, end_colx)
            ]

    ##
    # Returns a slice of the Cell objects in the given column.
    def col_slice(self, colx, start_rowx=0, end_rowx=None):
        nr = self.nrows
        if start_rowx < 0:
            start_rowx += nr
            if start_rowx < 0:
                start_rowx = 0
        if end_rowx is None or end_rowx > nr:
            end_rowx = nr
        elif end_rowx < 0:
            end_rowx += nr
        return [
            self.cell(rowx, colx)
            for rowx in xrange(start_rowx, end_rowx)
            ]

    ##
    # Returns a slice of the values of the cells in the given column.
    def col_values(self, colx, start_rowx=0, end_rowx=None):
        nr = self.nrows
        if start_rowx < 0:
            start_rowx += nr
            if start_rowx < 0:
                start_rowx = 0
        if end_rowx is None or end_rowx > nr:
            end_rowx = nr
        elif end_rowx < 0:
            end_rowx += nr
        return [
            self._cell_values[rowx][colx]
            for rowx in xrange(start_rowx, end_rowx)
            ]

    ##
    # Returns a slice of the types of the cells in the given column.
    def col_types(self, colx, start_rowx=0, end_rowx=None):
        nr = self.nrows
        if start_rowx < 0:
            start_rowx += nr
            if start_rowx < 0:
                start_rowx = 0
        if end_rowx is None or end_rowx > nr:
            end_rowx = nr
        elif end_rowx < 0:
            end_rowx += nr
        return [
            self._cell_types[rowx][colx]
            for rowx in xrange(start_rowx, end_rowx)
            ]

    ##
    # Returns a sequence of the Cell objects in the given column.
    def col(self, colx):
        return self.col_slice(colx)
    # Above two lines just for the docs. Here's the real McCoy:
    col = col_slice

    # === Following methods are used in building the worksheet.
    # === They are not part of the API.

    def extend_cells(self, nr, nc):
        # print "extend_cells_2", self.nrows, self.ncols, nr, nc
        assert 1 <= nc <= self.utter_max_cols
        assert 1 <= nr <= self.utter_max_rows
        if nr <= self.nrows:
            # New cell is in an existing row, so extend that row (if necessary).
            # Note that nr < self.nrows means that the cell data
            # is not in ascending row order!!
            self._need_fix_ragged_rows = 1
            nrx = nr - 1
            trow = self._cell_types[nrx]
            tlen = len(trow)
            nextra = max(nc, self.ncols) - tlen
            if nextra > 0:
                xce = XL_CELL_EMPTY
                if self.dont_use_array:
                    trow.extend([xce] * nextra)
                    if self.formatting_info:
                        self._cell_xf_indexes[nrx].extend([-1] * nextra)
                else:
                    aa = array_array
                    trow.extend(aa('B', [xce]) * nextra)
                    if self.formatting_info:
                        self._cell_xf_indexes[nrx].extend(aa('h', [-1]) * nextra)
                self._cell_values[nrx].extend([''] * nextra)
        if nc > self.ncols:
            self.ncols = nc
            self._need_fix_ragged_rows = 1
        if nr > self.nrows:
            scta = self._cell_types.append
            scva = self._cell_values.append
            scxa = self._cell_xf_indexes.append
            fmt_info = self.formatting_info
            xce = XL_CELL_EMPTY
            nc = self.ncols
            if self.dont_use_array:
                for _unused in xrange(self.nrows, nr):
                    scta([xce] * nc)
                    scva([''] * nc)
                    if fmt_info:
                        scxa([-1] * nc)
            else:
                aa = array_array
                for _unused in xrange(self.nrows, nr):
                    scta(aa('B', [xce]) * nc)
                    scva([''] * nc)
                    if fmt_info:
                        scxa(aa('h', [-1]) * nc)
            self.nrows = nr

    def fix_ragged_rows(self):
        t0 = time.time()
        ncols = self.ncols
        xce = XL_CELL_EMPTY
        aa = array_array
        s_cell_types = self._cell_types
        s_cell_values = self._cell_values
        s_cell_xf_indexes = self._cell_xf_indexes
        s_dont_use_array = self.dont_use_array
        s_fmt_info = self.formatting_info
        totrowlen = 0
        for rowx in xrange(self.nrows):
            trow = s_cell_types[rowx]
            rlen = len(trow)
            totrowlen += rlen
            nextra = ncols - rlen
            if nextra > 0:
                s_cell_values[rowx][rlen:] = [''] * nextra
                if s_dont_use_array:
                    trow[rlen:] = [xce] * nextra
                    if s_fmt_info:
                        s_cell_xf_indexes[rowx][rlen:] = [-1] * nextra
                else:
                    trow.extend(aa('B', [xce]) * nextra)
                    if s_fmt_info:
                        s_cell_xf_indexes[rowx][rlen:] = aa('h', [-1]) * nextra
        self._fix_ragged_rows_time = time.time() - t0
        if 0 and self.nrows:
            avgrowlen = float(totrowlen) / self.nrows
            print >> self.logfile, \
                "sheet %d: avg row len %.1f; max row len %d" \
                % (self.number, avgrowlen, self.ncols)

    def tidy_dimensions(self):
        if self.verbosity >= 3:
            fprintf(self.logfile,
                "tidy_dimensions: nrows=%d ncols=%d _need_fix_ragged_rows=%d\n",
                self.nrows, self.ncols, self._need_fix_ragged_rows,
                )
        if 1 and self.merged_cells:
            nr = nc = 0
            umaxrows = self.utter_max_rows
            umaxcols = self.utter_max_cols
            for crange in self.merged_cells:
                rlo, rhi, clo, chi = crange
                if not (0 <= rlo < rhi <= umaxrows) \
                or not (0 <= clo < chi <= umaxcols):
                    fprintf(self.logfile,
                        "*** WARNING: sheet #%d (%r), MERGEDCELLS bad range %r\n",
                        self.number, self.name, crange)
                if rhi > nr: nr = rhi
                if chi > nc: nc = chi
            self.extend_cells(nr, nc)
        if self.verbosity >= 1 \
        and (self.nrows != self._dimnrows or self.ncols != self._dimncols):
            fprintf(self.logfile,
                "NOTE *** sheet %d (%r): DIMENSIONS R,C = %d,%d should be %d,%d\n",
                self.number,
                self.name,
                self._dimnrows,
                self._dimncols,
                self.nrows,
                self.ncols,
                )
        if self._need_fix_ragged_rows:
            self.fix_ragged_rows()

    def put_cell(self, rowx, colx, ctype, value, xf_index):
        try:
            self._cell_types[rowx][colx] = ctype
            self._cell_values[rowx][colx] = value
            if self.formatting_info:
                self._cell_xf_indexes[rowx][colx] = xf_index
        except IndexError:
            # print >> self.logfile, "put_cell extending", rowx, colx
            self.extend_cells(rowx+1, colx+1)
            try:
                self._cell_types[rowx][colx] = ctype
                self._cell_values[rowx][colx] = value
                if self.formatting_info:
                    self._cell_xf_indexes[rowx][colx] = xf_index
            except:
                print >> self.logfile, "put_cell", rowx, colx
                raise
        except:
            print >> self.logfile, "put_cell", rowx, colx
            raise

    def put_blank_cell(self, rowx, colx, xf_index):
        # This is used for cells from BLANK and MULBLANK records
        ctype = XL_CELL_BLANK
        value = ''
        try:
            self._cell_types[rowx][colx] = ctype
            self._cell_values[rowx][colx] = value
            self._cell_xf_indexes[rowx][colx] = xf_index
        except IndexError:
            # print >> self.logfile, "put_cell extending", rowx, colx
            self.extend_cells(rowx+1, colx+1)
            try:
                self._cell_types[rowx][colx] = ctype
                self._cell_values[rowx][colx] = value
                self._cell_xf_indexes[rowx][colx] = xf_index
            except:
                print >> self.logfile, "put_cell", rowx, colx
                raise
        except:
            print >> self.logfile, "put_cell", rowx, colx
            raise

    def put_number_cell(self, rowx, colx, value, xf_index):
        ctype = self._xf_index_to_xl_type_map[xf_index]
        try:
            self._cell_types[rowx][colx] = ctype
            self._cell_values[rowx][colx] = value
            if self.formatting_info:
                self._cell_xf_indexes[rowx][colx] = xf_index
        except IndexError:
            # print >> self.logfile, "put_number_cell extending", rowx, colx
            self.extend_cells(rowx+1, colx+1)
            try:
                self._cell_types[rowx][colx] = ctype
                self._cell_values[rowx][colx] = value
                if self.formatting_info:
                    self._cell_xf_indexes[rowx][colx] = xf_index
            except:
                print >> self.logfile, "put_number_cell", rowx, colx
                raise
        except:
            print >> self.logfile, "put_number_cell", rowx, colx
            raise

    # === Methods after this line neither know nor care about how cells are stored.

    def read(self, bk):
        global rc_stats
        DEBUG = 0
        blah = DEBUG or self.verbosity >= 2
        blah_rows = DEBUG or self.verbosity >= 4
        blah_formulas = 1 and blah
        oldpos = bk._position
        bk._position = self._position
        XL_SHRFMLA_ETC_ETC = (
            XL_SHRFMLA, XL_ARRAY, XL_TABLEOP, XL_TABLEOP2,
            XL_ARRAY2, XL_TABLEOP_B2,
            )
        self_put_number_cell = self.put_number_cell
        self_put_cell = self.put_cell
        self_put_blank_cell = self.put_blank_cell
        local_unpack = unpack
        bk_get_record_parts = bk.get_record_parts
        bv = self.biff_version
        fmt_info = self.formatting_info
        eof_found = 0
        while 1:
            # if DEBUG: print "SHEET.READ: about to read from position %d" % bk._position
            rc, data_len, data = bk_get_record_parts()
            # if rc in rc_stats:
            #     rc_stats[rc] += 1
            # else:
            #     rc_stats[rc] = 1
            # if DEBUG: print "SHEET.READ: op 0x%04x, %d bytes %r" % (rc, data_len, data)
            if rc == XL_NUMBER:
                rowx, colx, xf_index, d = local_unpack('<HHHd', data)
                # if xf_index == 0:
                #     fprintf(self.logfile,
                #         "NUMBER: r=%d c=%d xfx=%d %f\n", rowx, colx, xf_index, d)
                self_put_number_cell(rowx, colx, d, xf_index)
            elif rc == XL_LABELSST:
                rowx, colx, xf_index, sstindex = local_unpack('<HHHi', data)
                # print "LABELSST", rowx, colx, sstindex, bk._sharedstrings[sstindex]
                self_put_cell(rowx, colx, XL_CELL_TEXT, bk._sharedstrings[sstindex], xf_index)
            elif rc == XL_LABEL or rc == XL_RSTRING:
                # RSTRING has extra richtext info at the end, but we ignore it.
                rowx, colx, xf_index = local_unpack('<HHH', data[0:6])
                if bv < BIFF_FIRST_UNICODE:
                    strg = unpack_string(data, 6, bk.encoding or bk.derive_encoding, lenlen=2)
                else:
                    strg = unpack_unicode(data, 6, lenlen=2)
                self_put_cell(rowx, colx, XL_CELL_TEXT, strg, xf_index)
            elif rc == XL_RK:
                rowx, colx, xf_index = local_unpack('<HHH', data[:6])
                d = unpack_RK(data[6:10])
                self_put_number_cell(rowx, colx, d, xf_index)
            elif rc == XL_MULRK:
                mulrk_row, mulrk_first = local_unpack('<HH', data[0:4])
                mulrk_last, = local_unpack('<H', data[-2:])
                pos = 4
                for colx in xrange(mulrk_first, mulrk_last+1):
                    xf_index, = local_unpack('<H', data[pos:pos+2])
                    d = unpack_RK(data[pos+2:pos+6])
                    pos += 6
                    self_put_number_cell(mulrk_row, colx, d, xf_index)
            elif rc == XL_ROW:
                # Version 0.6.0a3: ROW records are just not worth using (for memory allocation).
                # Version 0.6.1: now used for formatting info.
                if not fmt_info: continue
                rowx, bits1, bits2 = local_unpack('<H4xH4xi', data[0:16])
                if not(0 <= rowx < self.utter_max_rows):
                    print >> self.logfile, \
                        "*** NOTE: ROW record has row index %d; " \
                        "should have 0 <= rowx < %d -- record ignored!" \
                        % (rowx, self.utter_max_rows)
                    continue
                r = Rowinfo()
                # Using upkbits() is far too slow on a file
                # with 30 sheets each with 10K rows :-(
                #    upkbits(r, bits1, (
                #        ( 0, 0x7FFF, 'height'),
                #        (15, 0x8000, 'has_default_height'),
                #        ))
                #    upkbits(r, bits2, (
                #        ( 0, 0x00000007, 'outline_level'),
                #        ( 4, 0x00000010, 'outline_group_starts_ends'),
                #        ( 5, 0x00000020, 'hidden'),
                #        ( 6, 0x00000040, 'height_mismatch'),
                #        ( 7, 0x00000080, 'has_default_xf_index'),
                #        (16, 0x0FFF0000, 'xf_index'),
                #        (28, 0x10000000, 'additional_space_above'),
                #        (29, 0x20000000, 'additional_space_below'),
                #        ))
                # So:
                r.height = bits1 & 0x7fff
                r.has_default_height = (bits1 >> 15) & 1
                r.outline_level = bits2 & 7
                r.outline_group_starts_ends = (bits2 >> 4) & 1
                r.hidden = (bits2 >> 5) & 1
                r.height_mismatch = (bits2 >> 6) & 1
                r.has_default_xf_index = (bits2 >> 7) & 1
                r.xf_index = (bits2 >> 16) & 0xfff
                r.additional_space_above = (bits2 >> 28) & 1
                r.additional_space_below = (bits2 >> 29) & 1
                if not r.has_default_xf_index:
                    r.xf_index = -1
                self.rowinfo_map[rowx] = r
                if 0 and r.xf_index > -1:
                    fprintf(self.logfile,
                        "**ROW %d %d %d\n",
                        self.number, rowx, r.xf_index)
                if blah_rows:
                    print >> self.logfile, 'ROW', rowx, bits1, bits2
                    r.dump(self.logfile,
                        header="--- sh #%d, rowx=%d ---" % (self.number, rowx))
            elif rc in XL_FORMULA_OPCODES: # 06, 0206, 0406
                # DEBUG = 1
                # if DEBUG: print "FORMULA: rc: 0x%04x data: %r" % (rc, data)
                if bv >= 50:
                    rowx, colx, xf_index, result_str, flags = local_unpack('<HHH8sH', data[0:16])
                    lenlen = 2
                    tkarr_offset = 20
                elif bv >= 30:
                    rowx, colx, xf_index, result_str, flags = local_unpack('<HHH8sH', data[0:16])
                    lenlen = 2
                    tkarr_offset = 16
                else: # BIFF2
                    rowx, colx, cell_attr,  result_str, flags = local_unpack('<HH3s8sB', data[0:16])
                    xf_index =  self.fixed_BIFF2_xfindex(cell_attr, rowx, colx)
                    lenlen = 1
                    tkarr_offset = 16
                if blah_formulas: # testing formula dumper
                    #### XXXX FIXME
                    fprintf(self.logfile, "FORMULA: rowx=%d colx=%d\n", rowx, colx)
                    fmlalen = local_unpack("<H", data[20:22])[0]
                    decompile_formula(bk, data[22:], fmlalen,
                        reldelta=0, browx=rowx, bcolx=colx, blah=1)
                if result_str[6:8] == "\xFF\xFF":
                    if result_str[0]  == '\x00':
                        # need to read next record (STRING)
                        gotstring = 0
                        # if flags & 8:
                        if 1: # "flags & 8" applies only to SHRFMLA
                            # actually there's an optional SHRFMLA or ARRAY etc record to skip over
                            rc2, data2_len, data2 = bk.get_record_parts()
                            if rc2 == XL_STRING or rc2 == XL_STRING_B2:
                                gotstring = 1
                            elif rc2 == XL_ARRAY:
                                row1x, rownx, col1x, colnx, array_flags, tokslen = \
                                    local_unpack("<HHBBBxxxxxH", data2[:14])
                                if blah_formulas:
                                    fprintf(self.logfile, "ARRAY: %d %d %d %d %d\n",
                                        row1x, rownx, col1x, colnx, array_flags)
                                    dump_formula(bk, data2[14:], tokslen, bv, reldelta=0, blah=1)
                            elif rc2 == XL_SHRFMLA:
                                row1x, rownx, col1x, colnx, nfmlas, tokslen = \
                                    local_unpack("<HHBBxBH", data2[:10])
                                if blah_formulas:
                                    fprintf(self.logfile, "SHRFMLA (sub): %d %d %d %d %d\n",
                                        row1x, rownx, col1x, colnx, nfmlas)
                                    decompile_formula(bk, data2[10:], tokslen, reldelta=1, blah=1)
                            elif rc2 not in XL_SHRFMLA_ETC_ETC:
                                raise XLRDError(
                                    "Expected SHRFMLA, ARRAY, TABLEOP* or STRING record; found 0x%04x" % rc2)
                            # if DEBUG: print "gotstring:", gotstring
                        # now for the STRING record
                        if not gotstring:
                            rc2, _unused_len, data2 = bk.get_record_parts()
                            if rc2 not in (XL_STRING, XL_STRING_B2):
                                raise XLRDError("Expected STRING record; found 0x%04x" % rc2)
                        # if DEBUG: print "STRING: data=%r BIFF=%d cp=%d" % (data2, self.biff_version, bk.encoding)
                        if self.biff_version < BIFF_FIRST_UNICODE:
                            strg = unpack_string(data2, 0, bk.encoding or bk.derive_encoding, lenlen=1 + int(bv > 20))
                        else:
                            strg = unpack_unicode(data2, 0, lenlen=2)
                        self.put_cell(rowx, colx, XL_CELL_TEXT, strg, xf_index)
                        # if DEBUG: print "FORMULA strg %r" % strg
                    elif result_str[0] == '\x01':
                        # boolean formula result
                        value = ord(result_str[2])
                        self.put_cell(rowx, colx, XL_CELL_BOOLEAN, value, xf_index)
                    elif result_str[0] == '\x02':
                        # Error in cell
                        value = ord(result_str[2])
                        self.put_cell(rowx, colx, XL_CELL_ERROR, value, xf_index)
                    elif result_str[0] == '\x03':
                        # empty ... i.e. empty (zero-length) string, NOT an empty cell.
                        self.put_cell(rowx, colx, XL_CELL_TEXT, u"", xf_index)
                    else:
                        raise XLRDError("unexpected special case (0x%02x) in FORMULA" % ord(result_str[0]))
                else:
                    # it is a number
                    d = local_unpack('<d', result_str)[0]
                    self_put_number_cell(rowx, colx, d, xf_index)
            elif rc == XL_BOOLERR:
                rowx, colx, xf_index, value, is_err = local_unpack('<HHHBB', data[:8])
                # Note OOo Calc 2.0 writes 9-byte BOOLERR records.
                # OOo docs say 8. Excel writes 8.
                cellty = (XL_CELL_BOOLEAN, XL_CELL_ERROR)[is_err]
                # if DEBUG: print "XL_BOOLERR", rowx, colx, xf_index, value, is_err
                self.put_cell(rowx, colx, cellty, value, xf_index)
            elif rc == XL_COLINFO:
                if not fmt_info: continue
                c = Colinfo()
                first_colx, last_colx, c.width, c.xf_index, flags \
                    = local_unpack("<HHHHH", data[:10])
                #### Colinfo.width is denominated in 256ths of a character,
                #### *not* in characters.
                if not(0 <= first_colx <= last_colx <= 256):
                    # Note: 256 instead of 255 is a common mistake.
                    # We silently ignore the non-existing 257th column in that case.
                    print >> self.logfile, \
                        "*** NOTE: COLINFO record has first col index %d, last %d; " \
                        "should have 0 <= first <= last <= 255 -- record ignored!" \
                        % (first_colx, last_colx)
                    del c
                    continue
                upkbits(c, flags, (
                    ( 0, 0x0001, 'hidden'),
                    ( 1, 0x0002, 'bit1_flag'),
                    # *ALL* colinfos created by Excel in "default" cases are 0x0002!!
                    # Maybe it's "locked" by analogy with XFProtection data.
                    ( 8, 0x0700, 'outline_level'),
                    (12, 0x1000, 'collapsed'),
                    ))
                for colx in xrange(first_colx, last_colx+1):
                    if colx > 255: break # Excel does 0 to 256 inclusive
                    self.colinfo_map[colx] = c
                    if 0:
                        fprintf(self.logfile,
                            "**COL %d %d %d\n",
                            self.number, colx, c.xf_index)
                if blah:
                    fprintf(
                        self.logfile,
                        "COLINFO sheet #%d cols %d-%d: wid=%d xf_index=%d flags=0x%04x\n",
                        self.number, first_colx, last_colx, c.width, c.xf_index, flags,
                        )
                    c.dump(self.logfile, header='===')
            elif rc == XL_DEFCOLWIDTH:
                self.defcolwidth, = local_unpack("<H", data[:2])
                if 0: print >> self.logfile, 'DEFCOLWIDTH', self.defcolwidth
            elif rc == XL_STANDARDWIDTH:
                if data_len != 2:
                    print >> self.logfile, '*** ERROR *** STANDARDWIDTH', data_len, repr(data)
                self.standardwidth, = local_unpack("<H", data[:2])
                if 0: print >> self.logfile, 'STANDARDWIDTH', self.standardwidth
            elif rc == XL_GCW:
                if not fmt_info: continue # useless w/o COLINFO
                assert data_len == 34
                assert data[0:2] == "\x20\x00"
                iguff = unpack("<8i", data[2:34])
                gcw = []
                for bits in iguff:
                    for j in xrange(32):
                        gcw.append(bits & 1)
                        bits >>= 1
                self.gcw = tuple(gcw)
                if 0:
                    showgcw = "".join(map(lambda x: "F "[x], gcw)).rstrip().replace(' ', '.')
                    print "GCW:", showgcw
            elif rc == XL_BLANK:
                if not fmt_info: continue
                rowx, colx, xf_index = local_unpack('<HHH', data[:6])
                if 0: print >> self.logfile, "BLANK", rowx, colx, xf_index
                self_put_blank_cell(rowx, colx, xf_index)
            elif rc == XL_MULBLANK: # 00BE
                if not fmt_info: continue
                mul_row, mul_first = local_unpack('<HH', data[0:4])
                mul_last, = local_unpack('<H', data[-2:])
                if 0:
                    print >> self.logfile, "MULBLANK", mul_row, mul_first, mul_last
                pos = 4
                for colx in xrange(mul_first, mul_last+1):
                    xf_index, = local_unpack('<H', data[pos:pos+2])
                    pos += 2
                    self_put_blank_cell(mul_row, colx, xf_index)
            elif rc == XL_DIMENSION or rc == XL_DIMENSION2:
                # if data_len == 10:
                # Was crashing on BIFF 4.0 file w/o the two trailing unused bytes.
                # Reported by Ralph Heimburger.
                if bv < 80:
                    dim_tuple = local_unpack('<HxxH', data[2:8])
                else:
                    dim_tuple = local_unpack('<ixxH', data[4:12])
                self.nrows, self.ncols = 0, 0
                self._dimnrows, self._dimncols = dim_tuple
                if not self.book._xf_epilogue_done:
                    # Needed for bv <= 40
                    self.book.xf_epilogue()
                if blah:
                    fprintf(self.logfile,
                        "sheet %d(%r) DIMENSIONS: ncols=%d nrows=%d\n",
                        self.number, self.name, self._dimncols, self._dimnrows
                        )
            elif rc == XL_EOF:
                DEBUG = 0
                if DEBUG: print >> self.logfile, "SHEET.READ: EOF"
                eof_found = 1
                break
            elif rc == XL_OBJ:
                # handle SHEET-level objects; note there's a separate Book.handle_obj
                self.handle_obj(data)
            elif rc == XL_MSO_DRAWING:
                self.handle_msodrawingetc(rc, data_len, data)
            elif rc == XL_TXO:
                self.handle_txo(data)
            elif rc == XL_NOTE:
                self.handle_note(data)
            elif rc == XL_FEAT11:
                self.handle_feat11(data)
            elif rc in bofcodes: ##### EMBEDDED BOF #####
                version, boftype = local_unpack('<HH', data[0:4])
                if boftype != 0x20: # embedded chart
                    print >> self.logfile, \
                        "*** Unexpected embedded BOF (0x%04x) at offset %d: version=0x%04x type=0x%04x" \
                        % (rc, bk._position - data_len - 4, version, boftype)
                while 1:
                    code, data_len, data = bk.get_record_parts()
                    if code == XL_EOF:
                        break
                if DEBUG: print >> self.logfile, "---> found EOF"
            elif rc == XL_COUNTRY:
                bk.handle_country(data)
            elif rc == XL_LABELRANGES:
                pos = 0
                pos = unpack_cell_range_address_list_update_pos(
                        self.row_label_ranges, data, pos, bv, addr_size=8,
                        )
                pos = unpack_cell_range_address_list_update_pos(
                        self.col_label_ranges, data, pos, bv, addr_size=8,
                        )
                assert pos == data_len
            elif rc == XL_ARRAY:
                row1x, rownx, col1x, colnx, array_flags, tokslen = \
                    local_unpack("<HHBBBxxxxxH", data[:14])
                if blah_formulas:
                    print "ARRAY:", row1x, rownx, col1x, colnx, array_flags
                    dump_formula(bk, data[14:], tokslen, bv, reldelta=0, blah=1)
            elif rc == XL_SHRFMLA:
                row1x, rownx, col1x, colnx, nfmlas, tokslen = \
                    local_unpack("<HHBBxBH", data[:10])
                if blah_formulas:
                    print "SHRFMLA (main):", row1x, rownx, col1x, colnx, nfmlas
                    decompile_formula(bk, data[10:], tokslen, reldelta=0, blah=1)
            elif rc == XL_CONDFMT:
                if not fmt_info: continue
                assert bv >= 80
                num_CFs, needs_recalc, browx1, browx2, bcolx1, bcolx2 = \
                    unpack("<6H", data[0:12])
                if self.verbosity >= 1:
                    fprintf(self.logfile,
                        "\n*** WARNING: Ignoring CONDFMT (conditional formatting) record\n" \
                        "*** in Sheet %d (%r).\n" \
                        "*** %d CF record(s); needs_recalc_or_redraw = %d\n" \
                        "*** Bounding box is %s\n",
                        self.number, self.name, num_CFs, needs_recalc,
                        rangename2d(browx1, browx2+1, bcolx1, bcolx2+1),
                        )
                olist = [] # updated by the function
                pos = unpack_cell_range_address_list_update_pos(
                    olist, data, 12, bv, addr_size=8)
                # print >> self.logfile, repr(result), len(result)
                if self.verbosity >= 1:
                    fprintf(self.logfile,
                        "*** %d individual range(s):\n" \
                        "*** %s\n",
                        len(olist),
                        ", ".join([rangename2d(*coords) for coords in olist]),
                        )
            elif rc == XL_CF:
                if not fmt_info: continue
                cf_type, cmp_op, sz1, sz2, flags = unpack("<BBHHi", data[0:10])
                font_block = (flags >> 26) & 1
                bord_block = (flags >> 28) & 1
                patt_block = (flags >> 29) & 1
                if self.verbosity >= 1:
                    fprintf(self.logfile,
                        "\n*** WARNING: Ignoring CF (conditional formatting) sub-record.\n" \
                        "*** cf_type=%d, cmp_op=%d, sz1=%d, sz2=%d, flags=0x%08x\n" \
                        "*** optional data blocks: font=%d, border=%d, pattern=%d\n",
                        cf_type, cmp_op, sz1, sz2, flags,
                        font_block, bord_block, patt_block,
                        )
                # hex_char_dump(data, 0, data_len)
                pos = 12
                if font_block:
                    (font_height, font_options, weight, escapement, underline,
                    font_colour_index, two_bits, font_esc, font_underl) = \
                    unpack("<64x i i H H B 3x i 4x i i i 18x", data[pos:pos+118])
                    font_style = (two_bits > 1) & 1
                    posture = (font_options > 1) & 1
                    font_canc = (two_bits > 7) & 1
                    cancellation = (font_options > 7) & 1
                    if self.verbosity >= 1:
                        fprintf(self.logfile,
                            "*** Font info: height=%d, weight=%d, escapement=%d,\n" \
                            "*** underline=%d, colour_index=%d, esc=%d, underl=%d,\n" \
                            "*** style=%d, posture=%d, canc=%d, cancellation=%d\n",
                            font_height, weight, escapement, underline,
                            font_colour_index, font_esc, font_underl,
                            font_style, posture, font_canc, cancellation,
                            )
                    pos += 118
                if bord_block:
                    pos += 8
                if patt_block:
                    pos += 4
                fmla1 = data[pos:pos+sz1]
                pos += sz1
                if blah and sz1:
                    fprintf(self.logfile,
                        "*** formula 1:\n",
                        )
                    dump_formula(bk, fmla1, sz1, bv, reldelta=0, blah=1)
                fmla2 = data[pos:pos+sz2]
                pos += sz2
                assert pos == data_len
                if blah and sz2:
                    fprintf(self.logfile,
                        "*** formula 2:\n",
                        )
                    dump_formula(bk, fmla2, sz2, bv, reldelta=0, blah=1)
            elif rc == XL_DEFAULTROWHEIGHT:
                if data_len == 4:
                    bits, self.default_row_height = unpack("<HH", data[:4])
                elif data_len == 2:
                    self.default_row_height, = unpack("<H", data)
                    bits = 0
                    fprintf(self.logfile,
                        "*** WARNING: DEFAULTROWHEIGHT record len is 2, " \
                        "should be 4; assuming BIFF2 format\n")
                else:
                    bits = 0
                    fprintf(self.logfile,
                        "*** WARNING: DEFAULTROWHEIGHT record len is %d, " \
                        "should be 4; ignoring this record\n",
                        data_len)
                self.default_row_height_mismatch = bits & 1
                self.default_row_hidden = (bits >> 1) & 1
                self.default_additional_space_above = (bits >> 2) & 1
                self.default_additional_space_below = (bits >> 3) & 1
            elif rc == XL_MERGEDCELLS:
                if not fmt_info: continue
                pos = unpack_cell_range_address_list_update_pos(
                    self.merged_cells, data, 0, bv, addr_size=8)
                if blah:
                    fprintf(self.logfile,
                        "MERGEDCELLS: %d ranges\n", int_floor_div(pos - 2, 8))
                assert pos == data_len, \
                    "MERGEDCELLS: pos=%d data_len=%d" % (pos, data_len)
            elif rc == XL_WINDOW2:
                if bv >= 80:
                    (options,
                    self.first_visible_rowx, self.first_visible_colx,
                    self.gridline_colour_index,
                    self.cached_page_break_preview_mag_factor,
                    self.cached_normal_view_mag_factor
                    ) = unpack("<HHHHxxHH", data[:14])
                else: # BIFF3-7
                    (options,
                    self.first_visible_rowx, self.first_visible_colx,
                    ) = unpack("<HHH", data[:6])
                    self.gridline_colour_rgb = unpack("<BBB", data[6:9])
                    self.gridline_colour_index = \
                        nearest_colour_index(
                            self.book.colour_map,
                            self.gridline_colour_rgb,
                            debug=0)
                    self.cached_page_break_preview_mag_factor = 0 # default (60%)
                    self.cached_normal_view_mag_factor = 0 # default (100%)
                # options -- Bit, Mask, Contents:
                # 0 0001H 0 = Show formula results 1 = Show formulas
                # 1 0002H 0 = Do not show grid lines 1 = Show grid lines
                # 2 0004H 0 = Do not show sheet headers 1 = Show sheet headers
                # 3 0008H 0 = Panes are not frozen 1 = Panes are frozen (freeze)
                # 4 0010H 0 = Show zero values as empty cells 1 = Show zero values
                # 5 0020H 0 = Manual grid line colour 1 = Automatic grid line colour
                # 6 0040H 0 = Columns from left to right 1 = Columns from right to left
                # 7 0080H 0 = Do not show outline symbols 1 = Show outline symbols
                # 8 0100H 0 = Keep splits if pane freeze is removed 1 = Remove splits if pane freeze is removed
                # 9 0200H 0 = Sheet not selected 1 = Sheet selected (BIFF5-BIFF8)
                # 10 0400H 0 = Sheet not visible 1 = Sheet visible (BIFF5-BIFF8)
                # 11 0800H 0 = Show in normal view 1 = Show in page break preview (BIFF8)
                # The freeze flag specifies, if a following PANE record (6.71) describes unfrozen or frozen panes.
                for attr, _unused_defval in _WINDOW2_options:
                    setattr(self, attr, options & 1)
                    options >>= 1
                # print "WINDOW2: visible=%d selected=%d" \
                #     % (self.sheet_visible, self.sheet_selected)
            #### all of the following are for BIFF <= 4W
            elif bv <= 45:
                if rc == XL_FORMAT or rc == XL_FORMAT2:
                    bk.handle_format(data, rc)
                elif rc == XL_FONT or rc == XL_FONT_B3B4:
                    bk.handle_font(data)
                elif rc == XL_STYLE:
                    if not self.book._xf_epilogue_done:
                        self.book.xf_epilogue()
                    bk.handle_style(data)
                elif rc == XL_PALETTE:
                    bk.handle_palette(data)
                elif rc == XL_BUILTINFMTCOUNT:
                    bk.handle_builtinfmtcount(data)
                elif rc == XL_XF4 or rc == XL_XF3 or rc == XL_XF2: #### N.B. not XL_XF
                    bk.handle_xf(data)
                elif rc == XL_DATEMODE:
                    bk.handle_datemode(data)
                elif rc == XL_CODEPAGE:
                    bk.handle_codepage(data)
                elif rc == XL_FILEPASS:
                    bk.handle_filepass(data)
                elif rc == XL_WRITEACCESS:
                    bk.handle_writeaccess(data)
                elif rc == XL_IXFE:
                    self._ixfe = local_unpack('<H', data)[0]
                elif rc == XL_NUMBER_B2:
                    rowx, colx, cell_attr, d = local_unpack('<HH3sd', data)
                    self_put_number_cell(rowx, colx, d, self.fixed_BIFF2_xfindex(cell_attr, rowx, colx))
                elif rc == XL_INTEGER:
                    rowx, colx, cell_attr, d = local_unpack('<HH3sH', data)
                    self_put_number_cell(rowx, colx, float(d), self.fixed_BIFF2_xfindex(cell_attr, rowx, colx))
                elif rc == XL_LABEL_B2:
                    rowx, colx, cell_attr = local_unpack('<HH3s', data[0:7])
                    strg = unpack_string(data, 7, bk.encoding or bk.derive_encoding(), lenlen=1)
                    self_put_cell(rowx, colx, XL_CELL_TEXT, strg, self.fixed_BIFF2_xfindex(cell_attr, rowx, colx))
                elif rc == XL_BOOLERR_B2:
                    rowx, colx, cell_attr, value, is_err = local_unpack('<HH3sBB', data)
                    cellty = (XL_CELL_BOOLEAN, XL_CELL_ERROR)[is_err]
                    # if DEBUG: print "XL_BOOLERR_B2", rowx, colx, cell_attr, value, is_err
                    self.put_cell(rowx, colx, cellty, value, self.fixed_BIFF2_xfindex(cell_attr, rowx, colx))
                elif rc == XL_BLANK_B2:
                    if not fmt_info: continue
                    rowx, colx, cell_attr = local_unpack('<HH3s', data[:7])
                    self_put_blank_cell(rowx, colx, self.fixed_BIFF2_xfindex(cell_attr, rowx, colx))
                elif rc == XL_EFONT:
                    bk.handle_efont(data)
                elif rc == XL_ROW_B2:
                    if not fmt_info: continue
                    rowx, bits1, has_defaults = local_unpack('<H4xH2xB', data[0:11])
                    if not(0 <= rowx < self.utter_max_rows):
                        print >> self.logfile, \
                            "*** NOTE: ROW_B2 record has row index %d; " \
                            "should have 0 <= rowx < %d -- record ignored!" \
                            % (rowx, self.utter_max_rows)
                        continue
                    r = Rowinfo()
                    r.height = bits1 & 0x7fff
                    r.has_default_height = (bits1 >> 15) & 1
                    r.outline_level = 0
                    r.outline_group_starts_ends = 0
                    r.hidden = 0
                    r.height_mismatch = 0
                    r.has_default_xf_index = has_defaults & 1
                    r.additional_space_above = 0
                    r.additional_space_below = 0
                    if not r.has_default_xf_index:
                        r.xf_index = -1
                    elif data_len == 18:
                        # Seems the XF index in the cell_attr is dodgy
                         xfx = local_unpack('<H', data[16:18])[0]
                         r.xf_index = self.fixed_BIFF2_xfindex(cell_attr=None, rowx=rowx, colx=-1, true_xfx=xfx)
                    else:
                        cell_attr = data[13:16]
                        r.xf_index = self.fixed_BIFF2_xfindex(cell_attr, rowx, colx=-1)
                    self.rowinfo_map[rowx] = r
                    if 0 and r.xf_index > -1:
                        fprintf(self.logfile,
                            "**ROW %d %d %d\n",
                            self.number, rowx, r.xf_index)
                    if blah_rows:
                        print >> self.logfile, 'ROW_B2', rowx, bits1, has_defaults
                        r.dump(self.logfile,
                            header="--- sh #%d, rowx=%d ---" % (self.number, rowx))
                elif rc == XL_COLWIDTH: # BIFF2 only
                    if not fmt_info: continue
                    first_colx, last_colx, width\
                        = local_unpack("<BBH", data[:4])
                    if not(first_colx <= last_colx):
                        print >> self.logfile, \
                            "*** NOTE: COLWIDTH record has first col index %d, last %d; " \
                            "should have first <= last -- record ignored!" \
                            % (first_colx, last_colx)
                        continue
                    for colx in xrange(first_colx, last_colx+1):
                        if self.colinfo_map.has_key(colx):
                            c = self.colinfo_map[colx]
                        else:
                            c = Colinfo()
                            self.colinfo_map[colx] = c
                        c.width = width
                    if blah:
                        fprintf(
                            self.logfile,
                            "COLWIDTH sheet #%d cols %d-%d: wid=%d\n",
                            self.number, first_colx, last_colx, width
                            )
                elif rc == XL_COLUMNDEFAULT: # BIFF2 only
                    if not fmt_info: continue
                    first_colx, last_colx = local_unpack("<HH", data[:4])
                    #### Warning OOo docs wrong; first_colx <= colx < last_colx
                    if blah:
                        fprintf(
                            self.logfile,
                            "COLUMNDEFAULT sheet #%d cols in range(%d, %d)\n",
                            self.number, first_colx, last_colx
                            )
                    if not(0 <= first_colx < last_colx <= 256):
                        print >> self.logfile, \
                            "*** NOTE: COLUMNDEFAULT record has first col index %d, last %d; " \
                            "should have 0 <= first < last <= 256" \
                            % (first_colx, last_colx)
                        last_colx = min(last_colx, 256)
                    for colx in xrange(first_colx, last_colx):
                        offset = 4 + 3 * (colx - first_colx)
                        cell_attr = data[offset:offset+3]
                        xf_index = self.fixed_BIFF2_xfindex(cell_attr, rowx=-1, colx=colx)
                        if self.colinfo_map.has_key(colx):
                            c = self.colinfo_map[colx]
                        else:
                            c = Colinfo()
                            self.colinfo_map[colx] = c
                        c.xf_index = xf_index
            else:
                # if DEBUG: print "SHEET.READ: Unhandled record type %02x %d bytes %r" % (rc, data_len, data)
                pass
        if not eof_found:
            raise XLRDError("Sheet %d (%r) missing EOF record" \
                % (self.number, self.name))
        self.tidy_dimensions()
        bk._position = oldpos
        return 1

    def fixed_BIFF2_xfindex(self, cell_attr, rowx, colx, true_xfx=None):
        DEBUG = 0
        blah = DEBUG or self.verbosity >= 2
        if self.biff_version == 21:
            if self._xf_index_to_xl_type_map:
                if true_xfx is not None:
                    xfx = true_xfx
                else:
                    xfx = ord(cell_attr[0]) & 0x3F
                if xfx == 0x3F:
                    if self._ixfe is None:
                        raise XLRDError("BIFF2 cell record has XF index 63 but no preceding IXFE record.")
                    xfx = self._ixfe
                    # OOo docs are capable of interpretation that each
                    # cell record is preceded immediately by its own IXFE record.
                    # Empirical evidence is that (sensibly) an IXFE record applies to all
                    # following cell records until another IXFE comes along.
                return xfx
            # Have either Excel 2.0, or broken 2.1 w/o XF records -- same effect.
            self.biff_version = self.book.biff_version = 20
        #### check that XF slot in cell_attr is zero
        xfx_slot = ord(cell_attr[0]) & 0x3F
        assert xfx_slot == 0
        xfx = self._cell_attr_to_xfx.get(cell_attr)
        if xfx is not None:
            return xfx
        if blah:
            fprintf(self.logfile, "New cell_attr %r at (%r, %r)\n", cell_attr, rowx, colx)
        book = self.book
        xf = self.fake_XF_from_BIFF20_cell_attr(cell_attr)
        xfx = len(book.xf_list)
        xf.xf_index = xfx
        book.xf_list.append(xf)
        if blah:
            xf.dump(self.logfile, header="=== Faked XF %d ===" % xfx, footer="======")
        if not book.format_map.has_key(xf.format_key):
            msg = "ERROR *** XF[%d] unknown format key (%d, 0x%04x)\n"
            fprintf(self.logfile, msg,
                    xf.xf_index, xf.format_key, xf.format_key)
            fmt = Format(xf.format_key, FUN, u"General")
            book.format_map[xf.format_key] = fmt
            while len(book.format_list) <= xf.format_key:
                book.format_list.append(fmt)
        cellty_from_fmtty = {
            FNU: XL_CELL_NUMBER,
            FUN: XL_CELL_NUMBER,
            FGE: XL_CELL_NUMBER,
            FDT: XL_CELL_DATE,
            FTX: XL_CELL_NUMBER, # Yes, a number can be formatted as text.
            }
        fmt = book.format_map[xf.format_key]
        cellty = cellty_from_fmtty[fmt.type]
        self._xf_index_to_xl_type_map[xf.xf_index] = cellty
        self._cell_attr_to_xfx[cell_attr] = xfx
        return xfx

    def fake_XF_from_BIFF20_cell_attr(self, cell_attr):
        from formatting import XF, XFAlignment, XFBorder, XFBackground, XFProtection
        xf = XF()
        xf.alignment = XFAlignment()
        xf.alignment.indent_level = 0
        xf.alignment.shrink_to_fit = 0
        xf.alignment.text_direction = 0
        xf.border = XFBorder()
        xf.border.diag_up = 0
        xf.border.diag_down = 0
        xf.border.diag_colour_index = 0
        xf.border.diag_line_style = 0 # no line
        xf.background = XFBackground()
        xf.protection = XFProtection()
        (prot_bits, font_and_format, halign_etc) = unpack('<BBB', cell_attr)
        xf.format_key = font_and_format & 0x3F
        xf.font_index = (font_and_format & 0xC0) >> 6
        upkbits(xf.protection, prot_bits, (
            (6, 0x40, 'cell_locked'),
            (7, 0x80, 'formula_hidden'),
            ))
        xf.alignment.hor_align = halign_etc & 0x07
        for mask, side in ((0x08, 'left'), (0x10, 'right'), (0x20, 'top'), (0x40, 'bottom')):
            if halign_etc & mask:
                colour_index, line_style = 8, 1 # black, thin
            else:
                colour_index, line_style = 0, 0 # none, none
            setattr(xf.border, side + '_colour_index', colour_index)
            setattr(xf.border, side + '_line_style', line_style)
        bg = xf.background
        if halign_etc & 0x80:
            bg.fill_pattern = 17
        else:
            bg.fill_pattern = 0
        bg.background_colour_index = 9 # white
        bg.pattern_colour_index = 8 # black
        xf.parent_style_index = 0 # ???????????
        xf.alignment.vert_align = 2 # bottom
        xf.alignment.rotation = 0
        for attr_stem in \
            "format font alignment border background protection".split():
            attr = "_" + attr_stem + "_flag"
            setattr(xf, attr, 1)
        return xf

    def req_fmt_info(self):
        if not self.formatting_info:
            raise XLRDError("Feature requires open_workbook(..., formatting_info=True)")

    ##
    # Determine column display width.
    # <br /> -- New in version 0.6.1
    # <br />
    # @param colx Index of the queried column, range 0 to 255.
    # Note that it is possible to find out the width that will be used to display
    # columns with no cell information e.g. column IV (colx=255).
    # @return The column width that will be used for displaying
    # the given column by Excel, in units of 1/256th of the width of a
    # standard character (the digit zero in the first font).

    def computed_column_width(self, colx):
        self.req_fmt_info()
        if self.biff_version >= 80:
            colinfo = self.colinfo_map.get(colx, None)
            if colinfo is not None:
                return colinfo.width
            if self.standardwidth is not None:
                return self.standardwidth
        elif self.biff_version >= 40:
            if self.gcw[colx]:
                if self.standardwidth is not None:
                    return self.standardwidth
            else:
                colinfo = self.colinfo_map.get(colx, None)
                if colinfo is not None:
                    return colinfo.width
        elif self.biff_version == 30:
            colinfo = self.colinfo_map.get(colx, None)
            if colinfo is not None:
                return colinfo.width
        # All roads lead to Rome and the DEFCOLWIDTH ...
        if self.defcolwidth is not None:
            return self.defcolwidth * 256
        return 8 * 256 # 8 is what Excel puts in a DEFCOLWIDTH record

    def handle_msodrawingetc(self, recid, data_len, data):
        if not OBJ_MSO_DEBUG:
            return
        DEBUG = 1
        if self.biff_version < 80:
            return
        o = MSODrawing()
        pos = 0
        while pos < data_len:
            tmp, fbt, cb = unpack('<HHI', data[pos:pos+8])
            ver = tmp & 0xF
            inst = (tmp >> 4) & 0xFFF
            if ver == 0xF:
                ndb = 0 # container
            else:
                ndb = cb
            if DEBUG:
                hex_char_dump(data, pos, ndb + 8, base=0, fout=self.logfile)
                fprintf(self.logfile,
                    "fbt:0x%04X  inst:%d  ver:0x%X  cb:%d (0x%04X)\n",
                    fbt, inst, ver, cb, cb)
            if fbt == 0xF010: # Client Anchor
                assert ndb == 18
                (o.anchor_unk,
                o.anchor_colx_lo, o.anchor_rowx_lo,
                o.anchor_colx_hi, o.anchor_rowx_hi) = unpack('<Hiiii', data[pos+8:pos+8+ndb])
            elif fbt == 0xF011: # Client Data
                # must be followed by an OBJ record
                assert cb == 0
                assert pos + 8 == data_len
            else:
                pass
            pos += ndb + 8
        else:
            # didn't break out of while loop
            assert pos == data_len
        if DEBUG:
            o.dump(self.logfile, header="=== MSODrawing ===", footer= " ")


    def handle_obj(self, data):
        if not OBJ_MSO_DEBUG:
            return
        DEBUG = 1
        if self.biff_version < 80:
            return
        o = MSObj()
        data_len = len(data)
        pos = 0
        if DEBUG:
            fprintf(self.logfile, "... OBJ record ...\n")
        while pos < data_len:
            ft, cb = unpack('<HH', data[pos:pos+4])
            if DEBUG:
                hex_char_dump(data, pos, cb, base=0, fout=self.logfile)
            if ft == 0x15: # ftCmo ... s/b first
                assert pos == 0
                o.type, o.id, option_flags = unpack('<HHH', data[pos+4:pos+10])
                upkbits(o, option_flags, (
                    ( 0, 0x0001, 'locked'),
                    ( 4, 0x0010, 'printable'),
                    ( 8, 0x0100, 'autofilter'), # not documented in Excel 97 dev kit
                    ( 9, 0x0200, 'scrollbar_flag'), # not documented in Excel 97 dev kit
                    (13, 0x2000, 'autofill'),
                    (14, 0x4000, 'autoline'),
                    ))
            elif ft == 0x00:
                assert cb == 0
                assert pos + 4 == data_len
            elif ft == 0x0C: # Scrollbar
                values = unpack('<5H', data[pos+8:pos+18])
                for value, tag in zip(values, ('value', 'min', 'max', 'inc', 'page')):
                    setattr(o, 'scrollbar_' + tag, value)
            elif ft == 0x0D: # "Notes structure" [used for cell comments]
                pass ############## not documented in Excel 97 dev kit
            elif ft == 0x13: # list box data
                if o.autofilter: # non standard exit. NOT documented
                    break
            else:
                pass
            pos += cb + 4
        else:
            # didn't break out of while loop
            assert pos == data_len
        if DEBUG:
            o.dump(self.logfile, header="=== MSOBj ===", footer= " ")

    def handle_note(self, data):
        if not OBJ_MSO_DEBUG:
            return
        DEBUG = 1
        if self.biff_version < 80:
            return
        if DEBUG:
            fprintf(self.logfile, '... NOTE record ...\n')
            hex_char_dump(data, 0, len(data), base=0, fout=self.logfile)
        o = MSNote()
        data_len = len(data)
        o.rowx, o.colx, option_flags, o.object_id = unpack('<4H', data[:8])
        o.show = (option_flags >> 1) & 1
        # Docs say NULL [sic] bytes padding between string count and string data
        # to ensure that string is word-aligned. Appears to be nonsense.
        # There also seems to be a random(?) byte after the string (not counted in the
        # string length.
        o.original_author, endpos = unpack_unicode_update_pos(data, 8, lenlen=2)
        assert endpos == data_len - 1
        o.last_byte = data[-1]
        if DEBUG:
            o.dump(self.logfile, header="=== MSNote ===", footer= " ")

    def handle_txo(self, data):
        if not OBJ_MSO_DEBUG:
            return
        DEBUG = 1
        if self.biff_version < 80:
            return
        o = MSTxo()
        data_len = len(data)
        option_flags, o.rot, cchText, cbRuns = unpack('<HH6xHH4x', data)
        upkbits(o, option_flags, (
            (3, 0x000E, 'horz_align'),
            (6, 0x0070, 'vert_align'),
            (9, 0x0200, 'lock_text'),
            ))
        rc2, data2_len, data2 = self.book.get_record_parts()
        assert rc2 == XL_CONTINUE
        o.text, endpos = unpack_unicode_update_pos(data2, 0, known_len=cchText)
        assert endpos == data2_len
        rc3, data3_len, data3 = self.book.get_record_parts()
        assert rc3 == XL_CONTINUE
        # ignore the formatting runs for the moment
        if DEBUG:
            o.dump(self.logfile, header="=== MSTxo ===", footer= " ")

    def handle_feat11(self, data):
        if not OBJ_MSO_DEBUG:
            return
        # rt: Record type; this matches the BIFF rt in the first two bytes of the record; =0872h
        # grbitFrt: FRT cell reference flag (see table below for details)
        # Ref0: Range reference to a worksheet cell region if grbitFrt=1 (bitFrtRef). Otherwise blank.
        # isf: Shared feature type index =5 for Table
        # fHdr: =0 since this is for feat not feat header
        # reserved0: Reserved for future use =0 for Table
        # cref: Count of ref ranges this feature is on
        # cbFeatData: Count of byte for the current feature data.
        # reserved1: =0 currently not used
        # Ref1: Repeat of Ref0. UNDOCUMENTED
        rt, grbitFrt, Ref0, isf, fHdr, reserved0, cref, cbFeatData, reserved1, Ref1 = unpack('<HH8sHBiHiH8s', data[0:35])
        assert reserved0 == 0
        assert reserved1 == 0
        assert isf == 5
        assert rt == 0x872
        assert fHdr == 0
        assert Ref1 == Ref0
        print "FEAT11: grbitFrt=%d  Ref0=%r cref=%d cbFeatData=%d" % (grbitFrt, Ref0, cref, cbFeatData)
        # lt: Table data source type:
        #   =0 for Excel Worksheet Table =1 for read-write SharePoint linked List
        #   =2 for XML mapper Table =3 for Query Table
        # idList: The ID of the Table (unique per worksheet)
        # crwHeader: How many header/title rows the Table has at the top
        # crwTotals: How many total rows the Table has at the bottom
        # idFieldNext: Next id to try when assigning a unique id to a new field
        # cbFSData: The size of the Fixed Data portion of the Table data structure.
        # rupBuild: the rupBuild that generated the record
        # unusedShort: UNUSED short that can be used later. The value is reserved during round-tripping.
        # listFlags: Collection of bit flags: (see listFlags' bit setting table below for detail.)
        # lPosStmCache: Table data stream position of cached data
        # cbStmCache: Count of bytes of cached data
        # cchStmCache: Count of characters of uncompressed cached data in the stream
        # lem: Table edit mode (see List (Table) Editing Mode (lem) setting table below for details.)
        # rgbHashParam: Hash value for SharePoint Table
        # cchName: Count of characters in the Table name string rgbName
        (lt, idList, crwHeader, crwTotals, idFieldNext, cbFSData,
        rupBuild, unusedShort, listFlags, lPosStmCache, cbStmCache,
        cchStmCache, lem, rgbHashParam, cchName) = unpack('<iiiiiiHHiiiii16sH', data[35:35+66])
        print "lt=%d  idList=%d crwHeader=%d  crwTotals=%d  idFieldNext=%d cbFSData=%d\n"\
            "rupBuild=%d  unusedShort=%d listFlags=%04X  lPosStmCache=%d  cbStmCache=%d\n"\
            "cchStmCache=%d  lem=%d  rgbHashParam=%r  cchName=%d" % (
            lt, idList, crwHeader, crwTotals, idFieldNext, cbFSData,
            rupBuild, unusedShort,listFlags, lPosStmCache, cbStmCache,
            cchStmCache, lem, rgbHashParam, cchName)

class MSODrawing(BaseObject):
    pass

class MSObj(BaseObject):
    pass

class MSTxo(BaseObject):
    pass

class MSNote(BaseObject):
    pass

# === helpers ===

def unpack_RK(rk_str):
    flags = ord(rk_str[0])
    if flags & 2:
        # There's a SIGNED 30-bit integer in there!
        i,  = unpack('<i', rk_str)
        i >>= 2 # div by 4 to drop the 2 flag bits
        if flags & 1:
            return i / 100.0
        return float(i)
    else:
        # It's the most significant 30 bits of an IEEE 754 64-bit FP number
        d, = unpack('<d', '\0\0\0\0' + chr(flags & 252) + rk_str[1:4])
        if flags & 1:
            return d / 100.0
        return d

##### =============== Cell ======================================== #####

cellty_from_fmtty = {
    FNU: XL_CELL_NUMBER,
    FUN: XL_CELL_NUMBER,
    FGE: XL_CELL_NUMBER,
    FDT: XL_CELL_DATE,
    FTX: XL_CELL_NUMBER, # Yes, a number can be formatted as text.
    }

ctype_text = {
    XL_CELL_EMPTY: 'empty',
    XL_CELL_TEXT: 'text',
    XL_CELL_NUMBER: 'number',
    XL_CELL_DATE: 'xldate',
    XL_CELL_BOOLEAN: 'bool',
    XL_CELL_ERROR: 'error',
    XL_CELL_BLANK: 'blank',
    }

##
# <p>Contains the data for one cell.</p>
#
# <p>WARNING: You don't call this class yourself. You access Cell objects
# via methods of the Sheet object(s) that you found in the Book object that
# was returned when you called xlrd.open_workbook("myfile.xls").</p>
# <p> Cell objects have three attributes: <i>ctype</i> is an int, <i>value</i>
# (which depends on <i>ctype</i>) and <i>xf_index</i>.
# If "formatting_info" is not enabled when the workbook is opened, xf_index will be None.
# The following table describes the types of cells and how their values
# are represented in Python.</p>
#
# <table border="1" cellpadding="7">
# <tr>
# <th>Type symbol</th>
# <th>Type number</th>
# <th>Python value</th>
# </tr>
# <tr>
# <td>XL_CELL_EMPTY</td>
# <td align="center">0</td>
# <td>empty string u''</td>
# </tr>
# <tr>
# <td>XL_CELL_TEXT</td>
# <td align="center">1</td>
# <td>a Unicode string</td>
# </tr>
# <tr>
# <td>XL_CELL_NUMBER</td>
# <td align="center">2</td>
# <td>float</td>
# </tr>
# <tr>
# <td>XL_CELL_DATE</td>
# <td align="center">3</td>
# <td>float</td>
# </tr>
# <tr>
# <td>XL_CELL_BOOLEAN</td>
# <td align="center">4</td>
# <td>int; 1 means TRUE, 0 means FALSE</td>
# </tr>
# <tr>
# <td>XL_CELL_ERROR</td>
# <td align="center">5</td>
# <td>int representing internal Excel codes; for a text representation,
# refer to the supplied dictionary error_text_from_code</td>
# </tr>
# <tr>
# <td>XL_CELL_BLANK</td>
# <td align="center">6</td>
# <td>empty string u''. Note: this type will appear only when
# open_workbook(..., formatting_info=True) is used.</td>
# </tr>
# </table>
#<p></p>

class Cell(BaseObject):

    __slots__ = ['ctype', 'value', 'xf_index']

    def __init__(self, ctype, value, xf_index=None):
        self.ctype = ctype
        self.value = value
        self.xf_index = xf_index

    def __repr__(self):
        if self.xf_index is None:
            return "%s:%r" % (ctype_text[self.ctype], self.value)
        else:
            return "%s:%r (XF:%r)" % (ctype_text[self.ctype], self.value, self.xf_index)

##
# There is one and only one instance of an empty cell -- it's a singleton. This is it.
# You may use a test like "acell is empty_cell".
empty_cell = Cell(XL_CELL_EMPTY, '')

##### =============== Colinfo and Rowinfo ============================== #####

##
# Width and default formatting information that applies to one or
# more columns in a sheet. Derived from COLINFO records.
#
# <p> Here is the default hierarchy for width, according to the OOo docs:
#
# <br />"""In BIFF3, if a COLINFO record is missing for a column,
# the width specified in the record DEFCOLWIDTH is used instead.
#
# <br />In BIFF4-BIFF7, the width set in this [COLINFO] record is only used,
# if the corresponding bit for this column is cleared in the GCW
# record, otherwise the column width set in the DEFCOLWIDTH record
# is used (the STANDARDWIDTH record is always ignored in this case [see footnote!]).
#
# <br />In BIFF8, if a COLINFO record is missing for a column,
# the width specified in the record STANDARDWIDTH is used.
# If this [STANDARDWIDTH] record is also missing,
# the column width of the record DEFCOLWIDTH is used instead."""
# <br />
#
# Footnote:  The docs on the GCW record say this:
# """<br />
# If a bit is set, the corresponding column uses the width set in the STANDARDWIDTH
# record. If a bit is cleared, the corresponding column uses the width set in the
# COLINFO record for this column.
# <br />If a bit is set, and the worksheet does not contain the STANDARDWIDTH record, or if
# the bit is cleared, and the worksheet does not contain the COLINFO record, the DEFCOLWIDTH
# record of the worksheet will be used instead.
# <br />"""<br />
# At the moment (2007-01-17) xlrd is going with the GCW version of the story.
# Reference to the source may be useful: see the computed_column_width(colx) method
# of the Sheet class.
# <br />-- New in version 0.6.1
# </p>

class Colinfo(BaseObject):
    ##
    # Width of the column in 1/256 of the width of the zero character,
    # using default font (first FONT record in the file).
    width = 0
    ##
    # XF index to be used for formatting empty cells.
    xf_index = -1
    ##
    # 1 = column is hidden
    hidden = 0
    ##
    # Value of a 1-bit flag whose purpose is unknown
    # but is often seen set to 1
    bit1_flag = 0
    ##
    # Outline level of the column, in range(7).
    # (0 = no outline)
    outline_level = 0
    ##
    # 1 = column is collapsed
    collapsed = 0

##
# Height and default formatting information that applies to a row in a sheet.
# Derived from ROW records.
# <br /> -- New in version 0.6.1

class Rowinfo(BaseObject):
    ##
    # Height of the row, in twips. One twip == 1/20 of a point
    height = 0
    ##
    # 0 = Row has custom height; 1 = Row has default height
    has_default_height = 0
    ##
    # Outline level of the row
    outline_level = 0
    ##
    # 1 = Outline group starts or ends here (depending on where the
    # outline buttons are located, see WSBOOL record [TODO ??]),
    # <i>and</i> is collapsed
    outline_group_starts_ends = 0
    ##
    # 1 = Row is hidden (manually, or by a filter or outline group)
    hidden = 0
    ##
    # 1 = Row height and default font height do not match
    height_mismatch = 0
    ##
    # 1 = the xf_index attribute is usable; 0 = ignore it
    has_default_xf_index = 0
    ##
    # Index to default XF record for empty cells in this row.
    # Don't use this if has_default_xf_index == 0.
    xf_index = -9999
    ##
    # This flag is set, if the upper border of at least one cell in this row
    # or if the lower border of at least one cell in the row above is
    # formatted with a thick line style. Thin and medium line styles are not
    # taken into account.
    additional_space_above = 0
    ##
    # This flag is set, if the lower border of at least one cell in this row
    # or if the upper border of at least one cell in the row below is
    # formatted with a medium or thick line style. Thin line styles are not
    # taken into account.
    additional_space_below = 0