summaryrefslogtreecommitdiffstats
path: root/Lib/plistlib.py
blob: b9946fd313af0d67fddb74c1d0022f803f518ad6 (plain)
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
r"""plistlib.py -- a tool to generate and parse MacOSX .plist files.

The property list (.plist) file format is a simple XML pickle supporting
basic object types, like dictionaries, lists, numbers and strings.
Usually the top level object is a dictionary.

To write out a plist file, use the dump(value, file)
function. 'value' is the top level object, 'file' is
a (writable) file object.

To parse a plist from a file, use the load(file) function,
with a (readable) file object as the only argument. It
returns the top level object (again, usually a dictionary).

To work with plist data in bytes objects, you can use loads()
and dumps().

Values can be strings, integers, floats, booleans, tuples, lists,
dictionaries (but only with string keys), Data, bytes, bytearray, or
datetime.datetime objects.

Generate Plist example:

    pl = dict(
        aString = "Doodah",
        aList = ["A", "B", 12, 32.1, [1, 2, 3]],
        aFloat = 0.1,
        anInt = 728,
        aDict = dict(
            anotherString = "<hello & hi there!>",
            aUnicodeValue = "M\xe4ssig, Ma\xdf",
            aTrueValue = True,
            aFalseValue = False,
        ),
        someData = b"<binary gunk>",
        someMoreData = b"<lots of binary gunk>" * 10,
        aDate = datetime.datetime.fromtimestamp(time.mktime(time.gmtime())),
    )
    with open(fileName, 'wb') as fp:
        dump(pl, fp)

Parse Plist example:

    with open(fileName, 'rb') as fp:
        pl = load(fp)
    print(pl["aKey"])
"""
__all__ = [
    "readPlist", "writePlist", "readPlistFromBytes", "writePlistToBytes",
    "Plist", "Data", "Dict", "FMT_XML", "FMT_BINARY",
    "load", "dump", "loads", "dumps"
]

import binascii
import codecs
import contextlib
import datetime
import enum
from io import BytesIO
import itertools
import os
import re
import struct
from warnings import warn
from xml.parsers.expat import ParserCreate


PlistFormat = enum.Enum('PlistFormat', 'FMT_XML FMT_BINARY', module=__name__)
globals().update(PlistFormat.__members__)


#
#
# Deprecated functionality
#
#


class _InternalDict(dict):

    # This class is needed while Dict is scheduled for deprecation:
    # we only need to warn when a *user* instantiates Dict or when
    # the "attribute notation for dict keys" is used.
    __slots__ = ()

    def __getattr__(self, attr):
        try:
            value = self[attr]
        except KeyError:
            raise AttributeError(attr)
        warn("Attribute access from plist dicts is deprecated, use d[key] "
             "notation instead", DeprecationWarning, 2)
        return value

    def __setattr__(self, attr, value):
        warn("Attribute access from plist dicts is deprecated, use d[key] "
             "notation instead", DeprecationWarning, 2)
        self[attr] = value

    def __delattr__(self, attr):
        try:
            del self[attr]
        except KeyError:
            raise AttributeError(attr)
        warn("Attribute access from plist dicts is deprecated, use d[key] "
             "notation instead", DeprecationWarning, 2)


class Dict(_InternalDict):

    def __init__(self, **kwargs):
        warn("The plistlib.Dict class is deprecated, use builtin dict instead",
             DeprecationWarning, 2)
        super().__init__(**kwargs)


@contextlib.contextmanager
def _maybe_open(pathOrFile, mode):
    if isinstance(pathOrFile, str):
        with open(pathOrFile, mode) as fp:
            yield fp

    else:
        yield pathOrFile


class Plist(_InternalDict):
    """This class has been deprecated. Use dump() and load()
    functions instead, together with regular dict objects.
    """

    def __init__(self, **kwargs):
        warn("The Plist class is deprecated, use the load() and "
             "dump() functions instead", DeprecationWarning, 2)
        super().__init__(**kwargs)

    @classmethod
    def fromFile(cls, pathOrFile):
        """Deprecated. Use the load() function instead."""
        with _maybe_open(pathOrFile, 'rb') as fp:
            value = load(fp)
        plist = cls()
        plist.update(value)
        return plist

    def write(self, pathOrFile):
        """Deprecated. Use the dump() function instead."""
        with _maybe_open(pathOrFile, 'wb') as fp:
            dump(self, fp)


def readPlist(pathOrFile):
    """
    Read a .plist from a path or file. pathOrFile should either
    be a file name, or a readable binary file object.

    This function is deprecated, use load instead.
    """
    warn("The readPlist function is deprecated, use load() instead",
        DeprecationWarning, 2)

    with _maybe_open(pathOrFile, 'rb') as fp:
        return load(fp, fmt=None, use_builtin_types=False,
            dict_type=_InternalDict)

def writePlist(value, pathOrFile):
    """
    Write 'value' to a .plist file. 'pathOrFile' may either be a
    file name or a (writable) file object.

    This function is deprecated, use dump instead.
    """
    warn("The writePlist function is deprecated, use dump() instead",
        DeprecationWarning, 2)
    with _maybe_open(pathOrFile, 'wb') as fp:
        dump(value, fp, fmt=FMT_XML, sort_keys=True, skipkeys=False)


def readPlistFromBytes(data):
    """
    Read a plist data from a bytes object. Return the root object.

    This function is deprecated, use loads instead.
    """
    warn("The readPlistFromBytes function is deprecated, use loads() instead",
        DeprecationWarning, 2)
    return load(BytesIO(data), fmt=None, use_builtin_types=False,
        dict_type=_InternalDict)


def writePlistToBytes(value):
    """
    Return 'value' as a plist-formatted bytes object.

    This function is deprecated, use dumps instead.
    """
    warn("The writePlistToBytes function is deprecated, use dumps() instead",
        DeprecationWarning, 2)
    f = BytesIO()
    dump(value, f, fmt=FMT_XML, sort_keys=True, skipkeys=False)
    return f.getvalue()


class Data:
    """
    Wrapper for binary data.

    This class is deprecated, use a bytes object instead.
    """

    def __init__(self, data):
        if not isinstance(data, bytes):
            raise TypeError("data must be as bytes")
        self.data = data

    @classmethod
    def fromBase64(cls, data):
        # base64.decodebytes just calls binascii.a2b_base64;
        # it seems overkill to use both base64 and binascii.
        return cls(_decode_base64(data))

    def asBase64(self, maxlinelength=76):
        return _encode_base64(self.data, maxlinelength)

    def __eq__(self, other):
        if isinstance(other, self.__class__):
            return self.data == other.data
        elif isinstance(other, str):
            return self.data == other
        else:
            return id(self) == id(other)

    def __repr__(self):
        return "%s(%s)" % (self.__class__.__name__, repr(self.data))

#
#
# End of deprecated functionality
#
#


#
# XML support
#


# XML 'header'
PLISTHEADER = b"""\
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
"""


# Regex to find any control chars, except for \t \n and \r
_controlCharPat = re.compile(
    r"[\x00\x01\x02\x03\x04\x05\x06\x07\x08\x0b\x0c\x0e\x0f"
    r"\x10\x11\x12\x13\x14\x15\x16\x17\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f]")

def _encode_base64(s, maxlinelength=76):
    # copied from base64.encodebytes(), with added maxlinelength argument
    maxbinsize = (maxlinelength//4)*3
    pieces = []
    for i in range(0, len(s), maxbinsize):
        chunk = s[i : i + maxbinsize]
        pieces.append(binascii.b2a_base64(chunk))
    return b''.join(pieces)

def _decode_base64(s):
    if isinstance(s, str):
        return binascii.a2b_base64(s.encode("utf-8"))

    else:
        return binascii.a2b_base64(s)

# Contents should conform to a subset of ISO 8601
# (in particular, YYYY '-' MM '-' DD 'T' HH ':' MM ':' SS 'Z'.  Smaller units
# may be omitted with #  a loss of precision)
_dateParser = re.compile(r"(?P<year>\d\d\d\d)(?:-(?P<month>\d\d)(?:-(?P<day>\d\d)(?:T(?P<hour>\d\d)(?::(?P<minute>\d\d)(?::(?P<second>\d\d))?)?)?)?)?Z", re.ASCII)


def _date_from_string(s):
    order = ('year', 'month', 'day', 'hour', 'minute', 'second')
    gd = _dateParser.match(s).groupdict()
    lst = []
    for key in order:
        val = gd[key]
        if val is None:
            break
        lst.append(int(val))
    return datetime.datetime(*lst)


def _date_to_string(d):
    return '%04d-%02d-%02dT%02d:%02d:%02dZ' % (
        d.year, d.month, d.day,
        d.hour, d.minute, d.second
    )

def _escape(text):
    m = _controlCharPat.search(text)
    if m is not None:
        raise ValueError("strings can't contains control characters; "
                         "use bytes instead")
    text = text.replace("\r\n", "\n")       # convert DOS line endings
    text = text.replace("\r", "\n")         # convert Mac line endings
    text = text.replace("&", "&amp;")       # escape '&'
    text = text.replace("<", "&lt;")        # escape '<'
    text = text.replace(">", "&gt;")        # escape '>'
    return text

class _PlistParser:
    def __init__(self, use_builtin_types, dict_type):
        self.stack = []
        self.current_key = None
        self.root = None
        self._use_builtin_types = use_builtin_types
        self._dict_type = dict_type

    def parse(self, fileobj):
        self.parser = ParserCreate()
        self.parser.StartElementHandler = self.handle_begin_element
        self.parser.EndElementHandler = self.handle_end_element
        self.parser.CharacterDataHandler = self.handle_data
        self.parser.ParseFile(fileobj)
        return self.root

    def handle_begin_element(self, element, attrs):
        self.data = []
        handler = getattr(self, "begin_" + element, None)
        if handler is not None:
            handler(attrs)

    def handle_end_element(self, element):
        handler = getattr(self, "end_" + element, None)
        if handler is not None:
            handler()

    def handle_data(self, data):
        self.data.append(data)

    def add_object(self, value):
        if self.current_key is not None:
            if not isinstance(self.stack[-1], type({})):
                raise ValueError("unexpected element at line %d" %
                                 self.parser.CurrentLineNumber)
            self.stack[-1][self.current_key] = value
            self.current_key = None
        elif not self.stack:
            # this is the root object
            self.root = value
        else:
            if not isinstance(self.stack[-1], type([])):
                raise ValueError("unexpected element at line %d" %
                                 self.parser.CurrentLineNumber)
            self.stack[-1].append(value)

    def get_data(self):
        data = ''.join(self.data)
        self.data = []
        return data

    # element handlers

    def begin_dict(self, attrs):
        d = self._dict_type()
        self.add_object(d)
        self.stack.append(d)

    def end_dict(self):
        if self.current_key:
            raise ValueError("missing value for key '%s' at line %d" %
                             (self.current_key,self.parser.CurrentLineNumber))
        self.stack.pop()

    def end_key(self):
        if self.current_key or not isinstance(self.stack[-1], type({})):
            raise ValueError("unexpected key at line %d" %
                             self.parser.CurrentLineNumber)
        self.current_key = self.get_data()

    def begin_array(self, attrs):
        a = []
        self.add_object(a)
        self.stack.append(a)

    def end_array(self):
        self.stack.pop()

    def end_true(self):
        self.add_object(True)

    def end_false(self):
        self.add_object(False)

    def end_integer(self):
        self.add_object(int(self.get_data()))

    def end_real(self):
        self.add_object(float(self.get_data()))

    def end_string(self):
        self.add_object(self.get_data())

    def end_data(self):
        if self._use_builtin_types:
            self.add_object(_decode_base64(self.get_data()))

        else:
            self.add_object(Data.fromBase64(self.get_data()))

    def end_date(self):
        self.add_object(_date_from_string(self.get_data()))


class _DumbXMLWriter:
    def __init__(self, file, indent_level=0, indent="\t"):
        self.file = file
        self.stack = []
        self._indent_level = indent_level
        self.indent = indent

    def begin_element(self, element):
        self.stack.append(element)
        self.writeln("<%s>" % element)
        self._indent_level += 1

    def end_element(self, element):
        assert self._indent_level > 0
        assert self.stack.pop() == element
        self._indent_level -= 1
        self.writeln("</%s>" % element)

    def simple_element(self, element, value=None):
        if value is not None:
            value = _escape(value)
            self.writeln("<%s>%s</%s>" % (element, value, element))

        else:
            self.writeln("<%s/>" % element)

    def writeln(self, line):
        if line:
            # plist has fixed encoding of utf-8

            # XXX: is this test needed?
            if isinstance(line, str):
                line = line.encode('utf-8')
            self.file.write(self._indent_level * self.indent)
            self.file.write(line)
        self.file.write(b'\n')


class _PlistWriter(_DumbXMLWriter):
    def __init__(
            self, file, indent_level=0, indent=b"\t", writeHeader=1,
            sort_keys=True, skipkeys=False):

        if writeHeader:
            file.write(PLISTHEADER)
        _DumbXMLWriter.__init__(self, file, indent_level, indent)
        self._sort_keys = sort_keys
        self._skipkeys = skipkeys

    def write(self, value):
        self.writeln("<plist version=\"1.0\">")
        self.write_value(value)
        self.writeln("</plist>")

    def write_value(self, value):
        if isinstance(value, str):
            self.simple_element("string", value)

        elif value is True:
            self.simple_element("true")

        elif value is False:
            self.simple_element("false")

        elif isinstance(value, int):
            if -1 << 63 <= value < 1 << 64:
                self.simple_element("integer", "%d" % value)
            else:
                raise OverflowError(value)

        elif isinstance(value, float):
            self.simple_element("real", repr(value))

        elif isinstance(value, dict):
            self.write_dict(value)

        elif isinstance(value, Data):
            self.write_data(value)

        elif isinstance(value, (bytes, bytearray)):
            self.write_bytes(value)

        elif isinstance(value, datetime.datetime):
            self.simple_element("date", _date_to_string(value))

        elif isinstance(value, (tuple, list)):
            self.write_array(value)

        else:
            raise TypeError("unsupported type: %s" % type(value))

    def write_data(self, data):
        self.write_bytes(data.data)

    def write_bytes(self, data):
        self.begin_element("data")
        self._indent_level -= 1
        maxlinelength = max(
            16,
            76 - len(self.indent.replace(b"\t", b" " * 8) * self._indent_level))

        for line in _encode_base64(data, maxlinelength).split(b"\n"):
            if line:
                self.writeln(line)
        self._indent_level += 1
        self.end_element("data")

    def write_dict(self, d):
        if d:
            self.begin_element("dict")
            if self._sort_keys:
                items = sorted(d.items())
            else:
                items = d.items()

            for key, value in items:
                if not isinstance(key, str):
                    if self._skipkeys:
                        continue
                    raise TypeError("keys must be strings")
                self.simple_element("key", key)
                self.write_value(value)
            self.end_element("dict")

        else:
            self.simple_element("dict")

    def write_array(self, array):
        if array:
            self.begin_element("array")
            for value in array:
                self.write_value(value)
            self.end_element("array")

        else:
            self.simple_element("array")


def _is_fmt_xml(header):
    prefixes = (b'<?xml', b'<plist')

    for pfx in prefixes:
        if header.startswith(pfx):
            return True

    # Also check for alternative XML encodings, this is slightly
    # overkill because the Apple tools (and plistlib) will not
    # generate files with these encodings.
    for bom, encoding in (
                (codecs.BOM_UTF8, "utf-8"),
                (codecs.BOM_UTF16_BE, "utf-16-be"),
                (codecs.BOM_UTF16_LE, "utf-16-le"),
                # expat does not support utf-32
                #(codecs.BOM_UTF32_BE, "utf-32-be"),
                #(codecs.BOM_UTF32_LE, "utf-32-le"),
            ):
        if not header.startswith(bom):
            continue

        for start in prefixes:
            prefix = bom + start.decode('ascii').encode(encoding)
            if header[:len(prefix)] == prefix:
                return True

    return False

#
# Binary Plist
#


class InvalidFileException (ValueError):
    def __init__(self, message="Invalid file"):
        ValueError.__init__(self, message)

_BINARY_FORMAT = {1: 'B', 2: 'H', 4: 'L', 8: 'Q'}

class _BinaryPlistParser:
    """
    Read or write a binary plist file, following the description of the binary
    format.  Raise InvalidFileException in case of error, otherwise return the
    root object.

    see also: http://opensource.apple.com/source/CF/CF-744.18/CFBinaryPList.c
    """
    def __init__(self, use_builtin_types, dict_type):
        self._use_builtin_types = use_builtin_types
        self._dict_type = dict_type

    def parse(self, fp):
        try:
            # The basic file format:
            # HEADER
            # object...
            # refid->offset...
            # TRAILER
            self._fp = fp
            self._fp.seek(-32, os.SEEK_END)
            trailer = self._fp.read(32)
            if len(trailer) != 32:
                raise InvalidFileException()
            (
                offset_size, self._ref_size, num_objects, top_object,
                offset_table_offset
            ) = struct.unpack('>6xBBQQQ', trailer)
            self._fp.seek(offset_table_offset)
            self._object_offsets = self._read_ints(num_objects, offset_size)
            return self._read_object(self._object_offsets[top_object])

        except (OSError, IndexError, struct.error):
            raise InvalidFileException()

    def _get_size(self, tokenL):
        """ return the size of the next object."""
        if tokenL == 0xF:
            m = self._fp.read(1)[0] & 0x3
            s = 1 << m
            f = '>' + _BINARY_FORMAT[s]
            return struct.unpack(f, self._fp.read(s))[0]

        return tokenL

    def _read_ints(self, n, size):
        data = self._fp.read(size * n)
        if size in _BINARY_FORMAT:
            return struct.unpack('>' + _BINARY_FORMAT[size] * n, data)
        else:
            return tuple(int.from_bytes(data[i: i + size], 'big')
                         for i in range(0, size * n, size))

    def _read_refs(self, n):
        return self._read_ints(n, self._ref_size)

    def _read_object(self, offset):
        """
        read the object at offset.

        May recursively read sub-objects (content of an array/dict/set)
        """
        self._fp.seek(offset)
        token = self._fp.read(1)[0]
        tokenH, tokenL = token & 0xF0, token & 0x0F

        if token == 0x00:
            return None

        elif token == 0x08:
            return False

        elif token == 0x09:
            return True

        # The referenced source code also mentions URL (0x0c, 0x0d) and
        # UUID (0x0e), but neither can be generated using the Cocoa libraries.

        elif token == 0x0f:
            return b''

        elif tokenH == 0x10:  # int
            return int.from_bytes(self._fp.read(1 << tokenL),
                                  'big', signed=tokenL >= 3)

        elif token == 0x22: # real
            return struct.unpack('>f', self._fp.read(4))[0]

        elif token == 0x23: # real
            return struct.unpack('>d', self._fp.read(8))[0]

        elif token == 0x33:  # date
            f = struct.unpack('>d', self._fp.read(8))[0]
            # timestamp 0 of binary plists corresponds to 1/1/2001
            # (year of Mac OS X 10.0), instead of 1/1/1970.
            return datetime.datetime.utcfromtimestamp(f + (31 * 365 + 8) * 86400)

        elif tokenH == 0x40:  # data
            s = self._get_size(tokenL)
            if self._use_builtin_types:
                return self._fp.read(s)
            else:
                return Data(self._fp.read(s))

        elif tokenH == 0x50:  # ascii string
            s = self._get_size(tokenL)
            result =  self._fp.read(s).decode('ascii')
            return result

        elif tokenH == 0x60:  # unicode string
            s = self._get_size(tokenL)
            return self._fp.read(s * 2).decode('utf-16be')

        # tokenH == 0x80 is documented as 'UID' and appears to be used for
        # keyed-archiving, not in plists.

        elif tokenH == 0xA0:  # array
            s = self._get_size(tokenL)
            obj_refs = self._read_refs(s)
            return [self._read_object(self._object_offsets[x])
                for x in obj_refs]

        # tokenH == 0xB0 is documented as 'ordset', but is not actually
        # implemented in the Apple reference code.

        # tokenH == 0xC0 is documented as 'set', but sets cannot be used in
        # plists.

        elif tokenH == 0xD0:  # dict
            s = self._get_size(tokenL)
            key_refs = self._read_refs(s)
            obj_refs = self._read_refs(s)
            result = self._dict_type()
            for k, o in zip(key_refs, obj_refs):
                result[self._read_object(self._object_offsets[k])
                    ] = self._read_object(self._object_offsets[o])
            return result

        raise InvalidFileException()

def _count_to_size(count):
    if count < 1 << 8:
        return 1

    elif count < 1 << 16:
        return 2

    elif count << 1 << 32:
        return 4

    else:
        return 8

class _BinaryPlistWriter (object):
    def __init__(self, fp, sort_keys, skipkeys):
        self._fp = fp
        self._sort_keys = sort_keys
        self._skipkeys = skipkeys

    def write(self, value):

        # Flattened object list:
        self._objlist = []

        # Mappings from object->objectid
        # First dict has (type(object), object) as the key,
        # second dict is used when object is not hashable and
        # has id(object) as the key.
        self._objtable = {}
        self._objidtable = {}

        # Create list of all objects in the plist
        self._flatten(value)

        # Size of object references in serialized containers
        # depends on the number of objects in the plist.
        num_objects = len(self._objlist)
        self._object_offsets = [0]*num_objects
        self._ref_size = _count_to_size(num_objects)

        self._ref_format = _BINARY_FORMAT[self._ref_size]

        # Write file header
        self._fp.write(b'bplist00')

        # Write object list
        for obj in self._objlist:
            self._write_object(obj)

        # Write refnum->object offset table
        top_object = self._getrefnum(value)
        offset_table_offset = self._fp.tell()
        offset_size = _count_to_size(offset_table_offset)
        offset_format = '>' + _BINARY_FORMAT[offset_size] * num_objects
        self._fp.write(struct.pack(offset_format, *self._object_offsets))

        # Write trailer
        sort_version = 0
        trailer = (
            sort_version, offset_size, self._ref_size, num_objects,
            top_object, offset_table_offset
        )
        self._fp.write(struct.pack('>5xBBBQQQ', *trailer))

    def _flatten(self, value):
        # First check if the object is in the object table, not used for
        # containers to ensure that two subcontainers with the same contents
        # will be serialized as distinct values.
        if isinstance(value, (
                str, int, float, datetime.datetime, bytes, bytearray)):
            if (type(value), value) in self._objtable:
                return

        elif isinstance(value, Data):
            if (type(value.data), value.data) in self._objtable:
                return

        # Add to objectreference map
        refnum = len(self._objlist)
        self._objlist.append(value)
        try:
            if isinstance(value, Data):
                self._objtable[(type(value.data), value.data)] = refnum
            else:
                self._objtable[(type(value), value)] = refnum
        except TypeError:
            self._objidtable[id(value)] = refnum

        # And finally recurse into containers
        if isinstance(value, dict):
            keys = []
            values = []
            items = value.items()
            if self._sort_keys:
                items = sorted(items)

            for k, v in items:
                if not isinstance(k, str):
                    if self._skipkeys:
                        continue
                    raise TypeError("keys must be strings")
                keys.append(k)
                values.append(v)

            for o in itertools.chain(keys, values):
                self._flatten(o)

        elif isinstance(value, (list, tuple)):
            for o in value:
                self._flatten(o)

    def _getrefnum(self, value):
        try:
            if isinstance(value, Data):
                return self._objtable[(type(value.data), value.data)]
            else:
                return self._objtable[(type(value), value)]
        except TypeError:
            return self._objidtable[id(value)]

    def _write_size(self, token, size):
        if size < 15:
            self._fp.write(struct.pack('>B', token | size))

        elif size < 1 << 8:
            self._fp.write(struct.pack('>BBB', token | 0xF, 0x10, size))

        elif size < 1 << 16:
            self._fp.write(struct.pack('>BBH', token | 0xF, 0x11, size))

        elif size < 1 << 32:
            self._fp.write(struct.pack('>BBL', token | 0xF, 0x12, size))

        else:
            self._fp.write(struct.pack('>BBQ', token | 0xF, 0x13, size))

    def _write_object(self, value):
        ref = self._getrefnum(value)
        self._object_offsets[ref] = self._fp.tell()
        if value is None:
            self._fp.write(b'\x00')

        elif value is False:
            self._fp.write(b'\x08')

        elif value is True:
            self._fp.write(b'\x09')

        elif isinstance(value, int):
            if value < 0:
                try:
                    self._fp.write(struct.pack('>Bq', 0x13, value))
                except struct.error:
                    raise OverflowError(value) from None
            elif value < 1 << 8:
                self._fp.write(struct.pack('>BB', 0x10, value))
            elif value < 1 << 16:
                self._fp.write(struct.pack('>BH', 0x11, value))
            elif value < 1 << 32:
                self._fp.write(struct.pack('>BL', 0x12, value))
            elif value < 1 << 63:
                self._fp.write(struct.pack('>BQ', 0x13, value))
            elif value < 1 << 64:
                self._fp.write(b'\x14' + value.to_bytes(16, 'big', signed=True))
            else:
                raise OverflowError(value)

        elif isinstance(value, float):
            self._fp.write(struct.pack('>Bd', 0x23, value))

        elif isinstance(value, datetime.datetime):
            f = (value - datetime.datetime(2001, 1, 1)).total_seconds()
            self._fp.write(struct.pack('>Bd', 0x33, f))

        elif isinstance(value, Data):
            self._write_size(0x40, len(value.data))
            self._fp.write(value.data)

        elif isinstance(value, (bytes, bytearray)):
            self._write_size(0x40, len(value))
            self._fp.write(value)

        elif isinstance(value, str):
            try:
                t = value.encode('ascii')
                self._write_size(0x50, len(value))
            except UnicodeEncodeError:
                t = value.encode('utf-16be')
                self._write_size(0x60, len(value))

            self._fp.write(t)

        elif isinstance(value, (list, tuple)):
            refs = [self._getrefnum(o) for o in value]
            s = len(refs)
            self._write_size(0xA0, s)
            self._fp.write(struct.pack('>' + self._ref_format * s, *refs))

        elif isinstance(value, dict):
            keyRefs, valRefs = [], []

            if self._sort_keys:
                rootItems = sorted(value.items())
            else:
                rootItems = value.items()

            for k, v in rootItems:
                if not isinstance(k, str):
                    if self._skipkeys:
                        continue
                    raise TypeError("keys must be strings")
                keyRefs.append(self._getrefnum(k))
                valRefs.append(self._getrefnum(v))

            s = len(keyRefs)
            self._write_size(0xD0, s)
            self._fp.write(struct.pack('>' + self._ref_format * s, *keyRefs))
            self._fp.write(struct.pack('>' + self._ref_format * s, *valRefs))

        else:
            raise TypeError(value)


def _is_fmt_binary(header):
    return header[:8] == b'bplist00'


#
# Generic bits
#

_FORMATS={
    FMT_XML: dict(
        detect=_is_fmt_xml,
        parser=_PlistParser,
        writer=_PlistWriter,
    ),
    FMT_BINARY: dict(
        detect=_is_fmt_binary,
        parser=_BinaryPlistParser,
        writer=_BinaryPlistWriter,
    )
}


def load(fp, *, fmt=None, use_builtin_types=True, dict_type=dict):
    """Read a .plist file. 'fp' should be (readable) file object.
    Return the unpacked root object (which usually is a dictionary).
    """
    if fmt is None:
        header = fp.read(32)
        fp.seek(0)
        for info in _FORMATS.values():
            if info['detect'](header):
                P = info['parser']
                break

        else:
            raise InvalidFileException()

    else:
        P = _FORMATS[fmt]['parser']

    p = P(use_builtin_types=use_builtin_types, dict_type=dict_type)
    return p.parse(fp)


def loads(value, *, fmt=None, use_builtin_types=True, dict_type=dict):
    """Read a .plist file from a bytes object.
    Return the unpacked root object (which usually is a dictionary).
    """
    fp = BytesIO(value)
    return load(
        fp, fmt=fmt, use_builtin_types=use_builtin_types, dict_type=dict_type)


def dump(value, fp, *, fmt=FMT_XML, sort_keys=True, skipkeys=False):
    """Write 'value' to a .plist file. 'fp' should be a (writable)
    file object.
    """
    if fmt not in _FORMATS:
        raise ValueError("Unsupported format: %r"%(fmt,))

    writer = _FORMATS[fmt]["writer"](fp, sort_keys=sort_keys, skipkeys=skipkeys)
    writer.write(value)


def dumps(value, *, fmt=FMT_XML, skipkeys=False, sort_keys=True):
    """Return a bytes object with the contents for a .plist file.
    """
    fp = BytesIO()
    dump(value, fp, fmt=fmt, skipkeys=skipkeys, sort_keys=sort_keys)
    return fp.getvalue()