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
|
"""New I/O library conforming to PEP 3116.
This is a prototype; hopefully eventually some of this will be
reimplemented in C.
Conformance of alternative implementations: all arguments are intended
to be positional-only except the arguments of the open() function.
Argument names except those of the open() function are not part of the
specification. Instance variables and methods whose name starts with
a leading underscore are not part of the specification (except "magic"
names like __iter__). Only the top-level names listed in the __all__
variable are part of the specification.
XXX edge cases when switching between reading/writing
XXX need to default buffer size to 1 if isatty()
XXX need to support 1 meaning line-buffered
XXX don't use assert to validate input requirements
XXX whenever an argument is None, use the default value
XXX read/write ops should check readable/writable
XXX buffered readinto should work with arbitrary buffer objects
XXX use incremental encoder for text output, at least for UTF-16 and UTF-8-SIG
"""
__author__ = ("Guido van Rossum <guido@python.org>, "
"Mike Verdone <mike.verdone@gmail.com>, "
"Mark Russell <mark.russell@zen.co.uk>")
__all__ = ["BlockingIOError", "open", "IOBase", "RawIOBase", "FileIO",
"SocketIO", "BytesIO", "StringIO", "BufferedIOBase",
"BufferedReader", "BufferedWriter", "BufferedRWPair",
"BufferedRandom", "TextIOBase", "TextIOWrapper"]
import os
import sys
import codecs
import _fileio
import warnings
# XXX Shouldn't we use st_blksize whenever we can?
DEFAULT_BUFFER_SIZE = 8 * 1024 # bytes
class BlockingIOError(IOError):
"""Exception raised when I/O would block on a non-blocking I/O stream."""
def __init__(self, errno, strerror, characters_written=0):
IOError.__init__(self, errno, strerror)
self.characters_written = characters_written
def open(file, mode="r", buffering=None, encoding=None, newline=None):
"""Replacement for the built-in open function.
Args:
file: string giving the name of the file to be opened;
or integer file descriptor of the file to be wrapped (*).
mode: optional mode string; see below.
buffering: optional int >= 0 giving the buffer size; values
can be: 0 = unbuffered, 1 = line buffered,
larger = fully buffered.
encoding: optional string giving the text encoding.
newline: optional newlines specifier; must be None, '\n' or '\r\n';
specifies the line ending expected on input and written on
output. If None, use universal newlines on input and
use os.linesep on output.
(*) If a file descriptor is given, it is closed when the returned
I/O object is closed. If you don't want this to happen, use
os.dup() to create a duplicate file descriptor.
Mode strings characters:
'r': open for reading (default)
'w': open for writing, truncating the file first
'a': open for writing, appending to the end if the file exists
'b': binary mode
't': text mode (default)
'+': open a disk file for updating (implies reading and writing)
'U': universal newline mode (for backwards compatibility)
Constraints:
- encoding must not be given when a binary mode is given
- buffering must not be zero when a text mode is given
Returns:
Depending on the mode and buffering arguments, either a raw
binary stream, a buffered binary stream, or a buffered text
stream, open for reading and/or writing.
"""
# XXX Don't use asserts for these checks; raise TypeError or ValueError
assert isinstance(file, (basestring, int)), repr(file)
assert isinstance(mode, basestring), repr(mode)
assert buffering is None or isinstance(buffering, int), repr(buffering)
assert encoding is None or isinstance(encoding, basestring), repr(encoding)
modes = set(mode)
if modes - set("arwb+tU") or len(mode) > len(modes):
raise ValueError("invalid mode: %r" % mode)
reading = "r" in modes
writing = "w" in modes
appending = "a" in modes
updating = "+" in modes
text = "t" in modes
binary = "b" in modes
if "U" in modes:
if writing or appending:
raise ValueError("can't use U and writing mode at once")
reading = True
if text and binary:
raise ValueError("can't have text and binary mode at once")
if reading + writing + appending > 1:
raise ValueError("can't have read/write/append mode at once")
if not (reading or writing or appending):
raise ValueError("must have exactly one of read/write/append mode")
if binary and encoding is not None:
raise ValueError("binary mode doesn't take an encoding argument")
if binary and newline is not None:
raise ValueError("binary mode doesn't take a newline argument")
raw = FileIO(file,
(reading and "r" or "") +
(writing and "w" or "") +
(appending and "a" or "") +
(updating and "+" or ""))
if buffering is None:
buffering = -1
if buffering < 0:
buffering = DEFAULT_BUFFER_SIZE
# XXX Should default to line buffering if os.isatty(raw.fileno())
try:
bs = os.fstat(raw.fileno()).st_blksize
except (os.error, AttributeError):
pass
else:
if bs > 1:
buffering = bs
if buffering < 0:
raise ValueError("invalid buffering size")
if buffering == 0:
if binary:
raw._name = file
raw._mode = mode
return raw
raise ValueError("can't have unbuffered text I/O")
if updating:
buffer = BufferedRandom(raw, buffering)
elif writing or appending:
buffer = BufferedWriter(raw, buffering)
else:
assert reading
buffer = BufferedReader(raw, buffering)
if binary:
buffer.name = file
buffer.mode = mode
return buffer
text = TextIOWrapper(buffer, encoding, newline)
text.name = file
text.mode = mode
return text
class UnsupportedOperation(ValueError, IOError):
pass
class IOBase:
"""Base class for all I/O classes.
This class provides dummy implementations for many methods that
derived classes can override selectively; the default
implementations represent a file that cannot be read, written or
seeked.
This does not define read(), readinto() and write(), nor
readline() and friends, since their signatures vary per layer.
Not that calling any method (even inquiries) on a closed file is
undefined. Implementations may raise IOError in this case.
"""
### Internal ###
def _unsupported(self, name: str) -> IOError:
"""Internal: raise an exception for unsupported operations."""
raise UnsupportedOperation("%s.%s() not supported" %
(self.__class__.__name__, name))
### Positioning ###
def seek(self, pos: int, whence: int = 0) -> int:
"""seek(pos: int, whence: int = 0) -> int. Change stream position.
Seek to byte offset pos relative to position indicated by whence:
0 Start of stream (the default). pos should be >= 0;
1 Current position - whence may be negative;
2 End of stream - whence usually negative.
Returns the new absolute position.
"""
self._unsupported("seek")
def tell(self) -> int:
"""tell() -> int. Return current stream position."""
return self.seek(0, 1)
def truncate(self, pos: int = None) -> int:
"""truncate(size: int = None) -> int. Truncate file to size bytes.
Size defaults to the current IO position as reported by tell().
Returns the new size.
"""
self._unsupported("truncate")
### Flush and close ###
def flush(self) -> None:
"""flush() -> None. Flushes write buffers, if applicable.
This is a no-op for read-only and non-blocking streams.
"""
# XXX Should this return the number of bytes written???
__closed = False
def close(self) -> None:
"""close() -> None. Flushes and closes the IO object.
This must be idempotent. It should also set a flag for the
'closed' property (see below) to test.
"""
if not self.__closed:
try:
self.flush()
except IOError:
pass # If flush() fails, just give up
self.__closed = True
def __del__(self) -> None:
"""Destructor. Calls close()."""
# The try/except block is in case this is called at program
# exit time, when it's possible that globals have already been
# deleted, and then the close() call might fail. Since
# there's nothing we can do about such failures and they annoy
# the end users, we suppress the traceback.
try:
self.close()
except:
pass
### Inquiries ###
def seekable(self) -> bool:
"""seekable() -> bool. Return whether object supports random access.
If False, seek(), tell() and truncate() will raise IOError.
This method may need to do a test seek().
"""
return False
def readable(self) -> bool:
"""readable() -> bool. Return whether object was opened for reading.
If False, read() will raise IOError.
"""
return False
def writable(self) -> bool:
"""writable() -> bool. Return whether object was opened for writing.
If False, write() and truncate() will raise IOError.
"""
return False
@property
def closed(self):
"""closed: bool. True iff the file has been closed.
For backwards compatibility, this is a property, not a predicate.
"""
return self.__closed
### Context manager ###
def __enter__(self) -> "IOBase": # That's a forward reference
"""Context management protocol. Returns self."""
return self
def __exit__(self, *args) -> None:
"""Context management protocol. Calls close()"""
self.close()
### Lower-level APIs ###
# XXX Should these be present even if unimplemented?
def fileno(self) -> int:
"""fileno() -> int. Returns underlying file descriptor if one exists.
Raises IOError if the IO object does not use a file descriptor.
"""
self._unsupported("fileno")
def isatty(self) -> bool:
"""isatty() -> int. Returns whether this is an 'interactive' stream.
Returns False if we don't know.
"""
return False
### Readline[s] and writelines ###
def readline(self, limit: int = -1) -> bytes:
"""For backwards compatibility, a (slowish) readline()."""
if hasattr(self, "peek"):
def nreadahead():
readahead = self.peek(1, unsafe=True)
if not readahead:
return 1
n = (readahead.find(b"\n") + 1) or len(readahead)
if limit >= 0:
n = min(n, limit)
return n
else:
def nreadahead():
return 1
if limit is None:
limit = -1
res = bytes()
while limit < 0 or len(res) < limit:
b = self.read(nreadahead())
if not b:
break
res += b
if res.endswith(b"\n"):
break
return res
def __iter__(self):
if self.closed:
raise ValueError("__iter__ on closed file")
return self
def __next__(self):
line = self.readline()
if not line:
raise StopIteration
return line
def readlines(self, hint=None):
if hint is None:
return list(self)
n = 0
lines = []
for line in self:
lines.append(line)
n += len(line)
if n >= hint:
break
return lines
def writelines(self, lines):
if self.closed:
raise ValueError("write to closed file")
for line in lines:
self.write(line)
class RawIOBase(IOBase):
"""Base class for raw binary I/O.
The read() method is implemented by calling readinto(); derived
classes that want to support read() only need to implement
readinto() as a primitive operation. In general, readinto()
can be more efficient than read().
(It would be tempting to also provide an implementation of
readinto() in terms of read(), in case the latter is a more
suitable primitive operation, but that would lead to nasty
recursion in case a subclass doesn't implement either.)
"""
def read(self, n: int = -1) -> bytes:
"""read(n: int) -> bytes. Read and return up to n bytes.
Returns an empty bytes array on EOF, or None if the object is
set not to block and has no data to read.
"""
if n is None:
n = -1
if n < 0:
return self.readall()
b = bytes(n.__index__())
n = self.readinto(b)
del b[n:]
return b
def readall(self):
"""readall() -> bytes. Read until EOF, using multiple read() call."""
res = bytes()
while True:
data = self.read(DEFAULT_BUFFER_SIZE)
if not data:
break
res += data
return res
def readinto(self, b: bytes) -> int:
"""readinto(b: bytes) -> int. Read up to len(b) bytes into b.
Returns number of bytes read (0 for EOF), or None if the object
is set not to block as has no data to read.
"""
self._unsupported("readinto")
def write(self, b: bytes) -> int:
"""write(b: bytes) -> int. Write the given buffer to the IO stream.
Returns the number of bytes written, which may be less than len(b).
"""
self._unsupported("write")
class FileIO(_fileio._FileIO, RawIOBase):
"""Raw I/O implementation for OS files.
This multiply inherits from _FileIO and RawIOBase to make
isinstance(io.FileIO(), io.RawIOBase) return True without
requiring that _fileio._FileIO inherits from io.RawIOBase (which
would be hard to do since _fileio.c is written in C).
"""
def close(self):
_fileio._FileIO.close(self)
RawIOBase.close(self)
@property
def name(self):
return self._name
@property
def mode(self):
return self._mode
class BufferedIOBase(IOBase):
"""Base class for buffered IO objects.
The main difference with RawIOBase is that the read() method
supports omitting the size argument, and does not have a default
implementation that defers to readinto().
In addition, read(), readinto() and write() may raise
BlockingIOError if the underlying raw stream is in non-blocking
mode and not ready; unlike their raw counterparts, they will never
return None.
A typical implementation should not inherit from a RawIOBase
implementation, but wrap one.
"""
def read(self, n: int = None) -> bytes:
"""read(n: int = None) -> bytes. Read and return up to n bytes.
If the argument is omitted, None, or negative, reads and
returns all data until EOF.
If the argument is positive, and the underlying raw stream is
not 'interactive', multiple raw reads may be issued to satisfy
the byte count (unless EOF is reached first). But for
interactive raw streams (XXX and for pipes?), at most one raw
read will be issued, and a short result does not imply that
EOF is imminent.
Returns an empty bytes array on EOF.
Raises BlockingIOError if the underlying raw stream has no
data at the moment.
"""
self._unsupported("read")
def readinto(self, b: bytes) -> int:
"""readinto(b: bytes) -> int. Read up to len(b) bytes into b.
Like read(), this may issue multiple reads to the underlying
raw stream, unless the latter is 'interactive' (XXX or a
pipe?).
Returns the number of bytes read (0 for EOF).
Raises BlockingIOError if the underlying raw stream has no
data at the moment.
"""
# XXX This ought to work with anything that supports the buffer API
data = self.read(len(b))
n = len(data)
try:
b[:n] = data
except TypeError as err:
import array
if not isinstance(b, array.array):
raise err
b[:n] = array.array('b', data)
return n
def write(self, b: bytes) -> int:
"""write(b: bytes) -> int. Write the given buffer to the IO stream.
Returns the number of bytes written, which is never less than
len(b).
Raises BlockingIOError if the buffer is full and the
underlying raw stream cannot accept more data at the moment.
"""
self._unsupported("write")
class _BufferedIOMixin(BufferedIOBase):
"""A mixin implementation of BufferedIOBase with an underlying raw stream.
This passes most requests on to the underlying raw stream. It
does *not* provide implementations of read(), readinto() or
write().
"""
def __init__(self, raw):
self.raw = raw
### Positioning ###
def seek(self, pos, whence=0):
return self.raw.seek(pos, whence)
def tell(self):
return self.raw.tell()
def truncate(self, pos=None):
if pos is None:
pos = self.tell()
return self.raw.truncate(pos)
### Flush and close ###
def flush(self):
self.raw.flush()
def close(self):
if not self.closed:
try:
self.flush()
except IOError:
pass # If flush() fails, just give up
self.raw.close()
### Inquiries ###
def seekable(self):
return self.raw.seekable()
def readable(self):
return self.raw.readable()
def writable(self):
return self.raw.writable()
@property
def closed(self):
return self.raw.closed
### Lower-level APIs ###
def fileno(self):
return self.raw.fileno()
def isatty(self):
return self.raw.isatty()
class BytesIO(BufferedIOBase):
"""Buffered I/O implementation using an in-memory bytes buffer."""
# XXX More docs
def __init__(self, initial_bytes=None):
buffer = b""
if initial_bytes is not None:
buffer += initial_bytes
self._buffer = buffer
self._pos = 0
def getvalue(self):
return self._buffer
def read(self, n=None):
if n is None:
n = -1
if n < 0:
n = len(self._buffer)
newpos = min(len(self._buffer), self._pos + n)
b = self._buffer[self._pos : newpos]
self._pos = newpos
return b
def read1(self, n):
return self.read(n)
def write(self, b):
if self.closed:
raise ValueError("write to closed file")
n = len(b)
newpos = self._pos + n
if newpos > len(self._buffer):
# Inserts null bytes between the current end of the file
# and the new write position.
padding = '\x00' * (newpos - len(self._buffer) - n)
self._buffer[self._pos:newpos - n] = padding
self._buffer[self._pos:newpos] = b
self._pos = newpos
return n
def seek(self, pos, whence=0):
if whence == 0:
self._pos = max(0, pos)
elif whence == 1:
self._pos = max(0, self._pos + pos)
elif whence == 2:
self._pos = max(0, len(self._buffer) + pos)
else:
raise IOError("invalid whence value")
return self._pos
def tell(self):
return self._pos
def truncate(self, pos=None):
if pos is None:
pos = self._pos
del self._buffer[pos:]
return pos
def readable(self):
return True
def writable(self):
return True
def seekable(self):
return True
class BufferedReader(_BufferedIOMixin):
"""Buffer for a readable sequential RawIO object."""
def __init__(self, raw, buffer_size=DEFAULT_BUFFER_SIZE):
"""Create a new buffered reader using the given readable raw IO object.
"""
assert raw.readable()
_BufferedIOMixin.__init__(self, raw)
self._read_buf = b""
self.buffer_size = buffer_size
def read(self, n=None):
"""Read n bytes.
Returns exactly n bytes of data unless the underlying raw IO
stream reaches EOF or if the call would block in non-blocking
mode. If n is negative, read until EOF or until read() would
block.
"""
if n is None:
n = -1
nodata_val = b""
while n < 0 or len(self._read_buf) < n:
to_read = max(self.buffer_size,
n if n is not None else 2*len(self._read_buf))
current = self.raw.read(to_read)
if current in (b"", None):
nodata_val = current
break
self._read_buf += current
if self._read_buf:
if n < 0:
n = len(self._read_buf)
out = self._read_buf[:n]
self._read_buf = self._read_buf[n:]
else:
out = nodata_val
return out
def peek(self, n=0, *, unsafe=False):
"""Returns buffered bytes without advancing the position.
The argument indicates a desired minimal number of bytes; we
do at most one raw read to satisfy it. We never return more
than self.buffer_size.
Unless unsafe=True is passed, we return a copy.
"""
want = min(n, self.buffer_size)
have = len(self._read_buf)
if have < want:
to_read = self.buffer_size - have
current = self.raw.read(to_read)
if current:
self._read_buf += current
result = self._read_buf
if unsafe:
result = result[:]
return result
def read1(self, n):
"""Reads up to n bytes.
Returns up to n bytes. If at least one byte is buffered,
we only return buffered bytes. Otherwise, we do one
raw read.
"""
if n <= 0:
return b""
self.peek(1, unsafe=True)
return self.read(min(n, len(self._read_buf)))
def tell(self):
return self.raw.tell() - len(self._read_buf)
def seek(self, pos, whence=0):
if whence == 1:
pos -= len(self._read_buf)
pos = self.raw.seek(pos, whence)
self._read_buf = b""
return pos
class BufferedWriter(_BufferedIOMixin):
# XXX docstring
def __init__(self, raw,
buffer_size=DEFAULT_BUFFER_SIZE, max_buffer_size=None):
assert raw.writable()
_BufferedIOMixin.__init__(self, raw)
self.buffer_size = buffer_size
self.max_buffer_size = (2*buffer_size
if max_buffer_size is None
else max_buffer_size)
self._write_buf = b""
def write(self, b):
if self.closed:
raise ValueError("write to closed file")
if not isinstance(b, bytes):
if hasattr(b, "__index__"):
raise TypeError("Can't write object of type %s" %
type(b).__name__)
b = bytes(b)
# XXX we can implement some more tricks to try and avoid partial writes
if len(self._write_buf) > self.buffer_size:
# We're full, so let's pre-flush the buffer
try:
self.flush()
except BlockingIOError as e:
# We can't accept anything else.
# XXX Why not just let the exception pass through?
raise BlockingIOError(e.errno, e.strerror, 0)
before = len(self._write_buf)
self._write_buf.extend(b)
written = len(self._write_buf) - before
if len(self._write_buf) > self.buffer_size:
try:
self.flush()
except BlockingIOError as e:
if (len(self._write_buf) > self.max_buffer_size):
# We've hit max_buffer_size. We have to accept a partial
# write and cut back our buffer.
overage = len(self._write_buf) - self.max_buffer_size
self._write_buf = self._write_buf[:self.max_buffer_size]
raise BlockingIOError(e.errno, e.strerror, overage)
return written
def flush(self):
if self.closed:
raise ValueError("flush of closed file")
written = 0
try:
while self._write_buf:
n = self.raw.write(self._write_buf)
del self._write_buf[:n]
written += n
except BlockingIOError as e:
n = e.characters_written
del self._write_buf[:n]
written += n
raise BlockingIOError(e.errno, e.strerror, written)
def tell(self):
return self.raw.tell() + len(self._write_buf)
def seek(self, pos, whence=0):
self.flush()
return self.raw.seek(pos, whence)
class BufferedRWPair(BufferedIOBase):
"""A buffered reader and writer object together.
A buffered reader object and buffered writer object put together
to form a sequential IO object that can read and write.
This is typically used with a socket or two-way pipe.
XXX The usefulness of this (compared to having two separate IO
objects) is questionable.
"""
def __init__(self, reader, writer,
buffer_size=DEFAULT_BUFFER_SIZE, max_buffer_size=None):
"""Constructor.
The arguments are two RawIO instances.
"""
assert reader.readable()
assert writer.writable()
self.reader = BufferedReader(reader, buffer_size)
self.writer = BufferedWriter(writer, buffer_size, max_buffer_size)
def read(self, n=None):
if n is None:
n = -1
return self.reader.read(n)
def readinto(self, b):
return self.reader.readinto(b)
def write(self, b):
return self.writer.write(b)
def peek(self, n=0, *, unsafe=False):
return self.reader.peek(n, unsafe=unsafe)
def read1(self, n):
return self.reader.read1(n)
def readable(self):
return self.reader.readable()
def writable(self):
return self.writer.writable()
def flush(self):
return self.writer.flush()
def close(self):
self.writer.close()
self.reader.close()
def isatty(self):
return self.reader.isatty() or self.writer.isatty()
@property
def closed(self):
return self.writer.closed()
class BufferedRandom(BufferedWriter, BufferedReader):
# XXX docstring
def __init__(self, raw,
buffer_size=DEFAULT_BUFFER_SIZE, max_buffer_size=None):
assert raw.seekable()
BufferedReader.__init__(self, raw, buffer_size)
BufferedWriter.__init__(self, raw, buffer_size, max_buffer_size)
def seek(self, pos, whence=0):
self.flush()
# First do the raw seek, then empty the read buffer, so that
# if the raw seek fails, we don't lose buffered data forever.
pos = self.raw.seek(pos, whence)
self._read_buf = b""
return pos
def tell(self):
if (self._write_buf):
return self.raw.tell() + len(self._write_buf)
else:
return self.raw.tell() - len(self._read_buf)
def read(self, n=None):
if n is None:
n = -1
self.flush()
return BufferedReader.read(self, n)
def readinto(self, b):
self.flush()
return BufferedReader.readinto(self, b)
def peek(self, n=0, *, unsafe=False):
self.flush()
return BufferedReader.peek(self, n, unsafe=unsafe)
def read1(self, n):
self.flush()
return BufferedReader.read1(self, n)
def write(self, b):
if self._read_buf:
self.raw.seek(-len(self._read_buf), 1) # Undo readahead
self._read_buf = b""
return BufferedWriter.write(self, b)
class TextIOBase(IOBase):
"""Base class for text I/O.
This class provides a character and line based interface to stream I/O.
There is no readinto() method, as character strings are immutable.
"""
def read(self, n: int = -1) -> str:
"""read(n: int = -1) -> str. Read at most n characters from stream.
Read from underlying buffer until we have n characters or we hit EOF.
If n is negative or omitted, read until EOF.
"""
self._unsupported("read")
def write(self, s: str) -> int:
"""write(s: str) -> int. Write string s to stream."""
self._unsupported("write")
def truncate(self, pos: int = None) -> int:
"""truncate(pos: int = None) -> int. Truncate size to pos."""
self.flush()
if pos is None:
pos = self.tell()
self.seek(pos)
return self.buffer.truncate()
def readline(self) -> str:
"""readline() -> str. Read until newline or EOF.
Returns an empty string if EOF is hit immediately.
"""
self._unsupported("readline")
@property
def encoding(self):
"""Subclasses should override."""
return None
class TextIOWrapper(TextIOBase):
"""Buffered text stream.
Character and line based layer over a BufferedIOBase object.
"""
_CHUNK_SIZE = 128
def __init__(self, buffer, encoding=None, newline=None):
if newline not in (None, "\n", "\r\n"):
raise ValueError("illegal newline value: %r" % (newline,))
if encoding is None:
# XXX This is questionable
encoding = sys.getfilesystemencoding() or "latin-1"
self.buffer = buffer
self._encoding = encoding
self._newline = newline or os.linesep
self._fix_newlines = newline is None
self._decoder = None
self._pending = ""
self._snapshot = None
self._seekable = self._telling = self.buffer.seekable()
@property
def encoding(self):
return self._encoding
# A word about _snapshot. This attribute is either None, or a
# tuple (decoder_state, readahead, pending) where decoder_state is
# the second (integer) item of the decoder state, readahead is the
# chunk of bytes that was read, and pending is the characters that
# were rendered by the decoder after feeding it those bytes. We
# use this to reconstruct intermediate decoder states in tell().
def _seekable(self):
return self._seekable
def flush(self):
self.buffer.flush()
self._telling = self._seekable
def close(self):
try:
self.flush()
except:
pass # If flush() fails, just give up
self.buffer.close()
@property
def closed(self):
return self.buffer.closed
def fileno(self):
return self.buffer.fileno()
def isatty(self):
return self.buffer.isatty()
def write(self, s: str):
if self.closed:
raise ValueError("write to closed file")
# XXX What if we were just reading?
b = s.encode(self._encoding)
if isinstance(b, str):
b = bytes(b)
n = self.buffer.write(b)
if "\n" in s:
# XXX only if isatty
self.flush()
self._snapshot = self._decoder = None
return len(s)
def _get_decoder(self):
make_decoder = codecs.getincrementaldecoder(self._encoding)
if make_decoder is None:
raise IOError("Can't find an incremental decoder for encoding %s" %
self._encoding)
decoder = self._decoder = make_decoder() # XXX: errors
return decoder
def _read_chunk(self):
assert self._decoder is not None
if not self._telling:
readahead = self.buffer.read1(self._CHUNK_SIZE)
pending = self._decoder.decode(readahead, not readahead)
return readahead, pending
decoder_buffer, decoder_state = self._decoder.getstate()
readahead = self.buffer.read1(self._CHUNK_SIZE)
pending = self._decoder.decode(readahead, not readahead)
self._snapshot = (decoder_state, decoder_buffer + readahead, pending)
return readahead, pending
def _encode_decoder_state(self, ds, pos):
x = 0
for i in bytes(ds):
x = x<<8 | i
return (x<<64) | pos
def _decode_decoder_state(self, pos):
x, pos = divmod(pos, 1<<64)
if not x:
return None, pos
b = b""
while x:
b.append(x&0xff)
x >>= 8
return str(b[::-1]), pos
def tell(self):
if not self._seekable:
raise IOError("Underlying stream is not seekable")
if not self._telling:
raise IOError("Telling position disabled by next() call")
self.flush()
position = self.buffer.tell()
decoder = self._decoder
if decoder is None or self._snapshot is None:
assert self._pending == ""
return position
decoder_state, readahead, pending = self._snapshot
position -= len(readahead)
needed = len(pending) - len(self._pending)
if not needed:
return self._encode_decoder_state(decoder_state, position)
saved_state = decoder.getstate()
try:
decoder.setstate((b"", decoder_state))
n = 0
bb = bytes(1)
for i, bb[0] in enumerate(readahead):
n += len(decoder.decode(bb))
if n >= needed:
decoder_buffer, decoder_state = decoder.getstate()
return self._encode_decoder_state(
decoder_state,
position + (i+1) - len(decoder_buffer))
raise IOError("Can't reconstruct logical file position")
finally:
decoder.setstate(saved_state)
def seek(self, pos, whence=0):
if not self._seekable:
raise IOError("Underlying stream is not seekable")
if whence == 1:
if pos != 0:
raise IOError("Can't do nonzero cur-relative seeks")
pos = self.tell()
whence = 0
if whence == 2:
if pos != 0:
raise IOError("Can't do nonzero end-relative seeks")
self.flush()
pos = self.buffer.seek(0, 2)
self._snapshot = None
self._pending = ""
self._decoder = None
return pos
if whence != 0:
raise ValueError("Invalid whence (%r, should be 0, 1 or 2)" %
(whence,))
if pos < 0:
raise ValueError("Negative seek position %r" % (pos,))
self.flush()
orig_pos = pos
ds, pos = self._decode_decoder_state(pos)
if not ds:
self.buffer.seek(pos)
self._snapshot = None
self._pending = ""
self._decoder = None
return pos
decoder = self._decoder or self._get_decoder()
decoder.set_state(("", ds))
self.buffer.seek(pos)
self._snapshot = (ds, b"", "")
self._pending = ""
self._decoder = decoder
return orig_pos
def read(self, n=None):
if n is None:
n = -1
decoder = self._decoder or self._get_decoder()
res = self._pending
if n < 0:
res += decoder.decode(self.buffer.read(), True)
self._pending = ""
self._snapshot = None
return res.replace("\r\n", "\n")
else:
while len(res) < n:
readahead, pending = self._read_chunk()
res += pending
if not readahead:
break
self._pending = res[n:]
return res[:n].replace("\r\n", "\n")
def __next__(self):
self._telling = False
line = self.readline()
if not line:
self._snapshot = None
self._telling = self._seekable
raise StopIteration
return line
def readline(self, limit=None):
if limit is not None:
# XXX Hack to support limit argument, for backwards compatibility
line = self.readline()
if len(line) <= limit:
return line
line, self._pending = line[:limit], line[limit:] + self._pending
return line
line = self._pending
start = 0
decoder = self._decoder or self._get_decoder()
while True:
# In C we'd look for these in parallel of course.
nlpos = line.find("\n", start)
crpos = line.find("\r", start)
if nlpos >= 0 and crpos >= 0:
endpos = min(nlpos, crpos)
else:
endpos = nlpos if nlpos >= 0 else crpos
if endpos != -1:
endc = line[endpos]
if endc == "\n":
ending = "\n"
break
# We've seen \r - is it standalone, \r\n or \r at end of line?
if endpos + 1 < len(line):
if line[endpos+1] == "\n":
ending = "\r\n"
else:
ending = "\r"
break
# There might be a following \n in the next block of data ...
start = endpos
else:
start = len(line)
# No line ending seen yet - get more data
while True:
readahead, pending = self._read_chunk()
more_line = pending
if more_line or not readahead:
break
if not more_line:
ending = ""
endpos = len(line)
break
line += more_line
nextpos = endpos + len(ending)
self._pending = line[nextpos:]
# XXX Update self.newlines here if we want to support that
if self._fix_newlines and ending not in ("\n", ""):
return line[:endpos] + "\n"
else:
return line[:nextpos]
class StringIO(TextIOWrapper):
# XXX This is really slow, but fully functional
def __init__(self, initial_value="", encoding="utf-8", newline=None):
super(StringIO, self).__init__(BytesIO(),
encoding=encoding,
newline=newline)
if initial_value:
self.write(initial_value)
self.seek(0)
def getvalue(self):
return self.buffer.getvalue().decode(self._encoding)
|