This repository was archived by the owner on Nov 17, 2020. It is now read-only.
forked from glix/python
-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy path_pickle.c
More file actions
4677 lines (4013 loc) · 126 KB
/
_pickle.c
File metadata and controls
4677 lines (4013 loc) · 126 KB
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
#include "Python.h"
#include "structmember.h"
PyDoc_STRVAR(pickle_module_doc,
"Optimized C implementation for the Python pickle module.");
/* Bump this when new opcodes are added to the pickle protocol. */
enum {
HIGHEST_PROTOCOL = 3,
DEFAULT_PROTOCOL = 3
};
/* Pickle opcodes. These must be kept updated with pickle.py.
Extensive docs are in pickletools.py. */
enum opcode {
MARK = '(',
STOP = '.',
POP = '0',
POP_MARK = '1',
DUP = '2',
FLOAT = 'F',
INT = 'I',
BININT = 'J',
BININT1 = 'K',
LONG = 'L',
BININT2 = 'M',
NONE = 'N',
PERSID = 'P',
BINPERSID = 'Q',
REDUCE = 'R',
STRING = 'S',
BINSTRING = 'T',
SHORT_BINSTRING = 'U',
UNICODE = 'V',
BINUNICODE = 'X',
APPEND = 'a',
BUILD = 'b',
GLOBAL = 'c',
DICT = 'd',
EMPTY_DICT = '}',
APPENDS = 'e',
GET = 'g',
BINGET = 'h',
INST = 'i',
LONG_BINGET = 'j',
LIST = 'l',
EMPTY_LIST = ']',
OBJ = 'o',
PUT = 'p',
BINPUT = 'q',
LONG_BINPUT = 'r',
SETITEM = 's',
TUPLE = 't',
EMPTY_TUPLE = ')',
SETITEMS = 'u',
BINFLOAT = 'G',
/* Protocol 2. */
PROTO = '\x80',
NEWOBJ = '\x81',
EXT1 = '\x82',
EXT2 = '\x83',
EXT4 = '\x84',
TUPLE1 = '\x85',
TUPLE2 = '\x86',
TUPLE3 = '\x87',
NEWTRUE = '\x88',
NEWFALSE = '\x89',
LONG1 = '\x8a',
LONG4 = '\x8b',
/* Protocol 3 (Python 3.x) */
BINBYTES = 'B',
SHORT_BINBYTES = 'C',
};
/* These aren't opcodes -- they're ways to pickle bools before protocol 2
* so that unpicklers written before bools were introduced unpickle them
* as ints, but unpicklers after can recognize that bools were intended.
* Note that protocol 2 added direct ways to pickle bools.
*/
#undef TRUE
#define TRUE "I01\n"
#undef FALSE
#define FALSE "I00\n"
enum {
/* Keep in synch with pickle.Pickler._BATCHSIZE. This is how many elements
batch_list/dict() pumps out before doing APPENDS/SETITEMS. Nothing will
break if this gets out of synch with pickle.py, but it's unclear that would
help anything either. */
BATCHSIZE = 1000,
/* Nesting limit until Pickler, when running in "fast mode", starts
checking for self-referential data-structures. */
FAST_NESTING_LIMIT = 50,
/* Size of the write buffer of Pickler. Higher values will reduce the
number of calls to the write() method of the output stream. */
WRITE_BUF_SIZE = 256,
};
/* Exception classes for pickle. These should override the ones defined in
pickle.py, when the C-optimized Pickler and Unpickler are used. */
static PyObject *PickleError;
static PyObject *PicklingError;
static PyObject *UnpicklingError;
/* copyreg.dispatch_table, {type_object: pickling_function} */
static PyObject *dispatch_table;
/* For EXT[124] opcodes. */
/* copyreg._extension_registry, {(module_name, function_name): code} */
static PyObject *extension_registry;
/* copyreg._inverted_registry, {code: (module_name, function_name)} */
static PyObject *inverted_registry;
/* copyreg._extension_cache, {code: object} */
static PyObject *extension_cache;
/* XXX: Are these really nescessary? */
/* As the name says, an empty tuple. */
static PyObject *empty_tuple;
/* For looking up name pairs in copyreg._extension_registry. */
static PyObject *two_tuple;
static int
stack_underflow(void)
{
PyErr_SetString(UnpicklingError, "unpickling stack underflow");
return -1;
}
/* Internal data type used as the unpickling stack. */
typedef struct {
PyObject_HEAD
int length; /* number of initial slots in data currently used */
int size; /* number of slots in data allocated */
PyObject **data;
} Pdata;
static void
Pdata_dealloc(Pdata *self)
{
int i;
PyObject **p;
for (i = self->length, p = self->data; --i >= 0; p++) {
Py_DECREF(*p);
}
if (self->data)
PyMem_Free(self->data);
PyObject_Del(self);
}
static PyTypeObject Pdata_Type = {
PyVarObject_HEAD_INIT(NULL, 0)
"_pickle.Pdata", /*tp_name*/
sizeof(Pdata), /*tp_basicsize*/
0, /*tp_itemsize*/
(destructor)Pdata_dealloc, /*tp_dealloc*/
};
static PyObject *
Pdata_New(void)
{
Pdata *self;
if (!(self = PyObject_New(Pdata, &Pdata_Type)))
return NULL;
self->size = 8;
self->length = 0;
self->data = PyMem_Malloc(self->size * sizeof(PyObject *));
if (self->data)
return (PyObject *)self;
Py_DECREF(self);
return PyErr_NoMemory();
}
/* Retain only the initial clearto items. If clearto >= the current
* number of items, this is a (non-erroneous) NOP.
*/
static int
Pdata_clear(Pdata *self, int clearto)
{
int i;
PyObject **p;
if (clearto < 0)
return stack_underflow();
if (clearto >= self->length)
return 0;
for (i = self->length, p = self->data + clearto; --i >= clearto; p++) {
Py_CLEAR(*p);
}
self->length = clearto;
return 0;
}
static int
Pdata_grow(Pdata *self)
{
int bigger;
size_t nbytes;
PyObject **tmp;
bigger = (self->size << 1) + 1;
if (bigger <= 0) /* was 0, or new value overflows */
goto nomemory;
if ((int)(size_t)bigger != bigger)
goto nomemory;
nbytes = (size_t)bigger * sizeof(PyObject *);
if (nbytes / sizeof(PyObject *) != (size_t)bigger)
goto nomemory;
tmp = PyMem_Realloc(self->data, nbytes);
if (tmp == NULL)
goto nomemory;
self->data = tmp;
self->size = bigger;
return 0;
nomemory:
PyErr_NoMemory();
return -1;
}
/* D is a Pdata*. Pop the topmost element and store it into V, which
* must be an lvalue holding PyObject*. On stack underflow, UnpicklingError
* is raised and V is set to NULL.
*/
static PyObject *
Pdata_pop(Pdata *self)
{
if (self->length == 0) {
PyErr_SetString(UnpicklingError, "bad pickle data");
return NULL;
}
return self->data[--(self->length)];
}
#define PDATA_POP(D, V) do { (V) = Pdata_pop((D)); } while (0)
static int
Pdata_push(Pdata *self, PyObject *obj)
{
if (self->length == self->size && Pdata_grow(self) < 0) {
return -1;
}
self->data[self->length++] = obj;
return 0;
}
/* Push an object on stack, transferring its ownership to the stack. */
#define PDATA_PUSH(D, O, ER) do { \
if (Pdata_push((D), (O)) < 0) return (ER); } while(0)
/* Push an object on stack, adding a new reference to the object. */
#define PDATA_APPEND(D, O, ER) do { \
Py_INCREF((O)); \
if (Pdata_push((D), (O)) < 0) return (ER); } while(0)
static PyObject *
Pdata_poptuple(Pdata *self, Py_ssize_t start)
{
PyObject *tuple;
Py_ssize_t len, i, j;
len = self->length - start;
tuple = PyTuple_New(len);
if (tuple == NULL)
return NULL;
for (i = start, j = 0; j < len; i++, j++)
PyTuple_SET_ITEM(tuple, j, self->data[i]);
self->length = start;
return tuple;
}
static PyObject *
Pdata_poplist(Pdata *self, Py_ssize_t start)
{
PyObject *list;
Py_ssize_t len, i, j;
len = self->length - start;
list = PyList_New(len);
if (list == NULL)
return NULL;
for (i = start, j = 0; j < len; i++, j++)
PyList_SET_ITEM(list, j, self->data[i]);
self->length = start;
return list;
}
typedef struct PicklerObject {
PyObject_HEAD
PyObject *write; /* write() method of the output stream */
PyObject *memo; /* Memo dictionary, keep track of the seen
objects to support self-referential objects
pickling. */
PyObject *pers_func; /* persistent_id() method, can be NULL */
PyObject *arg;
int proto; /* Pickle protocol number, >= 0 */
int bin; /* Boolean, true if proto > 0 */
int buf_size; /* Size of the current buffered pickle data */
char *write_buf; /* Write buffer, this is to avoid calling the
write() method of the output stream too
often. */
int fast; /* Enable fast mode if set to a true value.
The fast mode disable the usage of memo,
therefore speeding the pickling process by
not generating superfluous PUT opcodes. It
should not be used if with self-referential
objects. */
int fast_nesting;
PyObject *fast_memo;
} PicklerObject;
typedef struct UnpicklerObject {
PyObject_HEAD
Pdata *stack; /* Pickle data stack, store unpickled objects. */
PyObject *readline; /* readline() method of the output stream */
PyObject *read; /* read() method of the output stream */
PyObject *memo; /* Memo dictionary, provide the objects stored
using the PUT opcodes. */
PyObject *arg;
PyObject *pers_func; /* persistent_load() method, can be NULL. */
PyObject *last_string; /* Reference to the last string read by the
readline() method. */
char *buffer; /* Reading buffer. */
char *encoding; /* Name of the encoding to be used for
decoding strings pickled using Python
2.x. The default value is "ASCII" */
char *errors; /* Name of errors handling scheme to used when
decoding strings. The default value is
"strict". */
int *marks; /* Mark stack, used for unpickling container
objects. */
Py_ssize_t num_marks; /* Number of marks in the mark stack. */
Py_ssize_t marks_size; /* Current allocated size of the mark stack. */
} UnpicklerObject;
/* Forward declarations */
static int save(PicklerObject *, PyObject *, int);
static int save_reduce(PicklerObject *, PyObject *, PyObject *);
static PyTypeObject Pickler_Type;
static PyTypeObject Unpickler_Type;
/* Helpers for creating the argument tuple passed to functions. This has the
performance advantage of calling PyTuple_New() only once. */
#define ARG_TUP(self, obj) do { \
if ((self)->arg || ((self)->arg=PyTuple_New(1))) { \
Py_XDECREF(PyTuple_GET_ITEM((self)->arg, 0)); \
PyTuple_SET_ITEM((self)->arg, 0, (obj)); \
} \
else { \
Py_DECREF((obj)); \
} \
} while (0)
#define FREE_ARG_TUP(self) do { \
if ((self)->arg->ob_refcnt > 1) \
Py_CLEAR((self)->arg); \
} while (0)
/* A temporary cleaner API for fast single argument function call.
XXX: Does caching the argument tuple provides any real performance benefits?
A quick benchmark, on a 2.0GHz Athlon64 3200+ running Linux 2.6.24 with
glibc 2.7, tells me that it takes roughly 20,000,000 PyTuple_New(1) calls
when the tuple is retrieved from the freelist (i.e, call PyTuple_New() then
immediately DECREF it) and 1,200,000 calls when allocating brand new tuples
(i.e, call PyTuple_New() and store the returned value in an array), to save
one second (wall clock time). Either ways, the loading time a pickle stream
large enough to generate this number of calls would be massively
overwhelmed by other factors, like I/O throughput, the GC traversal and
object allocation overhead. So, I really doubt these functions provide any
real benefits.
On the other hand, oprofile reports that pickle spends a lot of time in
these functions. But, that is probably more related to the function call
overhead, than the argument tuple allocation.
XXX: And, what is the reference behavior of these? Steal, borrow? At first
glance, it seems to steal the reference of 'arg' and borrow the reference
of 'func'.
*/
static PyObject *
pickler_call(PicklerObject *self, PyObject *func, PyObject *arg)
{
PyObject *result = NULL;
ARG_TUP(self, arg);
if (self->arg) {
result = PyObject_Call(func, self->arg, NULL);
FREE_ARG_TUP(self);
}
return result;
}
static PyObject *
unpickler_call(UnpicklerObject *self, PyObject *func, PyObject *arg)
{
PyObject *result = NULL;
ARG_TUP(self, arg);
if (self->arg) {
result = PyObject_Call(func, self->arg, NULL);
FREE_ARG_TUP(self);
}
return result;
}
static Py_ssize_t
pickler_write(PicklerObject *self, const char *s, Py_ssize_t n)
{
PyObject *data, *result;
if (self->write_buf == NULL) {
PyErr_SetString(PyExc_SystemError, "invalid write buffer");
return -1;
}
if (s == NULL) {
if (!(self->buf_size))
return 0;
data = PyBytes_FromStringAndSize(self->write_buf, self->buf_size);
if (data == NULL)
return -1;
}
else {
if (self->buf_size && (n + self->buf_size) > WRITE_BUF_SIZE) {
if (pickler_write(self, NULL, 0) < 0)
return -1;
}
if (n > WRITE_BUF_SIZE) {
if (!(data = PyBytes_FromStringAndSize(s, n)))
return -1;
}
else {
memcpy(self->write_buf + self->buf_size, s, n);
self->buf_size += n;
return n;
}
}
/* object with write method */
result = pickler_call(self, self->write, data);
if (result == NULL)
return -1;
Py_DECREF(result);
self->buf_size = 0;
return n;
}
/* XXX: These read/readline functions ought to be optimized. Buffered I/O
might help a lot, especially with the new (but much slower) io library.
On the other hand, the added complexity might not worth it.
*/
/* Read at least n characters from the input stream and set s to the current
reading position. */
static Py_ssize_t
unpickler_read(UnpicklerObject *self, char **s, Py_ssize_t n)
{
PyObject *len;
PyObject *data;
len = PyLong_FromSsize_t(n);
if (len == NULL)
return -1;
data = unpickler_call(self, self->read, len);
if (data == NULL)
return -1;
/* XXX: Should bytearray be supported too? */
if (!PyBytes_Check(data)) {
PyErr_SetString(PyExc_ValueError,
"read() from the underlying stream did not"
"return bytes");
Py_DECREF(data);
return -1;
}
if (PyBytes_GET_SIZE(data) != n) {
PyErr_SetNone(PyExc_EOFError);
Py_DECREF(data);
return -1;
}
Py_XDECREF(self->last_string);
self->last_string = data;
if (!(*s = PyBytes_AS_STRING(data)))
return -1;
return n;
}
static Py_ssize_t
unpickler_readline(UnpicklerObject *self, char **s)
{
PyObject *data;
data = PyObject_CallObject(self->readline, empty_tuple);
if (data == NULL)
return -1;
/* XXX: Should bytearray be supported too? */
if (!PyBytes_Check(data)) {
PyErr_SetString(PyExc_ValueError,
"readline() from the underlying stream did not"
"return bytes");
return -1;
}
Py_XDECREF(self->last_string);
self->last_string = data;
if (!(*s = PyBytes_AS_STRING(data)))
return -1;
return PyBytes_GET_SIZE(data);
}
/* Generate a GET opcode for an object stored in the memo. The 'key' argument
should be the address of the object as returned by PyLong_FromVoidPtr(). */
static int
memo_get(PicklerObject *self, PyObject *key)
{
PyObject *value;
PyObject *memo_id;
long x;
char pdata[30];
int len;
value = PyDict_GetItemWithError(self->memo, key);
if (value == NULL) {
if (!PyErr_Occurred())
PyErr_SetObject(PyExc_KeyError, key);
return -1;
}
memo_id = PyTuple_GetItem(value, 0);
if (memo_id == NULL)
return -1;
if (!PyLong_Check(memo_id)) {
PyErr_SetString(PicklingError, "memo id must be an integer");
return -1;
}
x = PyLong_AsLong(memo_id);
if (x == -1 && PyErr_Occurred())
return -1;
if (!self->bin) {
pdata[0] = GET;
PyOS_snprintf(pdata + 1, sizeof(pdata) - 1, "%ld\n", x);
len = (int)strlen(pdata);
}
else {
if (x < 256) {
pdata[0] = BINGET;
pdata[1] = (unsigned char)(x & 0xff);
len = 2;
}
else if (x <= 0xffffffffL) {
pdata[0] = LONG_BINGET;
pdata[1] = (unsigned char)(x & 0xff);
pdata[2] = (unsigned char)((x >> 8) & 0xff);
pdata[3] = (unsigned char)((x >> 16) & 0xff);
pdata[4] = (unsigned char)((x >> 24) & 0xff);
len = 5;
}
else { /* unlikely */
PyErr_SetString(PicklingError,
"memo id too large for LONG_BINGET");
return -1;
}
}
if (pickler_write(self, pdata, len) < 0)
return -1;
return 0;
}
/* Store an object in the memo, assign it a new unique ID based on the number
of objects currently stored in the memo and generate a PUT opcode. */
static int
memo_put(PicklerObject *self, PyObject *obj)
{
PyObject *key = NULL;
PyObject *memo_id = NULL;
PyObject *tuple = NULL;
long x;
char pdata[30];
int len;
int status = 0;
if (self->fast)
return 0;
key = PyLong_FromVoidPtr(obj);
if (key == NULL)
goto error;
if ((x = PyDict_Size(self->memo)) < 0)
goto error;
memo_id = PyLong_FromLong(x);
if (memo_id == NULL)
goto error;
tuple = PyTuple_New(2);
if (tuple == NULL)
goto error;
Py_INCREF(memo_id);
PyTuple_SET_ITEM(tuple, 0, memo_id);
Py_INCREF(obj);
PyTuple_SET_ITEM(tuple, 1, obj);
if (PyDict_SetItem(self->memo, key, tuple) < 0)
goto error;
if (!self->bin) {
pdata[0] = PUT;
PyOS_snprintf(pdata + 1, sizeof(pdata) - 1, "%ld\n", x);
len = strlen(pdata);
}
else {
if (x < 256) {
pdata[0] = BINPUT;
pdata[1] = (unsigned char)x;
len = 2;
}
else if (x <= 0xffffffffL) {
pdata[0] = LONG_BINPUT;
pdata[1] = (unsigned char)(x & 0xff);
pdata[2] = (unsigned char)((x >> 8) & 0xff);
pdata[3] = (unsigned char)((x >> 16) & 0xff);
pdata[4] = (unsigned char)((x >> 24) & 0xff);
len = 5;
}
else { /* unlikely */
PyErr_SetString(PicklingError,
"memo id too large for LONG_BINPUT");
return -1;
}
}
if (pickler_write(self, pdata, len) < 0)
goto error;
if (0) {
error:
status = -1;
}
Py_XDECREF(key);
Py_XDECREF(memo_id);
Py_XDECREF(tuple);
return status;
}
static PyObject *
whichmodule(PyObject *global, PyObject *global_name)
{
Py_ssize_t i, j;
static PyObject *module_str = NULL;
static PyObject *main_str = NULL;
PyObject *module_name;
PyObject *modules_dict;
PyObject *module;
PyObject *obj;
if (module_str == NULL) {
module_str = PyUnicode_InternFromString("__module__");
if (module_str == NULL)
return NULL;
main_str = PyUnicode_InternFromString("__main__");
if (main_str == NULL)
return NULL;
}
module_name = PyObject_GetAttr(global, module_str);
/* In some rare cases (e.g., random.getrandbits), __module__ can be
None. If it is so, then search sys.modules for the module of
global. */
if (module_name == Py_None) {
Py_DECREF(module_name);
goto search;
}
if (module_name) {
return module_name;
}
if (PyErr_ExceptionMatches(PyExc_AttributeError))
PyErr_Clear();
else
return NULL;
search:
modules_dict = PySys_GetObject("modules");
if (modules_dict == NULL)
return NULL;
i = 0;
module_name = NULL;
while ((j = PyDict_Next(modules_dict, &i, &module_name, &module))) {
if (PyObject_RichCompareBool(module_name, main_str, Py_EQ) == 1)
continue;
obj = PyObject_GetAttr(module, global_name);
if (obj == NULL) {
if (PyErr_ExceptionMatches(PyExc_AttributeError))
PyErr_Clear();
else
return NULL;
continue;
}
if (obj != global) {
Py_DECREF(obj);
continue;
}
Py_DECREF(obj);
break;
}
/* If no module is found, use __main__. */
if (!j) {
module_name = main_str;
}
Py_INCREF(module_name);
return module_name;
}
/* fast_save_enter() and fast_save_leave() are guards against recursive
objects when Pickler is used with the "fast mode" (i.e., with object
memoization disabled). If the nesting of a list or dict object exceed
FAST_NESTING_LIMIT, these guards will start keeping an internal
reference to the seen list or dict objects and check whether these objects
are recursive. These are not strictly necessary, since save() has a
hard-coded recursion limit, but they give a nicer error message than the
typical RuntimeError. */
static int
fast_save_enter(PicklerObject *self, PyObject *obj)
{
/* if fast_nesting < 0, we're doing an error exit. */
if (++self->fast_nesting >= FAST_NESTING_LIMIT) {
PyObject *key = NULL;
if (self->fast_memo == NULL) {
self->fast_memo = PyDict_New();
if (self->fast_memo == NULL) {
self->fast_nesting = -1;
return 0;
}
}
key = PyLong_FromVoidPtr(obj);
if (key == NULL)
return 0;
if (PyDict_GetItem(self->fast_memo, key)) {
Py_DECREF(key);
PyErr_Format(PyExc_ValueError,
"fast mode: can't pickle cyclic objects "
"including object type %.200s at %p",
obj->ob_type->tp_name, obj);
self->fast_nesting = -1;
return 0;
}
if (PyDict_SetItem(self->fast_memo, key, Py_None) < 0) {
Py_DECREF(key);
self->fast_nesting = -1;
return 0;
}
Py_DECREF(key);
}
return 1;
}
static int
fast_save_leave(PicklerObject *self, PyObject *obj)
{
if (self->fast_nesting-- >= FAST_NESTING_LIMIT) {
PyObject *key = PyLong_FromVoidPtr(obj);
if (key == NULL)
return 0;
if (PyDict_DelItem(self->fast_memo, key) < 0) {
Py_DECREF(key);
return 0;
}
Py_DECREF(key);
}
return 1;
}
static int
save_none(PicklerObject *self, PyObject *obj)
{
const char none_op = NONE;
if (pickler_write(self, &none_op, 1) < 0)
return -1;
return 0;
}
static int
save_bool(PicklerObject *self, PyObject *obj)
{
static const char *buf[2] = { FALSE, TRUE };
const char len[2] = {sizeof(FALSE) - 1, sizeof(TRUE) - 1};
int p = (obj == Py_True);
if (self->proto >= 2) {
const char bool_op = p ? NEWTRUE : NEWFALSE;
if (pickler_write(self, &bool_op, 1) < 0)
return -1;
}
else if (pickler_write(self, buf[p], len[p]) < 0)
return -1;
return 0;
}
static int
save_int(PicklerObject *self, long x)
{
char pdata[32];
int len = 0;
if (!self->bin
#if SIZEOF_LONG > 4
|| x > 0x7fffffffL || x < -0x80000000L
#endif
) {
/* Text-mode pickle, or long too big to fit in the 4-byte
* signed BININT format: store as a string.
*/
pdata[0] = LONG; /* use LONG for consistency with pickle.py */
PyOS_snprintf(pdata + 1, sizeof(pdata) - 1, "%ldL\n", x);
if (pickler_write(self, pdata, strlen(pdata)) < 0)
return -1;
}
else {
/* Binary pickle and x fits in a signed 4-byte int. */
pdata[1] = (unsigned char)(x & 0xff);
pdata[2] = (unsigned char)((x >> 8) & 0xff);
pdata[3] = (unsigned char)((x >> 16) & 0xff);
pdata[4] = (unsigned char)((x >> 24) & 0xff);
if ((pdata[4] == 0) && (pdata[3] == 0)) {
if (pdata[2] == 0) {
pdata[0] = BININT1;
len = 2;
}
else {
pdata[0] = BININT2;
len = 3;
}
}
else {
pdata[0] = BININT;
len = 5;
}
if (pickler_write(self, pdata, len) < 0)
return -1;
}
return 0;
}
static int
save_long(PicklerObject *self, PyObject *obj)
{
PyObject *repr = NULL;
Py_ssize_t size;
long val = PyLong_AsLong(obj);
int status = 0;
const char long_op = LONG;
if (val == -1 && PyErr_Occurred()) {
/* out of range for int pickling */
PyErr_Clear();
}
else
return save_int(self, val);
if (self->proto >= 2) {
/* Linear-time pickling. */
size_t nbits;
size_t nbytes;
unsigned char *pdata;
char header[5];
int i;
int sign = _PyLong_Sign(obj);
if (sign == 0) {
header[0] = LONG1;
header[1] = 0; /* It's 0 -- an empty bytestring. */
if (pickler_write(self, header, 2) < 0)
goto error;
return 0;
}
nbits = _PyLong_NumBits(obj);
if (nbits == (size_t)-1 && PyErr_Occurred())
goto error;
/* How many bytes do we need? There are nbits >> 3 full
* bytes of data, and nbits & 7 leftover bits. If there
* are any leftover bits, then we clearly need another
* byte. Wnat's not so obvious is that we *probably*
* need another byte even if there aren't any leftovers:
* the most-significant bit of the most-significant byte
* acts like a sign bit, and it's usually got a sense
* opposite of the one we need. The exception is longs
* of the form -(2**(8*j-1)) for j > 0. Such a long is
* its own 256's-complement, so has the right sign bit
* even without the extra byte. That's a pain to check
* for in advance, though, so we always grab an extra
* byte at the start, and cut it back later if possible.
*/
nbytes = (nbits >> 3) + 1;
if (nbytes > INT_MAX) {
PyErr_SetString(PyExc_OverflowError,
"long too large to pickle");
goto error;
}
repr = PyBytes_FromStringAndSize(NULL, (Py_ssize_t)nbytes);
if (repr == NULL)
goto error;
pdata = (unsigned char *)PyBytes_AS_STRING(repr);
i = _PyLong_AsByteArray((PyLongObject *)obj,
pdata, nbytes,
1 /* little endian */ , 1 /* signed */ );
if (i < 0)
goto error;
/* If the long is negative, this may be a byte more than
* needed. This is so iff the MSB is all redundant sign
* bits.
*/
if (sign < 0 &&
nbytes > 1 &&
pdata[nbytes - 1] == 0xff &&
(pdata[nbytes - 2] & 0x80) != 0) {
nbytes--;
}
if (nbytes < 256) {
header[0] = LONG1;
header[1] = (unsigned char)nbytes;
size = 2;
}
else {
header[0] = LONG4;
size = (int)nbytes;
for (i = 1; i < 5; i++) {
header[i] = (unsigned char)(size & 0xff);
size >>= 8;
}
size = 5;
}
if (pickler_write(self, header, size) < 0 ||
pickler_write(self, (char *)pdata, (int)nbytes) < 0)
goto error;
}
else {
char *string;
/* proto < 2: write the repr and newline. This is quadratic-time (in
the number of digits), in both directions. We add a trailing 'L'
to the repr, for compatibility with Python 2.x. */
repr = PyObject_Repr(obj);
if (repr == NULL)
goto error;
string = _PyUnicode_AsStringAndSize(repr, &size);
if (string == NULL)
goto error;
if (pickler_write(self, &long_op, 1) < 0 ||
pickler_write(self, string, size) < 0 ||
pickler_write(self, "L\n", 2) < 0)
goto error;
}
if (0) {
error:
status = -1;