2000-02-04 23:10:34 +08:00
|
|
|
"""Create portable serialized representations of Python objects.
|
1997-12-06 03:42:42 +08:00
|
|
|
|
|
|
|
See module cPickle for a (much) faster implementation.
|
|
|
|
See module copy_reg for a mechanism for registering custom picklers.
|
2003-01-28 04:16:36 +08:00
|
|
|
See module pickletools source for extensive comments.
|
1997-12-06 03:42:42 +08:00
|
|
|
|
|
|
|
Classes:
|
|
|
|
|
|
|
|
Pickler
|
|
|
|
Unpickler
|
|
|
|
|
|
|
|
Functions:
|
|
|
|
|
|
|
|
dump(object, file)
|
|
|
|
dumps(object) -> string
|
|
|
|
load(file) -> object
|
|
|
|
loads(string) -> object
|
|
|
|
|
|
|
|
Misc variables:
|
|
|
|
|
1998-02-13 11:24:48 +08:00
|
|
|
__version__
|
1997-12-06 03:42:42 +08:00
|
|
|
format_version
|
|
|
|
compatible_formats
|
|
|
|
|
1995-01-10 08:31:14 +08:00
|
|
|
"""
|
|
|
|
|
1998-09-16 04:25:57 +08:00
|
|
|
__version__ = "$Revision$" # Code version
|
1995-01-10 08:31:14 +08:00
|
|
|
|
|
|
|
from types import *
|
1997-09-13 04:07:24 +08:00
|
|
|
from copy_reg import dispatch_table, safe_constructors
|
1998-10-23 04:15:36 +08:00
|
|
|
import marshal
|
|
|
|
import sys
|
|
|
|
import struct
|
2001-02-18 11:10:09 +08:00
|
|
|
import re
|
1995-01-10 08:31:14 +08:00
|
|
|
|
2001-02-08 07:14:30 +08:00
|
|
|
__all__ = ["PickleError", "PicklingError", "UnpicklingError", "Pickler",
|
|
|
|
"Unpickler", "dump", "dumps", "load", "loads"]
|
|
|
|
|
2003-01-28 06:47:53 +08:00
|
|
|
# These are purely informational; no code usues these
|
|
|
|
format_version = "2.0" # File format version we write
|
|
|
|
compatible_formats = ["1.0", # Original protocol 0
|
|
|
|
"1.1", # Protocol 0 with class supprt added
|
|
|
|
"1.2", # Original protocol 1
|
|
|
|
"1.3", # Protocol 1 with BINFLOAT added
|
|
|
|
"2.0", # Protocol 2
|
|
|
|
] # Old format versions we can read
|
1995-03-14 23:09:05 +08:00
|
|
|
|
1997-04-10 01:32:51 +08:00
|
|
|
mdumps = marshal.dumps
|
|
|
|
mloads = marshal.loads
|
1995-03-09 22:08:35 +08:00
|
|
|
|
2002-05-30 00:18:42 +08:00
|
|
|
class PickleError(Exception):
|
2002-05-30 20:12:04 +08:00
|
|
|
"""A common base class for the other pickling exceptions."""
|
2002-05-30 00:18:42 +08:00
|
|
|
pass
|
|
|
|
|
|
|
|
class PicklingError(PickleError):
|
|
|
|
"""This exception is raised when an unpicklable object is passed to the
|
|
|
|
dump() method.
|
|
|
|
|
|
|
|
"""
|
|
|
|
pass
|
|
|
|
|
|
|
|
class UnpicklingError(PickleError):
|
|
|
|
"""This exception is raised when there is a problem unpickling an object,
|
|
|
|
such as a security violation.
|
|
|
|
|
|
|
|
Note that other exceptions may also be raised during unpickling, including
|
|
|
|
(but not necessarily limited to) AttributeError, EOFError, ImportError,
|
|
|
|
and IndexError.
|
|
|
|
|
|
|
|
"""
|
|
|
|
pass
|
1997-04-10 01:32:51 +08:00
|
|
|
|
2000-12-14 02:11:56 +08:00
|
|
|
class _Stop(Exception):
|
|
|
|
def __init__(self, value):
|
|
|
|
self.value = value
|
|
|
|
|
1998-05-28 06:38:22 +08:00
|
|
|
try:
|
|
|
|
from org.python.core import PyStringMap
|
|
|
|
except ImportError:
|
|
|
|
PyStringMap = None
|
|
|
|
|
2001-09-22 03:22:34 +08:00
|
|
|
try:
|
|
|
|
UnicodeType
|
|
|
|
except NameError:
|
|
|
|
UnicodeType = None
|
|
|
|
|
2003-01-28 04:16:36 +08:00
|
|
|
# Pickle opcodes. See pickletools.py for extensive docs. The listing
|
|
|
|
# here is in kind-of alphabetical order of 1-character pickle code.
|
|
|
|
# pickletools groups them by purpose.
|
|
|
|
|
|
|
|
MARK = '(' # push special markobject on stack
|
|
|
|
STOP = '.' # every pickle ends with STOP
|
|
|
|
POP = '0' # discard topmost stack item
|
|
|
|
POP_MARK = '1' # discard stack top through topmost markobject
|
|
|
|
DUP = '2' # duplicate top stack item
|
|
|
|
FLOAT = 'F' # push float object; decimal string argument
|
|
|
|
INT = 'I' # push integer or bool; decimal string argument
|
|
|
|
BININT = 'J' # push four-byte signed int
|
|
|
|
BININT1 = 'K' # push 1-byte unsigned int
|
|
|
|
LONG = 'L' # push long; decimal string argument
|
|
|
|
BININT2 = 'M' # push 2-byte unsigned int
|
|
|
|
NONE = 'N' # push None
|
|
|
|
PERSID = 'P' # push persistent object; id is taken from string arg
|
|
|
|
BINPERSID = 'Q' # " " " ; " " " " stack
|
|
|
|
REDUCE = 'R' # apply callable to argtuple, both on stack
|
|
|
|
STRING = 'S' # push string; NL-terminated string argument
|
|
|
|
BINSTRING = 'T' # push string; counted binary string argument
|
|
|
|
SHORT_BINSTRING = 'U' # " " ; " " " " < 256 bytes
|
|
|
|
UNICODE = 'V' # push Unicode string; raw-unicode-escaped'd argument
|
|
|
|
BINUNICODE = 'X' # " " " ; counted UTF-8 string argument
|
|
|
|
APPEND = 'a' # append stack top to list below it
|
|
|
|
BUILD = 'b' # call __setstate__ or __dict__.update()
|
|
|
|
GLOBAL = 'c' # push self.find_class(modname, name); 2 string args
|
|
|
|
DICT = 'd' # build a dict from stack items
|
|
|
|
EMPTY_DICT = '}' # push empty dict
|
|
|
|
APPENDS = 'e' # extend list on stack by topmost stack slice
|
|
|
|
GET = 'g' # push item from memo on stack; index is string arg
|
|
|
|
BINGET = 'h' # " " " " " " ; " " 1-byte arg
|
|
|
|
INST = 'i' # build & push class instance
|
|
|
|
LONG_BINGET = 'j' # push item from memo on stack; index is 4-byte arg
|
|
|
|
LIST = 'l' # build list from topmost stack items
|
|
|
|
EMPTY_LIST = ']' # push empty list
|
|
|
|
OBJ = 'o' # build & push class instance
|
|
|
|
PUT = 'p' # store stack top in memo; index is string arg
|
|
|
|
BINPUT = 'q' # " " " " " ; " " 1-byte arg
|
|
|
|
LONG_BINPUT = 'r' # " " " " " ; " " 4-byte arg
|
|
|
|
SETITEM = 's' # add key+value pair to dict
|
|
|
|
TUPLE = 't' # build tuple from topmost stack items
|
|
|
|
EMPTY_TUPLE = ')' # push empty tuple
|
|
|
|
SETITEMS = 'u' # modify dict by adding topmost key+value pairs
|
|
|
|
BINFLOAT = 'G' # push float; arg is 8-byte float encoding
|
|
|
|
|
|
|
|
TRUE = 'I01\n' # not an opcode; see INT docs in pickletools.py
|
|
|
|
FALSE = 'I00\n' # not an opcode; see INT docs in pickletools.py
|
2002-04-04 06:41:51 +08:00
|
|
|
|
2003-01-28 08:22:12 +08:00
|
|
|
# Protocol 2 (not yet implemented).
|
|
|
|
|
|
|
|
PROTO = '\x80' # identify pickle protocol
|
|
|
|
NEWOBJ = '\x81' # build object by applying cls.__new__ to argtuple
|
|
|
|
EXT1 = '\x82' # push object from extension registry; 1-byte index
|
|
|
|
EXT2 = '\x83' # ditto, but 2-byte index
|
|
|
|
EXT4 = '\x84' # ditto, but 4-byte index
|
|
|
|
TUPLE1 = '\x85' # build 1-tuple from stack top
|
|
|
|
TUPLE2 = '\x86' # build 2-tuple from two topmost stack items
|
|
|
|
TUPLE3 = '\x87' # build 3-tuple from three topmost stack items
|
|
|
|
NEWTRUE = '\x88' # push True
|
|
|
|
NEWFALSE = '\x89' # push False
|
|
|
|
LONG1 = '\x8a' # push long from < 256 bytes
|
|
|
|
LONG4 = '\x8b' # push really big long
|
2003-01-28 05:44:25 +08:00
|
|
|
|
1995-01-10 08:31:14 +08:00
|
|
|
|
2001-02-18 11:10:09 +08:00
|
|
|
__all__.extend([x for x in dir() if re.match("[A-Z][A-Z0-9_]+$",x)])
|
2002-02-12 02:12:06 +08:00
|
|
|
del x
|
2001-02-18 11:10:09 +08:00
|
|
|
|
2002-08-14 15:46:28 +08:00
|
|
|
_quotes = ["'", '"']
|
|
|
|
|
1995-01-10 08:31:14 +08:00
|
|
|
class Pickler:
|
|
|
|
|
2003-01-28 06:47:53 +08:00
|
|
|
def __init__(self, file, proto=1):
|
2002-05-30 00:18:42 +08:00
|
|
|
"""This takes a file-like object for writing a pickle data stream.
|
|
|
|
|
2003-01-28 06:47:53 +08:00
|
|
|
The optional proto argument tells the pickler to use the given
|
|
|
|
protocol; supported protocols are 0, 1, 2. The default
|
|
|
|
protocol is 1 (in previous Python versions the default was 0).
|
|
|
|
|
|
|
|
Protocol 1 is more efficient than protocol 0; protocol 2 is
|
|
|
|
more efficient than protocol 1. Protocol 2 is not the default
|
|
|
|
because it is not supported by older Python versions.
|
|
|
|
|
|
|
|
XXX Protocol 2 is not yet implemented.
|
2002-05-30 00:18:42 +08:00
|
|
|
|
|
|
|
The file parameter must have a write() method that accepts a single
|
|
|
|
string argument. It can thus be an open file object, a StringIO
|
|
|
|
object, or any other custom object that meets this interface.
|
|
|
|
|
|
|
|
"""
|
1997-04-10 01:32:51 +08:00
|
|
|
self.write = file.write
|
|
|
|
self.memo = {}
|
2003-01-28 06:47:53 +08:00
|
|
|
self.proto = proto
|
|
|
|
self.bin = proto >= 1
|
1997-04-10 01:32:51 +08:00
|
|
|
|
2002-05-02 04:33:53 +08:00
|
|
|
def clear_memo(self):
|
2002-05-30 00:18:42 +08:00
|
|
|
"""Clears the pickler's "memo".
|
|
|
|
|
|
|
|
The memo is the data structure that remembers which objects the
|
2003-01-28 08:23:36 +08:00
|
|
|
pickler has already seen, so that shared or recursive objects are
|
|
|
|
pickled by reference and not by value. This method is useful when
|
|
|
|
re-using picklers.
|
2002-05-30 00:18:42 +08:00
|
|
|
|
|
|
|
"""
|
2002-05-02 04:33:53 +08:00
|
|
|
self.memo.clear()
|
|
|
|
|
1997-04-10 01:32:51 +08:00
|
|
|
def dump(self, object):
|
2002-05-30 00:18:42 +08:00
|
|
|
"""Write a pickled representation of object to the open file object.
|
|
|
|
|
|
|
|
Either the binary or ASCII format will be used, depending on the
|
|
|
|
value of the bin flag passed to the constructor.
|
|
|
|
|
|
|
|
"""
|
1997-04-10 01:32:51 +08:00
|
|
|
self.save(object)
|
|
|
|
self.write(STOP)
|
|
|
|
|
2003-01-25 03:29:52 +08:00
|
|
|
def memoize(self, obj):
|
|
|
|
"""Store an object in the memo."""
|
|
|
|
|
2003-01-28 05:22:10 +08:00
|
|
|
# The Pickler memo is a dictionary mapping object ids to 2-tuples
|
|
|
|
# that contain the Unpickler memo key and the object being memoized.
|
|
|
|
# The memo key is written to the pickle and will become
|
2003-01-25 03:29:52 +08:00
|
|
|
# the key in the Unpickler's memo. The object is stored in the
|
2003-01-28 05:22:10 +08:00
|
|
|
# Pickler memo so that transient objects are kept alive during
|
|
|
|
# pickling.
|
|
|
|
|
|
|
|
# The use of the Unpickler memo length as the memo key is just a
|
|
|
|
# convention. The only requirement is that the memo values be unique.
|
|
|
|
# But there appears no advantage to any other scheme, and this
|
2003-01-28 08:24:43 +08:00
|
|
|
# scheme allows the Unpickler memo to be implemented as a plain (but
|
2003-01-28 05:22:10 +08:00
|
|
|
# growable) array, indexed by memo key.
|
2003-01-25 03:29:52 +08:00
|
|
|
memo_len = len(self.memo)
|
|
|
|
self.write(self.put(memo_len))
|
2003-01-28 09:00:38 +08:00
|
|
|
self.memo[id(obj)] = memo_len, obj
|
2003-01-25 03:29:52 +08:00
|
|
|
|
2003-01-28 05:25:41 +08:00
|
|
|
# Return a PUT (BINPUT, LONG_BINPUT) opcode string, with argument i.
|
1997-04-10 01:32:51 +08:00
|
|
|
def put(self, i):
|
2001-04-10 10:48:53 +08:00
|
|
|
if self.bin:
|
1997-04-10 01:32:51 +08:00
|
|
|
s = mdumps(i)[1:]
|
2001-04-10 10:48:53 +08:00
|
|
|
if i < 256:
|
1997-04-10 01:32:51 +08:00
|
|
|
return BINPUT + s[0]
|
|
|
|
|
|
|
|
return LONG_BINPUT + s
|
|
|
|
|
|
|
|
return PUT + `i` + '\n'
|
|
|
|
|
2003-01-28 05:25:41 +08:00
|
|
|
# Return a GET (BINGET, LONG_BINGET) opcode string, with argument i.
|
1997-04-10 01:32:51 +08:00
|
|
|
def get(self, i):
|
2001-04-10 10:48:53 +08:00
|
|
|
if self.bin:
|
1997-04-10 01:32:51 +08:00
|
|
|
s = mdumps(i)[1:]
|
|
|
|
|
2001-04-10 10:48:53 +08:00
|
|
|
if i < 256:
|
1997-04-10 01:32:51 +08:00
|
|
|
return BINGET + s[0]
|
|
|
|
|
|
|
|
return LONG_BINGET + s
|
|
|
|
|
|
|
|
return GET + `i` + '\n'
|
2001-01-15 08:50:52 +08:00
|
|
|
|
2002-11-14 06:01:27 +08:00
|
|
|
def save(self, object):
|
1997-04-10 01:32:51 +08:00
|
|
|
memo = self.memo
|
|
|
|
|
2002-11-14 06:01:27 +08:00
|
|
|
pid = self.persistent_id(object)
|
|
|
|
if pid is not None:
|
|
|
|
self.save_pers(pid)
|
|
|
|
return
|
1997-04-10 01:32:51 +08:00
|
|
|
|
|
|
|
d = id(object)
|
2001-01-15 08:50:52 +08:00
|
|
|
|
1997-04-10 01:32:51 +08:00
|
|
|
t = type(object)
|
|
|
|
|
2003-01-28 08:43:26 +08:00
|
|
|
# XXX Why are tuples a special case here?
|
2001-04-10 10:48:53 +08:00
|
|
|
if (t is TupleType) and (len(object) == 0):
|
|
|
|
if self.bin:
|
1997-04-10 01:32:51 +08:00
|
|
|
self.save_empty_tuple(object)
|
|
|
|
else:
|
|
|
|
self.save_tuple(object)
|
|
|
|
return
|
|
|
|
|
2002-06-01 22:18:47 +08:00
|
|
|
if d in memo:
|
1997-04-10 01:32:51 +08:00
|
|
|
self.write(self.get(memo[d][0]))
|
|
|
|
return
|
|
|
|
|
|
|
|
try:
|
|
|
|
f = self.dispatch[t]
|
|
|
|
except KeyError:
|
2003-01-28 08:48:09 +08:00
|
|
|
pass
|
|
|
|
else:
|
|
|
|
f(self, object)
|
|
|
|
return
|
|
|
|
|
|
|
|
# The dispatch table doesn't know about type t.
|
|
|
|
try:
|
|
|
|
issc = issubclass(t, TypeType)
|
|
|
|
except TypeError: # t is not a class
|
|
|
|
issc = 0
|
|
|
|
if issc:
|
|
|
|
self.save_global(object)
|
|
|
|
return
|
2001-12-20 00:55:02 +08:00
|
|
|
|
2003-01-28 08:48:09 +08:00
|
|
|
try:
|
|
|
|
reduce = dispatch_table[t]
|
|
|
|
except KeyError:
|
1997-04-10 01:32:51 +08:00
|
|
|
try:
|
2003-01-28 08:48:09 +08:00
|
|
|
reduce = object.__reduce__
|
|
|
|
except AttributeError:
|
|
|
|
raise PicklingError, \
|
|
|
|
"can't pickle %s object: %s" % (`t.__name__`,
|
|
|
|
`object`)
|
1997-04-10 01:32:51 +08:00
|
|
|
else:
|
2003-01-28 08:48:09 +08:00
|
|
|
tup = reduce()
|
|
|
|
else:
|
|
|
|
tup = reduce(object)
|
1997-12-11 07:40:18 +08:00
|
|
|
|
2003-01-28 08:48:09 +08:00
|
|
|
if type(tup) is StringType:
|
|
|
|
self.save_global(object, tup)
|
|
|
|
return
|
1997-04-10 01:32:51 +08:00
|
|
|
|
2003-01-28 08:48:09 +08:00
|
|
|
if type(tup) is not TupleType:
|
|
|
|
raise PicklingError, "Value returned by %s must be a " \
|
|
|
|
"tuple" % reduce
|
2001-01-15 08:50:52 +08:00
|
|
|
|
2003-01-28 08:48:09 +08:00
|
|
|
l = len(tup)
|
1997-04-10 01:32:51 +08:00
|
|
|
|
2003-01-28 08:48:09 +08:00
|
|
|
if (l != 2) and (l != 3):
|
|
|
|
raise PicklingError, "tuple returned by %s must contain " \
|
|
|
|
"only two or three elements" % reduce
|
2001-01-15 08:50:52 +08:00
|
|
|
|
2003-01-28 08:48:09 +08:00
|
|
|
callable = tup[0]
|
|
|
|
arg_tup = tup[1]
|
1997-04-10 01:32:51 +08:00
|
|
|
|
2003-01-28 08:48:09 +08:00
|
|
|
if l > 2:
|
|
|
|
state = tup[2]
|
|
|
|
else:
|
|
|
|
state = None
|
1997-04-10 01:32:51 +08:00
|
|
|
|
2003-01-28 08:48:09 +08:00
|
|
|
if type(arg_tup) is not TupleType and arg_tup is not None:
|
|
|
|
raise PicklingError, "Second element of tuple returned " \
|
|
|
|
"by %s must be a tuple" % reduce
|
1997-04-10 01:32:51 +08:00
|
|
|
|
2003-01-28 08:48:09 +08:00
|
|
|
self.save_reduce(callable, arg_tup, state)
|
2003-01-28 09:00:38 +08:00
|
|
|
self.memoize(object)
|
1997-04-10 01:32:51 +08:00
|
|
|
|
|
|
|
def persistent_id(self, object):
|
|
|
|
return None
|
|
|
|
|
|
|
|
def save_pers(self, pid):
|
2003-01-28 09:03:10 +08:00
|
|
|
if self.bin:
|
2002-11-14 06:01:27 +08:00
|
|
|
self.save(pid)
|
1997-04-10 01:32:51 +08:00
|
|
|
self.write(BINPERSID)
|
2003-01-28 09:03:10 +08:00
|
|
|
else:
|
|
|
|
self.write(PERSID + str(pid) + '\n')
|
1997-04-10 01:32:51 +08:00
|
|
|
|
2003-01-25 03:29:52 +08:00
|
|
|
def save_reduce(self, acallable, arg_tup, state = None):
|
1997-04-10 01:32:51 +08:00
|
|
|
write = self.write
|
|
|
|
save = self.save
|
|
|
|
|
2003-01-25 03:29:52 +08:00
|
|
|
if not callable(acallable):
|
|
|
|
raise PicklingError("__reduce__() must return callable as "
|
|
|
|
"first argument, not %s" % `acallable`)
|
2003-01-28 04:16:36 +08:00
|
|
|
|
2003-01-25 03:29:52 +08:00
|
|
|
save(acallable)
|
1997-04-10 01:32:51 +08:00
|
|
|
save(arg_tup)
|
|
|
|
write(REDUCE)
|
2001-01-15 08:50:52 +08:00
|
|
|
|
2001-04-10 10:48:53 +08:00
|
|
|
if state is not None:
|
1997-04-10 01:32:51 +08:00
|
|
|
save(state)
|
|
|
|
write(BUILD)
|
|
|
|
|
|
|
|
dispatch = {}
|
|
|
|
|
|
|
|
def save_none(self, object):
|
|
|
|
self.write(NONE)
|
|
|
|
dispatch[NoneType] = save_none
|
|
|
|
|
2002-04-04 06:41:51 +08:00
|
|
|
def save_bool(self, object):
|
2003-01-28 08:26:14 +08:00
|
|
|
self.write(object and TRUE or FALSE)
|
2002-04-04 06:41:51 +08:00
|
|
|
dispatch[bool] = save_bool
|
|
|
|
|
1997-04-10 01:32:51 +08:00
|
|
|
def save_int(self, object):
|
2001-04-10 10:48:53 +08:00
|
|
|
if self.bin:
|
2001-04-10 13:02:52 +08:00
|
|
|
# If the int is small enough to fit in a signed 4-byte 2's-comp
|
|
|
|
# format, we can store it more efficiently than the general
|
|
|
|
# case.
|
|
|
|
high_bits = object >> 31 # note that Python shift sign-extends
|
|
|
|
if high_bits == 0 or high_bits == -1:
|
|
|
|
# All high bits are copies of bit 2**31, so the value
|
|
|
|
# fits in a 4-byte signed int.
|
|
|
|
i = mdumps(object)[1:]
|
|
|
|
assert len(i) == 4
|
|
|
|
if i[-2:] == '\000\000': # fits in 2-byte unsigned int
|
|
|
|
if i[-3] == '\000': # fits in 1-byte unsigned int
|
|
|
|
self.write(BININT1 + i[0])
|
|
|
|
else:
|
|
|
|
self.write(BININT2 + i[:2])
|
|
|
|
else:
|
|
|
|
self.write(BININT + i)
|
1997-04-10 01:32:51 +08:00
|
|
|
return
|
2001-04-10 13:02:52 +08:00
|
|
|
# Text pickle, or int too big to fit in signed 4-byte format.
|
|
|
|
self.write(INT + `object` + '\n')
|
1997-04-10 01:32:51 +08:00
|
|
|
dispatch[IntType] = save_int
|
|
|
|
|
|
|
|
def save_long(self, object):
|
|
|
|
self.write(LONG + `object` + '\n')
|
|
|
|
dispatch[LongType] = save_long
|
|
|
|
|
1998-10-23 04:15:36 +08:00
|
|
|
def save_float(self, object, pack=struct.pack):
|
|
|
|
if self.bin:
|
|
|
|
self.write(BINFLOAT + pack('>d', object))
|
|
|
|
else:
|
|
|
|
self.write(FLOAT + `object` + '\n')
|
1997-04-10 01:32:51 +08:00
|
|
|
dispatch[FloatType] = save_float
|
|
|
|
|
|
|
|
def save_string(self, object):
|
2001-04-10 10:48:53 +08:00
|
|
|
if self.bin:
|
2003-01-28 05:15:36 +08:00
|
|
|
n = len(object)
|
|
|
|
if n < 256:
|
|
|
|
self.write(SHORT_BINSTRING + chr(n) + object)
|
1997-04-10 01:32:51 +08:00
|
|
|
else:
|
2003-01-28 05:15:36 +08:00
|
|
|
self.write(BINSTRING + mdumps(n)[1:] + object)
|
1997-04-10 01:32:51 +08:00
|
|
|
else:
|
|
|
|
self.write(STRING + `object` + '\n')
|
2003-01-25 03:29:52 +08:00
|
|
|
self.memoize(object)
|
1997-04-10 01:32:51 +08:00
|
|
|
dispatch[StringType] = save_string
|
|
|
|
|
2000-03-11 07:20:09 +08:00
|
|
|
def save_unicode(self, object):
|
2001-04-10 10:48:53 +08:00
|
|
|
if self.bin:
|
2000-03-11 07:20:09 +08:00
|
|
|
encoding = object.encode('utf-8')
|
2003-01-28 05:15:36 +08:00
|
|
|
n = len(encoding)
|
|
|
|
s = mdumps(n)[1:]
|
2000-03-11 07:20:09 +08:00
|
|
|
self.write(BINUNICODE + s + encoding)
|
|
|
|
else:
|
2001-09-22 03:22:34 +08:00
|
|
|
object = object.replace("\\", "\\u005c")
|
|
|
|
object = object.replace("\n", "\\u000a")
|
2000-03-11 07:20:09 +08:00
|
|
|
self.write(UNICODE + object.encode('raw-unicode-escape') + '\n')
|
2003-01-25 03:29:52 +08:00
|
|
|
self.memoize(object)
|
2000-03-11 07:20:09 +08:00
|
|
|
dispatch[UnicodeType] = save_unicode
|
|
|
|
|
2001-01-22 22:53:29 +08:00
|
|
|
if StringType == UnicodeType:
|
|
|
|
# This is true for Jython
|
|
|
|
def save_string(self, object):
|
|
|
|
unicode = object.isunicode()
|
|
|
|
|
2001-04-10 10:48:53 +08:00
|
|
|
if self.bin:
|
2001-01-22 22:53:29 +08:00
|
|
|
if unicode:
|
|
|
|
object = object.encode("utf-8")
|
|
|
|
l = len(object)
|
|
|
|
s = mdumps(l)[1:]
|
2001-04-10 10:48:53 +08:00
|
|
|
if l < 256 and not unicode:
|
2001-01-22 22:53:29 +08:00
|
|
|
self.write(SHORT_BINSTRING + s[0] + object)
|
|
|
|
else:
|
|
|
|
if unicode:
|
|
|
|
self.write(BINUNICODE + s + object)
|
|
|
|
else:
|
|
|
|
self.write(BINSTRING + s + object)
|
|
|
|
else:
|
2001-02-10 04:06:00 +08:00
|
|
|
if unicode:
|
2001-09-22 03:22:34 +08:00
|
|
|
object = object.replace("\\", "\\u005c")
|
|
|
|
object = object.replace("\n", "\\u000a")
|
2001-01-22 22:53:29 +08:00
|
|
|
object = object.encode('raw-unicode-escape')
|
|
|
|
self.write(UNICODE + object + '\n')
|
|
|
|
else:
|
|
|
|
self.write(STRING + `object` + '\n')
|
2003-01-25 03:29:52 +08:00
|
|
|
self.memoize(object)
|
2001-01-22 22:53:29 +08:00
|
|
|
dispatch[StringType] = save_string
|
2001-02-10 04:06:00 +08:00
|
|
|
|
1997-04-10 01:32:51 +08:00
|
|
|
def save_tuple(self, object):
|
|
|
|
write = self.write
|
|
|
|
save = self.save
|
|
|
|
memo = self.memo
|
|
|
|
|
|
|
|
d = id(object)
|
|
|
|
|
|
|
|
write(MARK)
|
|
|
|
|
|
|
|
for element in object:
|
|
|
|
save(element)
|
|
|
|
|
2002-06-01 22:18:47 +08:00
|
|
|
if len(object) and d in memo:
|
2001-04-10 10:48:53 +08:00
|
|
|
if self.bin:
|
1997-04-10 01:32:51 +08:00
|
|
|
write(POP_MARK + self.get(memo[d][0]))
|
|
|
|
return
|
2001-01-15 08:50:52 +08:00
|
|
|
|
1998-04-01 00:30:28 +08:00
|
|
|
write(POP * (len(object) + 1) + self.get(memo[d][0]))
|
1997-04-10 01:32:51 +08:00
|
|
|
return
|
|
|
|
|
2003-01-28 09:00:38 +08:00
|
|
|
self.write(TUPLE)
|
|
|
|
self.memoize(object)
|
2003-01-25 03:29:52 +08:00
|
|
|
|
1997-04-10 01:32:51 +08:00
|
|
|
dispatch[TupleType] = save_tuple
|
|
|
|
|
|
|
|
def save_empty_tuple(self, object):
|
|
|
|
self.write(EMPTY_TUPLE)
|
|
|
|
|
|
|
|
def save_list(self, object):
|
|
|
|
write = self.write
|
|
|
|
save = self.save
|
|
|
|
|
2001-04-10 10:48:53 +08:00
|
|
|
if self.bin:
|
1997-04-10 01:32:51 +08:00
|
|
|
write(EMPTY_LIST)
|
2003-01-28 09:15:46 +08:00
|
|
|
self.memoize(object)
|
|
|
|
n = len(object)
|
|
|
|
if n > 1:
|
|
|
|
write(MARK)
|
|
|
|
for element in object:
|
|
|
|
save(element)
|
|
|
|
write(APPENDS)
|
|
|
|
elif n:
|
|
|
|
assert n == 1
|
|
|
|
save(object[0])
|
|
|
|
write(APPEND)
|
|
|
|
# else the list is empty, and we're already done
|
2001-01-15 08:50:52 +08:00
|
|
|
|
2003-01-28 09:15:46 +08:00
|
|
|
else: # proto 0 -- can't use EMPTY_LIST or APPENDS
|
|
|
|
write(MARK + LIST)
|
|
|
|
self.memoize(object)
|
|
|
|
for element in object:
|
|
|
|
save(element)
|
1997-04-10 01:32:51 +08:00
|
|
|
write(APPEND)
|
|
|
|
|
|
|
|
dispatch[ListType] = save_list
|
|
|
|
|
|
|
|
def save_dict(self, object):
|
|
|
|
write = self.write
|
|
|
|
save = self.save
|
2003-01-28 09:34:43 +08:00
|
|
|
items = object.iteritems()
|
1997-04-10 01:32:51 +08:00
|
|
|
|
2001-04-10 10:48:53 +08:00
|
|
|
if self.bin:
|
1997-04-10 01:32:51 +08:00
|
|
|
write(EMPTY_DICT)
|
2003-01-28 09:34:43 +08:00
|
|
|
self.memoize(object)
|
|
|
|
if len(object) > 1:
|
|
|
|
write(MARK)
|
|
|
|
for key, value in items:
|
|
|
|
save(key)
|
|
|
|
save(value)
|
|
|
|
write(SETITEMS)
|
|
|
|
return
|
1997-04-10 01:32:51 +08:00
|
|
|
|
2003-01-28 09:34:43 +08:00
|
|
|
else: # proto 0 -- can't use EMPTY_DICT or SETITEMS
|
|
|
|
write(MARK + DICT)
|
|
|
|
self.memoize(object)
|
1997-04-10 01:32:51 +08:00
|
|
|
|
2003-01-28 09:34:43 +08:00
|
|
|
# proto 0 or len(object) < 2
|
1997-04-10 01:32:51 +08:00
|
|
|
for key, value in items:
|
|
|
|
save(key)
|
|
|
|
save(value)
|
2003-01-28 09:34:43 +08:00
|
|
|
write(SETITEM)
|
1997-04-10 01:32:51 +08:00
|
|
|
|
|
|
|
dispatch[DictionaryType] = save_dict
|
1998-05-28 06:38:22 +08:00
|
|
|
if not PyStringMap is None:
|
|
|
|
dispatch[PyStringMap] = save_dict
|
1997-04-10 01:32:51 +08:00
|
|
|
|
|
|
|
def save_inst(self, object):
|
|
|
|
d = id(object)
|
|
|
|
cls = object.__class__
|
|
|
|
|
|
|
|
memo = self.memo
|
|
|
|
write = self.write
|
|
|
|
save = self.save
|
|
|
|
|
|
|
|
if hasattr(object, '__getinitargs__'):
|
|
|
|
args = object.__getinitargs__()
|
|
|
|
len(args) # XXX Assert it's a sequence
|
1998-03-27 05:13:24 +08:00
|
|
|
_keep_alive(args, memo)
|
1997-04-10 01:32:51 +08:00
|
|
|
else:
|
|
|
|
args = ()
|
|
|
|
|
|
|
|
write(MARK)
|
|
|
|
|
2001-04-10 10:48:53 +08:00
|
|
|
if self.bin:
|
1997-04-10 01:32:51 +08:00
|
|
|
save(cls)
|
|
|
|
|
|
|
|
for arg in args:
|
|
|
|
save(arg)
|
|
|
|
|
2003-01-25 03:29:52 +08:00
|
|
|
# This method does not use memoize() so that it can handle
|
|
|
|
# the special case for non-binary mode.
|
2003-01-28 09:00:38 +08:00
|
|
|
# XXX What did that comment mean? That is, what "special case for
|
2003-01-28 09:41:51 +08:00
|
|
|
# XXX non-binary mode"? It sure *looks* like nothing special is
|
2003-01-28 09:00:38 +08:00
|
|
|
# XXX happening in the INST case.
|
1997-04-10 01:32:51 +08:00
|
|
|
memo_len = len(memo)
|
2001-04-10 10:48:53 +08:00
|
|
|
if self.bin:
|
1997-04-10 01:32:51 +08:00
|
|
|
write(OBJ + self.put(memo_len))
|
|
|
|
else:
|
1997-09-13 04:07:24 +08:00
|
|
|
write(INST + cls.__module__ + '\n' + cls.__name__ + '\n' +
|
1997-04-10 01:32:51 +08:00
|
|
|
self.put(memo_len))
|
|
|
|
|
|
|
|
memo[d] = (memo_len, object)
|
|
|
|
|
|
|
|
try:
|
|
|
|
getstate = object.__getstate__
|
|
|
|
except AttributeError:
|
|
|
|
stuff = object.__dict__
|
|
|
|
else:
|
|
|
|
stuff = getstate()
|
1998-03-27 05:13:24 +08:00
|
|
|
_keep_alive(stuff, memo)
|
1997-04-10 01:32:51 +08:00
|
|
|
save(stuff)
|
|
|
|
write(BUILD)
|
|
|
|
dispatch[InstanceType] = save_inst
|
|
|
|
|
|
|
|
def save_global(self, object, name = None):
|
|
|
|
write = self.write
|
|
|
|
memo = self.memo
|
|
|
|
|
2001-04-10 10:48:53 +08:00
|
|
|
if name is None:
|
1997-04-10 01:32:51 +08:00
|
|
|
name = object.__name__
|
|
|
|
|
1998-03-27 05:13:24 +08:00
|
|
|
try:
|
|
|
|
module = object.__module__
|
|
|
|
except AttributeError:
|
|
|
|
module = whichmodule(object, name)
|
1997-04-10 01:32:51 +08:00
|
|
|
|
2001-08-18 02:49:52 +08:00
|
|
|
try:
|
|
|
|
__import__(module)
|
|
|
|
mod = sys.modules[module]
|
|
|
|
klass = getattr(mod, name)
|
|
|
|
except (ImportError, KeyError, AttributeError):
|
|
|
|
raise PicklingError(
|
|
|
|
"Can't pickle %r: it's not found as %s.%s" %
|
|
|
|
(object, module, name))
|
|
|
|
else:
|
|
|
|
if klass is not object:
|
|
|
|
raise PicklingError(
|
|
|
|
"Can't pickle %r: it's not the same object as %s.%s" %
|
|
|
|
(object, module, name))
|
|
|
|
|
2003-01-28 09:00:38 +08:00
|
|
|
write(GLOBAL + module + '\n' + name + '\n')
|
|
|
|
self.memoize(object)
|
1997-04-10 01:32:51 +08:00
|
|
|
dispatch[ClassType] = save_global
|
|
|
|
dispatch[FunctionType] = save_global
|
|
|
|
dispatch[BuiltinFunctionType] = save_global
|
2001-08-02 12:15:00 +08:00
|
|
|
dispatch[TypeType] = save_global
|
1995-03-14 23:09:05 +08:00
|
|
|
|
1995-01-10 08:31:14 +08:00
|
|
|
|
1997-09-03 08:23:54 +08:00
|
|
|
def _keep_alive(x, memo):
|
|
|
|
"""Keeps a reference to the object x in the memo.
|
|
|
|
|
|
|
|
Because we remember objects by their id, we have
|
|
|
|
to assure that possibly temporary objects are kept
|
|
|
|
alive by referencing them.
|
|
|
|
We store a reference at the id of the memo, which should
|
|
|
|
normally not be used unless someone tries to deepcopy
|
|
|
|
the memo itself...
|
|
|
|
"""
|
|
|
|
try:
|
1998-03-27 05:13:24 +08:00
|
|
|
memo[id(memo)].append(x)
|
1997-09-03 08:23:54 +08:00
|
|
|
except KeyError:
|
1998-03-27 05:13:24 +08:00
|
|
|
# aha, this is the first one :-)
|
|
|
|
memo[id(memo)]=[x]
|
1997-09-03 08:23:54 +08:00
|
|
|
|
|
|
|
|
2002-09-20 07:00:12 +08:00
|
|
|
classmap = {} # called classmap for backwards compatibility
|
1995-01-10 08:31:14 +08:00
|
|
|
|
2002-09-20 07:00:12 +08:00
|
|
|
def whichmodule(func, funcname):
|
|
|
|
"""Figure out the module in which a function occurs.
|
1995-01-10 08:31:14 +08:00
|
|
|
|
1997-04-10 01:32:51 +08:00
|
|
|
Search sys.modules for the module.
|
|
|
|
Cache in classmap.
|
|
|
|
Return a module name.
|
2002-09-20 07:00:12 +08:00
|
|
|
If the function cannot be found, return __main__.
|
1997-04-10 01:32:51 +08:00
|
|
|
"""
|
2002-09-20 07:00:12 +08:00
|
|
|
if func in classmap:
|
|
|
|
return classmap[func]
|
1995-01-10 08:31:14 +08:00
|
|
|
|
1997-04-10 01:32:51 +08:00
|
|
|
for name, module in sys.modules.items():
|
2002-09-20 07:00:12 +08:00
|
|
|
if module is None:
|
2002-09-20 06:57:26 +08:00
|
|
|
continue # skip dummy package entries
|
1998-03-27 05:13:24 +08:00
|
|
|
if name != '__main__' and \
|
2002-09-20 07:00:12 +08:00
|
|
|
hasattr(module, funcname) and \
|
|
|
|
getattr(module, funcname) is func:
|
1997-04-10 01:32:51 +08:00
|
|
|
break
|
|
|
|
else:
|
|
|
|
name = '__main__'
|
2002-09-20 07:00:12 +08:00
|
|
|
classmap[func] = name
|
1997-04-10 01:32:51 +08:00
|
|
|
return name
|
1995-01-10 08:31:14 +08:00
|
|
|
|
|
|
|
|
1997-04-10 01:32:51 +08:00
|
|
|
class Unpickler:
|
1995-01-10 08:31:14 +08:00
|
|
|
|
1997-04-10 01:32:51 +08:00
|
|
|
def __init__(self, file):
|
2002-05-30 00:18:42 +08:00
|
|
|
"""This takes a file-like object for reading a pickle data stream.
|
|
|
|
|
|
|
|
This class automatically determines whether the data stream was
|
|
|
|
written in binary mode or not, so it does not need a flag as in
|
|
|
|
the Pickler class factory.
|
|
|
|
|
|
|
|
The file-like object must have two methods, a read() method that
|
|
|
|
takes an integer argument, and a readline() method that requires no
|
|
|
|
arguments. Both methods should return a string. Thus file-like
|
|
|
|
object can be a file object opened for reading, a StringIO object,
|
|
|
|
or any other custom object that meets this interface.
|
|
|
|
|
|
|
|
"""
|
1997-04-10 01:32:51 +08:00
|
|
|
self.readline = file.readline
|
|
|
|
self.read = file.read
|
|
|
|
self.memo = {}
|
|
|
|
|
|
|
|
def load(self):
|
2002-05-30 00:18:42 +08:00
|
|
|
"""Read a pickled object representation from the open file object.
|
|
|
|
|
|
|
|
Return the reconstituted object hierarchy specified in the file
|
|
|
|
object.
|
|
|
|
|
|
|
|
"""
|
2001-11-10 00:15:04 +08:00
|
|
|
self.mark = object() # any new unique object
|
1997-04-10 01:32:51 +08:00
|
|
|
self.stack = []
|
|
|
|
self.append = self.stack.append
|
|
|
|
read = self.read
|
|
|
|
dispatch = self.dispatch
|
|
|
|
try:
|
|
|
|
while 1:
|
|
|
|
key = read(1)
|
|
|
|
dispatch[key](self)
|
2000-12-14 02:11:56 +08:00
|
|
|
except _Stop, stopinst:
|
|
|
|
return stopinst.value
|
1997-04-10 01:32:51 +08:00
|
|
|
|
2003-01-28 09:41:51 +08:00
|
|
|
# Return largest index k such that self.stack[k] is self.mark.
|
|
|
|
# If the stack doesn't contain a mark, eventually raises IndexError.
|
|
|
|
# This could be sped by maintaining another stack, of indices at which
|
|
|
|
# the mark appears. For that matter, the latter stack would suffice,
|
|
|
|
# and we wouldn't need to push mark objects on self.stack at all.
|
|
|
|
# Doing so is probably a good thing, though, since if the pickle is
|
|
|
|
# corrupt (or hostile) we may get a clue from finding self.mark embedded
|
|
|
|
# in unpickled objects.
|
1997-04-10 01:32:51 +08:00
|
|
|
def marker(self):
|
|
|
|
stack = self.stack
|
|
|
|
mark = self.mark
|
|
|
|
k = len(stack)-1
|
|
|
|
while stack[k] is not mark: k = k-1
|
|
|
|
return k
|
|
|
|
|
|
|
|
dispatch = {}
|
|
|
|
|
|
|
|
def load_eof(self):
|
|
|
|
raise EOFError
|
|
|
|
dispatch[''] = load_eof
|
|
|
|
|
|
|
|
def load_persid(self):
|
|
|
|
pid = self.readline()[:-1]
|
|
|
|
self.append(self.persistent_load(pid))
|
|
|
|
dispatch[PERSID] = load_persid
|
|
|
|
|
|
|
|
def load_binpersid(self):
|
2002-06-30 11:39:14 +08:00
|
|
|
pid = self.stack.pop()
|
1997-04-10 01:32:51 +08:00
|
|
|
self.append(self.persistent_load(pid))
|
|
|
|
dispatch[BINPERSID] = load_binpersid
|
|
|
|
|
|
|
|
def load_none(self):
|
|
|
|
self.append(None)
|
|
|
|
dispatch[NONE] = load_none
|
|
|
|
|
|
|
|
def load_int(self):
|
2001-08-29 06:21:18 +08:00
|
|
|
data = self.readline()
|
2002-04-06 03:30:08 +08:00
|
|
|
if data == FALSE[1:]:
|
|
|
|
val = False
|
|
|
|
elif data == TRUE[1:]:
|
|
|
|
val = True
|
|
|
|
else:
|
|
|
|
try:
|
|
|
|
val = int(data)
|
|
|
|
except ValueError:
|
|
|
|
val = long(data)
|
|
|
|
self.append(val)
|
1997-04-10 01:32:51 +08:00
|
|
|
dispatch[INT] = load_int
|
|
|
|
|
|
|
|
def load_binint(self):
|
|
|
|
self.append(mloads('i' + self.read(4)))
|
|
|
|
dispatch[BININT] = load_binint
|
|
|
|
|
|
|
|
def load_binint1(self):
|
2003-01-28 05:15:36 +08:00
|
|
|
self.append(ord(self.read(1)))
|
1997-04-10 01:32:51 +08:00
|
|
|
dispatch[BININT1] = load_binint1
|
|
|
|
|
|
|
|
def load_binint2(self):
|
|
|
|
self.append(mloads('i' + self.read(2) + '\000\000'))
|
|
|
|
dispatch[BININT2] = load_binint2
|
2001-01-15 08:50:52 +08:00
|
|
|
|
1997-04-10 01:32:51 +08:00
|
|
|
def load_long(self):
|
2000-12-14 02:11:56 +08:00
|
|
|
self.append(long(self.readline()[:-1], 0))
|
1997-04-10 01:32:51 +08:00
|
|
|
dispatch[LONG] = load_long
|
|
|
|
|
|
|
|
def load_float(self):
|
2000-12-14 02:11:56 +08:00
|
|
|
self.append(float(self.readline()[:-1]))
|
1997-04-10 01:32:51 +08:00
|
|
|
dispatch[FLOAT] = load_float
|
|
|
|
|
1998-10-23 04:15:36 +08:00
|
|
|
def load_binfloat(self, unpack=struct.unpack):
|
|
|
|
self.append(unpack('>d', self.read(8))[0])
|
|
|
|
dispatch[BINFLOAT] = load_binfloat
|
|
|
|
|
1997-04-10 01:32:51 +08:00
|
|
|
def load_string(self):
|
2000-09-15 23:14:51 +08:00
|
|
|
rep = self.readline()[:-1]
|
2002-08-14 15:46:28 +08:00
|
|
|
for q in _quotes:
|
|
|
|
if rep.startswith(q):
|
|
|
|
if not rep.endswith(q):
|
|
|
|
raise ValueError, "insecure string pickle"
|
|
|
|
rep = rep[len(q):-len(q)]
|
|
|
|
break
|
|
|
|
else:
|
2000-09-15 23:14:51 +08:00
|
|
|
raise ValueError, "insecure string pickle"
|
2002-08-14 15:46:28 +08:00
|
|
|
self.append(rep.decode("string-escape"))
|
1997-04-10 01:32:51 +08:00
|
|
|
dispatch[STRING] = load_string
|
|
|
|
|
2000-09-15 23:14:51 +08:00
|
|
|
def _is_string_secure(self, s):
|
|
|
|
"""Return true if s contains a string that is safe to eval
|
|
|
|
|
|
|
|
The definition of secure string is based on the implementation
|
|
|
|
in cPickle. s is secure as long as it only contains a quoted
|
|
|
|
string and optional trailing whitespace.
|
|
|
|
"""
|
|
|
|
q = s[0]
|
|
|
|
if q not in ("'", '"'):
|
|
|
|
return 0
|
|
|
|
# find the closing quote
|
|
|
|
offset = 1
|
|
|
|
i = None
|
|
|
|
while 1:
|
|
|
|
try:
|
|
|
|
i = s.index(q, offset)
|
|
|
|
except ValueError:
|
|
|
|
# if there is an error the first time, there is no
|
|
|
|
# close quote
|
|
|
|
if offset == 1:
|
|
|
|
return 0
|
|
|
|
if s[i-1] != '\\':
|
|
|
|
break
|
|
|
|
# check to see if this one is escaped
|
|
|
|
nslash = 0
|
|
|
|
j = i - 1
|
|
|
|
while j >= offset and s[j] == '\\':
|
|
|
|
j = j - 1
|
|
|
|
nslash = nslash + 1
|
|
|
|
if nslash % 2 == 0:
|
|
|
|
break
|
|
|
|
offset = i + 1
|
|
|
|
for c in s[i+1:]:
|
|
|
|
if ord(c) > 32:
|
|
|
|
return 0
|
|
|
|
return 1
|
|
|
|
|
1997-04-10 01:32:51 +08:00
|
|
|
def load_binstring(self):
|
|
|
|
len = mloads('i' + self.read(4))
|
|
|
|
self.append(self.read(len))
|
|
|
|
dispatch[BINSTRING] = load_binstring
|
|
|
|
|
2000-03-11 07:20:09 +08:00
|
|
|
def load_unicode(self):
|
|
|
|
self.append(unicode(self.readline()[:-1],'raw-unicode-escape'))
|
|
|
|
dispatch[UNICODE] = load_unicode
|
|
|
|
|
|
|
|
def load_binunicode(self):
|
|
|
|
len = mloads('i' + self.read(4))
|
|
|
|
self.append(unicode(self.read(len),'utf-8'))
|
|
|
|
dispatch[BINUNICODE] = load_binunicode
|
|
|
|
|
1997-04-10 01:32:51 +08:00
|
|
|
def load_short_binstring(self):
|
2003-01-28 05:15:36 +08:00
|
|
|
len = ord(self.read(1))
|
1997-04-10 01:32:51 +08:00
|
|
|
self.append(self.read(len))
|
|
|
|
dispatch[SHORT_BINSTRING] = load_short_binstring
|
|
|
|
|
|
|
|
def load_tuple(self):
|
|
|
|
k = self.marker()
|
|
|
|
self.stack[k:] = [tuple(self.stack[k+1:])]
|
|
|
|
dispatch[TUPLE] = load_tuple
|
|
|
|
|
|
|
|
def load_empty_tuple(self):
|
|
|
|
self.stack.append(())
|
|
|
|
dispatch[EMPTY_TUPLE] = load_empty_tuple
|
|
|
|
|
|
|
|
def load_empty_list(self):
|
|
|
|
self.stack.append([])
|
|
|
|
dispatch[EMPTY_LIST] = load_empty_list
|
|
|
|
|
|
|
|
def load_empty_dictionary(self):
|
|
|
|
self.stack.append({})
|
|
|
|
dispatch[EMPTY_DICT] = load_empty_dictionary
|
|
|
|
|
|
|
|
def load_list(self):
|
|
|
|
k = self.marker()
|
|
|
|
self.stack[k:] = [self.stack[k+1:]]
|
|
|
|
dispatch[LIST] = load_list
|
|
|
|
|
|
|
|
def load_dict(self):
|
|
|
|
k = self.marker()
|
|
|
|
d = {}
|
|
|
|
items = self.stack[k+1:]
|
|
|
|
for i in range(0, len(items), 2):
|
|
|
|
key = items[i]
|
|
|
|
value = items[i+1]
|
|
|
|
d[key] = value
|
|
|
|
self.stack[k:] = [d]
|
|
|
|
dispatch[DICT] = load_dict
|
|
|
|
|
|
|
|
def load_inst(self):
|
|
|
|
k = self.marker()
|
|
|
|
args = tuple(self.stack[k+1:])
|
|
|
|
del self.stack[k:]
|
|
|
|
module = self.readline()[:-1]
|
|
|
|
name = self.readline()[:-1]
|
|
|
|
klass = self.find_class(module, name)
|
1998-03-27 05:13:24 +08:00
|
|
|
instantiated = 0
|
|
|
|
if (not args and type(klass) is ClassType and
|
|
|
|
not hasattr(klass, "__getinitargs__")):
|
|
|
|
try:
|
|
|
|
value = _EmptyClass()
|
|
|
|
value.__class__ = klass
|
1998-04-14 02:08:45 +08:00
|
|
|
instantiated = 1
|
1998-03-27 05:13:24 +08:00
|
|
|
except RuntimeError:
|
|
|
|
# In restricted execution, assignment to inst.__class__ is
|
|
|
|
# prohibited
|
|
|
|
pass
|
|
|
|
if not instantiated:
|
1998-09-16 04:25:57 +08:00
|
|
|
try:
|
2001-11-16 07:42:58 +08:00
|
|
|
if not hasattr(klass, '__safe_for_unpickling__'):
|
|
|
|
raise UnpicklingError('%s is not safe for unpickling' %
|
|
|
|
klass)
|
1998-09-16 04:25:57 +08:00
|
|
|
value = apply(klass, args)
|
|
|
|
except TypeError, err:
|
|
|
|
raise TypeError, "in constructor for %s: %s" % (
|
|
|
|
klass.__name__, str(err)), sys.exc_info()[2]
|
1997-04-10 01:32:51 +08:00
|
|
|
self.append(value)
|
|
|
|
dispatch[INST] = load_inst
|
|
|
|
|
|
|
|
def load_obj(self):
|
|
|
|
stack = self.stack
|
|
|
|
k = self.marker()
|
|
|
|
klass = stack[k + 1]
|
|
|
|
del stack[k + 1]
|
2001-01-15 08:50:52 +08:00
|
|
|
args = tuple(stack[k + 1:])
|
1997-04-10 01:32:51 +08:00
|
|
|
del stack[k:]
|
1998-03-27 05:13:24 +08:00
|
|
|
instantiated = 0
|
|
|
|
if (not args and type(klass) is ClassType and
|
|
|
|
not hasattr(klass, "__getinitargs__")):
|
|
|
|
try:
|
|
|
|
value = _EmptyClass()
|
|
|
|
value.__class__ = klass
|
|
|
|
instantiated = 1
|
|
|
|
except RuntimeError:
|
|
|
|
# In restricted execution, assignment to inst.__class__ is
|
|
|
|
# prohibited
|
|
|
|
pass
|
|
|
|
if not instantiated:
|
|
|
|
value = apply(klass, args)
|
1997-04-10 01:32:51 +08:00
|
|
|
self.append(value)
|
2001-01-15 08:50:52 +08:00
|
|
|
dispatch[OBJ] = load_obj
|
1997-04-10 01:32:51 +08:00
|
|
|
|
|
|
|
def load_global(self):
|
|
|
|
module = self.readline()[:-1]
|
|
|
|
name = self.readline()[:-1]
|
|
|
|
klass = self.find_class(module, name)
|
|
|
|
self.append(klass)
|
|
|
|
dispatch[GLOBAL] = load_global
|
|
|
|
|
|
|
|
def find_class(self, module, name):
|
2001-11-16 07:42:58 +08:00
|
|
|
__import__(module)
|
|
|
|
mod = sys.modules[module]
|
|
|
|
klass = getattr(mod, name)
|
1997-04-10 01:32:51 +08:00
|
|
|
return klass
|
|
|
|
|
|
|
|
def load_reduce(self):
|
|
|
|
stack = self.stack
|
|
|
|
|
|
|
|
callable = stack[-2]
|
|
|
|
arg_tup = stack[-1]
|
1998-03-27 05:13:24 +08:00
|
|
|
del stack[-2:]
|
1997-04-10 01:32:51 +08:00
|
|
|
|
1998-03-27 05:13:24 +08:00
|
|
|
if type(callable) is not ClassType:
|
2002-06-01 22:18:47 +08:00
|
|
|
if not callable in safe_constructors:
|
1998-03-27 05:13:24 +08:00
|
|
|
try:
|
1997-04-10 01:32:51 +08:00
|
|
|
safe = callable.__safe_for_unpickling__
|
|
|
|
except AttributeError:
|
|
|
|
safe = None
|
|
|
|
|
2001-04-10 10:48:53 +08:00
|
|
|
if not safe:
|
2001-01-15 08:50:52 +08:00
|
|
|
raise UnpicklingError, "%s is not safe for " \
|
|
|
|
"unpickling" % callable
|
1997-04-10 01:32:51 +08:00
|
|
|
|
1998-03-27 05:13:24 +08:00
|
|
|
if arg_tup is None:
|
2002-05-22 01:22:02 +08:00
|
|
|
import warnings
|
|
|
|
warnings.warn("The None return argument form of __reduce__ is "
|
|
|
|
"deprecated. Return a tuple of arguments instead.",
|
2002-05-23 23:15:30 +08:00
|
|
|
DeprecationWarning)
|
1998-03-27 05:13:24 +08:00
|
|
|
value = callable.__basicnew__()
|
|
|
|
else:
|
|
|
|
value = apply(callable, arg_tup)
|
1997-04-10 01:32:51 +08:00
|
|
|
self.append(value)
|
|
|
|
dispatch[REDUCE] = load_reduce
|
|
|
|
|
|
|
|
def load_pop(self):
|
|
|
|
del self.stack[-1]
|
|
|
|
dispatch[POP] = load_pop
|
|
|
|
|
|
|
|
def load_pop_mark(self):
|
|
|
|
k = self.marker()
|
1998-03-27 05:13:24 +08:00
|
|
|
del self.stack[k:]
|
1997-04-10 01:32:51 +08:00
|
|
|
dispatch[POP_MARK] = load_pop_mark
|
|
|
|
|
|
|
|
def load_dup(self):
|
1998-04-01 01:00:46 +08:00
|
|
|
self.append(self.stack[-1])
|
1997-04-10 01:32:51 +08:00
|
|
|
dispatch[DUP] = load_dup
|
|
|
|
|
|
|
|
def load_get(self):
|
|
|
|
self.append(self.memo[self.readline()[:-1]])
|
|
|
|
dispatch[GET] = load_get
|
|
|
|
|
|
|
|
def load_binget(self):
|
2003-01-28 05:15:36 +08:00
|
|
|
i = ord(self.read(1))
|
1997-04-10 01:32:51 +08:00
|
|
|
self.append(self.memo[`i`])
|
|
|
|
dispatch[BINGET] = load_binget
|
|
|
|
|
|
|
|
def load_long_binget(self):
|
|
|
|
i = mloads('i' + self.read(4))
|
|
|
|
self.append(self.memo[`i`])
|
|
|
|
dispatch[LONG_BINGET] = load_long_binget
|
|
|
|
|
|
|
|
def load_put(self):
|
|
|
|
self.memo[self.readline()[:-1]] = self.stack[-1]
|
|
|
|
dispatch[PUT] = load_put
|
|
|
|
|
|
|
|
def load_binput(self):
|
2003-01-28 05:15:36 +08:00
|
|
|
i = ord(self.read(1))
|
1997-04-10 01:32:51 +08:00
|
|
|
self.memo[`i`] = self.stack[-1]
|
|
|
|
dispatch[BINPUT] = load_binput
|
|
|
|
|
|
|
|
def load_long_binput(self):
|
|
|
|
i = mloads('i' + self.read(4))
|
|
|
|
self.memo[`i`] = self.stack[-1]
|
|
|
|
dispatch[LONG_BINPUT] = load_long_binput
|
|
|
|
|
|
|
|
def load_append(self):
|
|
|
|
stack = self.stack
|
2002-06-30 11:39:14 +08:00
|
|
|
value = stack.pop()
|
1997-04-10 01:32:51 +08:00
|
|
|
list = stack[-1]
|
|
|
|
list.append(value)
|
|
|
|
dispatch[APPEND] = load_append
|
|
|
|
|
|
|
|
def load_appends(self):
|
|
|
|
stack = self.stack
|
|
|
|
mark = self.marker()
|
|
|
|
list = stack[mark - 1]
|
1998-03-27 05:13:24 +08:00
|
|
|
for i in range(mark + 1, len(stack)):
|
1997-04-10 01:32:51 +08:00
|
|
|
list.append(stack[i])
|
|
|
|
|
|
|
|
del stack[mark:]
|
|
|
|
dispatch[APPENDS] = load_appends
|
2001-01-15 08:50:52 +08:00
|
|
|
|
1997-04-10 01:32:51 +08:00
|
|
|
def load_setitem(self):
|
|
|
|
stack = self.stack
|
2002-06-30 11:39:14 +08:00
|
|
|
value = stack.pop()
|
|
|
|
key = stack.pop()
|
1997-04-10 01:32:51 +08:00
|
|
|
dict = stack[-1]
|
|
|
|
dict[key] = value
|
|
|
|
dispatch[SETITEM] = load_setitem
|
|
|
|
|
|
|
|
def load_setitems(self):
|
|
|
|
stack = self.stack
|
|
|
|
mark = self.marker()
|
|
|
|
dict = stack[mark - 1]
|
1998-03-27 05:13:24 +08:00
|
|
|
for i in range(mark + 1, len(stack), 2):
|
1997-04-10 01:32:51 +08:00
|
|
|
dict[stack[i]] = stack[i + 1]
|
|
|
|
|
|
|
|
del stack[mark:]
|
|
|
|
dispatch[SETITEMS] = load_setitems
|
|
|
|
|
|
|
|
def load_build(self):
|
|
|
|
stack = self.stack
|
2002-06-30 11:39:14 +08:00
|
|
|
value = stack.pop()
|
1997-04-10 01:32:51 +08:00
|
|
|
inst = stack[-1]
|
|
|
|
try:
|
|
|
|
setstate = inst.__setstate__
|
|
|
|
except AttributeError:
|
1998-03-27 05:13:24 +08:00
|
|
|
try:
|
|
|
|
inst.__dict__.update(value)
|
|
|
|
except RuntimeError:
|
|
|
|
# XXX In restricted execution, the instance's __dict__ is not
|
|
|
|
# accessible. Use the old way of unpickling the instance
|
|
|
|
# variables. This is a semantic different when unpickling in
|
|
|
|
# restricted vs. unrestricted modes.
|
|
|
|
for k, v in value.items():
|
|
|
|
setattr(inst, k, v)
|
1997-04-10 01:32:51 +08:00
|
|
|
else:
|
|
|
|
setstate(value)
|
|
|
|
dispatch[BUILD] = load_build
|
|
|
|
|
|
|
|
def load_mark(self):
|
|
|
|
self.append(self.mark)
|
|
|
|
dispatch[MARK] = load_mark
|
|
|
|
|
|
|
|
def load_stop(self):
|
2002-06-30 11:39:14 +08:00
|
|
|
value = self.stack.pop()
|
2000-12-14 02:11:56 +08:00
|
|
|
raise _Stop(value)
|
1997-04-10 01:32:51 +08:00
|
|
|
dispatch[STOP] = load_stop
|
1995-01-10 08:31:14 +08:00
|
|
|
|
1997-12-06 03:42:42 +08:00
|
|
|
# Helper class for load_inst/load_obj
|
|
|
|
|
|
|
|
class _EmptyClass:
|
|
|
|
pass
|
1995-01-10 08:31:14 +08:00
|
|
|
|
1995-03-14 23:09:05 +08:00
|
|
|
# Shorthands
|
|
|
|
|
2001-10-16 05:29:28 +08:00
|
|
|
try:
|
|
|
|
from cStringIO import StringIO
|
|
|
|
except ImportError:
|
|
|
|
from StringIO import StringIO
|
1996-07-23 06:26:07 +08:00
|
|
|
|
2003-01-28 06:47:53 +08:00
|
|
|
def dump(object, file, proto=1):
|
|
|
|
Pickler(file, proto).dump(object)
|
1995-03-14 23:09:05 +08:00
|
|
|
|
2003-01-28 06:47:53 +08:00
|
|
|
def dumps(object, proto=1):
|
1997-04-10 01:32:51 +08:00
|
|
|
file = StringIO()
|
2003-01-28 06:47:53 +08:00
|
|
|
Pickler(file, proto).dump(object)
|
1997-04-10 01:32:51 +08:00
|
|
|
return file.getvalue()
|
1995-03-14 23:09:05 +08:00
|
|
|
|
|
|
|
def load(file):
|
1997-04-10 01:32:51 +08:00
|
|
|
return Unpickler(file).load()
|
1995-03-14 23:09:05 +08:00
|
|
|
|
|
|
|
def loads(str):
|
1997-04-10 01:32:51 +08:00
|
|
|
file = StringIO(str)
|
|
|
|
return Unpickler(file).load()
|