"""Create portable serialized representations of Python objects. See module cPickle for a (much) faster implementation. See module copy_reg for a mechanism for registering custom picklers. See module pickletools source for extensive comments. Classes: Pickler Unpickler Functions: dump(object, file) dumps(object) -> string load(file) -> object loads(string) -> object Misc variables: __version__ format_version compatible_formats """ __version__ = "$Revision$" # Code version from types import * from copy_reg import dispatch_table, safe_constructors import marshal import sys import struct import re __all__ = ["PickleError", "PicklingError", "UnpicklingError", "Pickler", "Unpickler", "dump", "dumps", "load", "loads"] # 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 # Why use struct.pack() for pickling but marshal.loads() for # unpickling? struct.pack() is 40% faster than marshal.loads(), but # marshal.loads() is twice as fast as struct.unpack()! mloads = marshal.loads class PickleError(Exception): """A common base class for the other pickling exceptions.""" 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 class _Stop(Exception): def __init__(self, value): self.value = value try: from org.python.core import PyStringMap except ImportError: PyStringMap = None try: UnicodeType except NameError: UnicodeType = None # 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 # 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 __all__.extend([x for x in dir() if re.match("[A-Z][A-Z0-9_]+$",x)]) del x _quotes = ["'", '"'] class Pickler: def __init__(self, file, proto=1): """This takes a file-like object for writing a pickle data stream. 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. 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. """ if not 0 <= proto <= 2: raise ValueError, "pickle protocol must be 0, 1 or 2" self.write = file.write self.memo = {} self.proto = proto self.bin = proto >= 1 def clear_memo(self): """Clears the pickler's "memo". The memo is the data structure that remembers which objects the 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. """ self.memo.clear() def dump(self, object): """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. """ if self.proto >= 2: self.write(PROTO + chr(self.proto)) self.save(object) self.write(STOP) def memoize(self, obj): """Store an object in the memo.""" # 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 # the key in the Unpickler's memo. The object is stored in the # 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 # scheme allows the Unpickler memo to be implemented as a plain (but # growable) array, indexed by memo key. memo_len = len(self.memo) self.write(self.put(memo_len)) self.memo[id(obj)] = memo_len, obj # Return a PUT (BINPUT, LONG_BINPUT) opcode string, with argument i. def put(self, i, pack=struct.pack): if self.bin: if i < 256: return BINPUT + chr(i) else: return LONG_BINPUT + pack(" 2: state = tup[2] else: state = None 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 self.save_reduce(callable, arg_tup, state) self.memoize(object) def persistent_id(self, object): return None def save_pers(self, pid): if self.bin: self.save(pid) self.write(BINPERSID) else: self.write(PERSID + str(pid) + '\n') def save_reduce(self, acallable, arg_tup, state = None): write = self.write save = self.save if not callable(acallable): raise PicklingError("__reduce__() must return callable as " "first argument, not %s" % `acallable`) save(acallable) save(arg_tup) write(REDUCE) if state is not None: save(state) write(BUILD) dispatch = {} def save_none(self, object): self.write(NONE) dispatch[NoneType] = save_none def save_bool(self, object): self.write(object and TRUE or FALSE) dispatch[bool] = save_bool def save_int(self, object, pack=struct.pack): if self.bin: # 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. # First one- and two-byte unsigned ints: if object >= 0: if object <= 0xff: self.write(BININT1 + chr(object)) return if object <= 0xffff: self.write(BININT2 + chr(object&0xff) + chr(object>>8)) return # Next check for 4-byte signed ints: 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. self.write(BININT + pack("= 2: bytes = encode_long(object) n = len(bytes) if n < 256: self.write(LONG1 + chr(n) + bytes) else: self.write(LONG4 + pack("d', object)) else: self.write(FLOAT + `object` + '\n') dispatch[FloatType] = save_float def save_string(self, object, pack=struct.pack): if self.bin: n = len(object) if n < 256: self.write(SHORT_BINSTRING + chr(n) + object) else: self.write(BINSTRING + pack(" 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 else: # proto 0 -- can't use EMPTY_LIST or APPENDS write(MARK + LIST) self.memoize(object) for element in object: save(element) write(APPEND) dispatch[ListType] = save_list def save_dict(self, object): write = self.write save = self.save items = object.iteritems() if self.bin: write(EMPTY_DICT) self.memoize(object) if len(object) > 1: write(MARK) for key, value in items: save(key) save(value) write(SETITEMS) return else: # proto 0 -- can't use EMPTY_DICT or SETITEMS write(MARK + DICT) self.memoize(object) # proto 0 or len(object) < 2 for key, value in items: save(key) save(value) write(SETITEM) dispatch[DictionaryType] = save_dict if not PyStringMap is None: dispatch[PyStringMap] = save_dict def save_inst(self, 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 _keep_alive(args, memo) else: args = () write(MARK) if self.bin: save(cls) for arg in args: save(arg) write(OBJ) else: for arg in args: save(arg) write(INST + cls.__module__ + '\n' + cls.__name__ + '\n') self.memoize(object) try: getstate = object.__getstate__ except AttributeError: stuff = object.__dict__ else: stuff = getstate() _keep_alive(stuff, memo) save(stuff) write(BUILD) dispatch[InstanceType] = save_inst def save_global(self, object, name = None): write = self.write memo = self.memo if name is None: name = object.__name__ try: module = object.__module__ except AttributeError: module = whichmodule(object, name) 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)) write(GLOBAL + module + '\n' + name + '\n') self.memoize(object) dispatch[ClassType] = save_global dispatch[FunctionType] = save_global dispatch[BuiltinFunctionType] = save_global dispatch[TypeType] = save_global 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: memo[id(memo)].append(x) except KeyError: # aha, this is the first one :-) memo[id(memo)]=[x] classmap = {} # called classmap for backwards compatibility def whichmodule(func, funcname): """Figure out the module in which a function occurs. Search sys.modules for the module. Cache in classmap. Return a module name. If the function cannot be found, return __main__. """ if func in classmap: return classmap[func] for name, module in sys.modules.items(): if module is None: continue # skip dummy package entries if name != '__main__' and \ hasattr(module, funcname) and \ getattr(module, funcname) is func: break else: name = '__main__' classmap[func] = name return name class Unpickler: def __init__(self, file): """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. """ self.readline = file.readline self.read = file.read self.memo = {} def load(self): """Read a pickled object representation from the open file object. Return the reconstituted object hierarchy specified in the file object. """ self.mark = object() # any new unique object self.stack = [] self.append = self.stack.append read = self.read dispatch = self.dispatch try: while 1: key = read(1) dispatch[key](self) except _Stop, stopinst: return stopinst.value # 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. 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_proto(self): proto = ord(self.read(1)) if not 0 <= proto <= 2: raise ValueError, "unsupported pickle protocol: %d" % proto dispatch[PROTO] = load_proto def load_persid(self): pid = self.readline()[:-1] self.append(self.persistent_load(pid)) dispatch[PERSID] = load_persid def load_binpersid(self): pid = self.stack.pop() 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): data = self.readline() 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) dispatch[INT] = load_int def load_binint(self): self.append(mloads('i' + self.read(4))) dispatch[BININT] = load_binint def load_binint1(self): self.append(ord(self.read(1))) dispatch[BININT1] = load_binint1 def load_binint2(self): self.append(mloads('i' + self.read(2) + '\000\000')) dispatch[BININT2] = load_binint2 def load_long(self): self.append(long(self.readline()[:-1], 0)) dispatch[LONG] = load_long def load_long1(self): n = ord(self.read(1)) bytes = self.read(n) return decode_long(bytes) dispatch[LONG1] = load_long1 def load_long4(self): n = mloads('i' + self.read(4)) bytes = self.read(n) return decode_long(bytes) dispatch[LONG4] = load_long4 def load_float(self): self.append(float(self.readline()[:-1])) dispatch[FLOAT] = load_float def load_binfloat(self, unpack=struct.unpack): self.append(unpack('>d', self.read(8))[0]) dispatch[BINFLOAT] = load_binfloat def load_string(self): rep = self.readline()[:-1] 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: raise ValueError, "insecure string pickle" self.append(rep.decode("string-escape")) dispatch[STRING] = load_string 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 def load_binstring(self): len = mloads('i' + self.read(4)) self.append(self.read(len)) dispatch[BINSTRING] = load_binstring 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 def load_short_binstring(self): len = ord(self.read(1)) 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) 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: try: if not hasattr(klass, '__safe_for_unpickling__'): raise UnpicklingError('%s is not safe for unpickling' % klass) value = apply(klass, args) except TypeError, err: raise TypeError, "in constructor for %s: %s" % ( klass.__name__, str(err)), sys.exc_info()[2] 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] args = tuple(stack[k + 1:]) del stack[k:] 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) self.append(value) dispatch[OBJ] = load_obj 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): __import__(module) mod = sys.modules[module] klass = getattr(mod, name) return klass def load_reduce(self): stack = self.stack callable = stack[-2] arg_tup = stack[-1] del stack[-2:] if type(callable) is not ClassType: if not callable in safe_constructors: try: safe = callable.__safe_for_unpickling__ except AttributeError: safe = None if not safe: raise UnpicklingError, "%s is not safe for " \ "unpickling" % callable if arg_tup is None: import warnings warnings.warn("The None return argument form of __reduce__ is " "deprecated. Return a tuple of arguments instead.", DeprecationWarning) value = callable.__basicnew__() else: value = apply(callable, arg_tup) 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() del self.stack[k:] dispatch[POP_MARK] = load_pop_mark def load_dup(self): self.append(self.stack[-1]) dispatch[DUP] = load_dup def load_get(self): self.append(self.memo[self.readline()[:-1]]) dispatch[GET] = load_get def load_binget(self): i = ord(self.read(1)) 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): i = ord(self.read(1)) 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 value = stack.pop() list = stack[-1] list.append(value) dispatch[APPEND] = load_append def load_appends(self): stack = self.stack mark = self.marker() list = stack[mark - 1] list.extend(stack[mark + 1:]) del stack[mark:] dispatch[APPENDS] = load_appends def load_setitem(self): stack = self.stack value = stack.pop() key = stack.pop() dict = stack[-1] dict[key] = value dispatch[SETITEM] = load_setitem def load_setitems(self): stack = self.stack mark = self.marker() dict = stack[mark - 1] for i in range(mark + 1, len(stack), 2): dict[stack[i]] = stack[i + 1] del stack[mark:] dispatch[SETITEMS] = load_setitems def load_build(self): stack = self.stack value = stack.pop() inst = stack[-1] try: setstate = inst.__setstate__ except AttributeError: 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) else: setstate(value) dispatch[BUILD] = load_build def load_mark(self): self.append(self.mark) dispatch[MARK] = load_mark def load_stop(self): value = self.stack.pop() raise _Stop(value) dispatch[STOP] = load_stop # Helper class for load_inst/load_obj class _EmptyClass: pass # Encode/decode longs. def encode_long(x): r"""Encode a long to a two's complement little-ending binary string. >>> encode_long(255L) '\xff\x00' >>> encode_long(32767L) '\xff\x7f' >>> encode_long(-256L) '\x00\xff' >>> encode_long(-32768L) '\x00\x80' >>> encode_long(-128L) '\x80' >>> encode_long(127L) '\x7f' >>> """ digits = [] while not -128 <= x < 128: digits.append(x & 0xff) x >>= 8 digits.append(x & 0xff) return "".join(map(chr, digits)) def decode_long(data): r"""Decode a long from a two's complement little-endian binary string. >>> decode_long("\xff\x00") 255L >>> decode_long("\xff\x7f") 32767L >>> decode_long("\x00\xff") -256L >>> decode_long("\x00\x80") -32768L >>> decode_long("\x80") -128L >>> decode_long("\x7f") 127L """ x = 0L i = 0L for c in data: x |= long(ord(c)) << i i += 8L if data and ord(c) >= 0x80: x -= 1L << i return x # Shorthands try: from cStringIO import StringIO except ImportError: from StringIO import StringIO def dump(object, file, proto=1): Pickler(file, proto).dump(object) def dumps(object, proto=1): file = StringIO() Pickler(file, proto).dump(object) return file.getvalue() def load(file): return Unpickler(file).load() def loads(str): file = StringIO(str) return Unpickler(file).load() # Doctest def _test(): import doctest return doctest.testmod() if __name__ == "__main__": _test()