B
fTd I @ sj
d Z ddlZddlZddlZddlZddlZdddgZejZdZdZ dZ
d Z d
Z G d d d e
Zdd
lmZ dd ZeddeddZdd ZeddeddZdd Zedd ed dZd d Zed d ed!dZd"d# Zed$d%ed&dZdd(d)Z ed*ee d+dZ d,d- Z ed.ee d/dZ d0d1 Z ed2ee d3dZ!d4d5 Z"ed6e e"d7dZ#d8d9 Z$ed:e
e$d;dZ%de e&d?dZ'd@dA Z(edBe e(dCdZ)dDdE Z*edFe e*dGdZ+dHdI Z,edJee,dKdZ-dLdM Z.edNe e.dOdZ/dPdQ Z0edRe e0dSdZ1dTdU Z2edVe e2dWdZ3dXdY Z4dZd[ Z5ed\ee4d]dZ6ed^ee5d_dZ7d`da Z8edbee8dcdZ9ddde Z:edfd%e:dgdZ;ddhlmdmdn Z?edoe
e?dpdZ@G dqdr dre
ZAeAdseBdtdu ZCZDeAdveBeEfdwduZFeAdxeEdyduZGeAdzeHd{duZIeAd|eJeKfd}du ZLZMeAd~eJdduZNeAdeKdduZOeAdePddduZQeAdeRdduZSeAdeTdduZUeAdeVdduZWeAdeXdduZYeAdeXdduZZeAde
dduZ[eAdeAdduZ\eAdeAdduZ]G dd de
Z^e^Z_e_dde6g eFgddde_ddeg eCgddde_ddeg eCgddde_ddeg eCgddde_dde7g eCgddde_dde>g eCgddde_dde@g eCgddde_dde g eLgddde_dde%g eLgddde_dde#g eLgddde_dde)g eNgddde_dde'g eNgddde_dde+g eNgd dde_dddg eQgddde_dddg eGgddde_dddg eGgddde_dde-g eOgddde_dde/g eOgd dde_dde1g eOgddde_dde3g eOgd dde_dde9g eIgddde_dde;g eIgddde_dddg eUgddde_dddeUe[geUgddde_dddeUe\e]geUgddde_ddde\e]geUgddde_dddg eSgddde_ddde\e]geSgddde_ddde[geSgddde_ddde[e[geSgddde_ddde[e[e[geSgddde_dddg eWgddde_ddde\e]geWgddde_dddeWe[e[geWgddde_dddeWe\e]geWgdd de_dddg eYgd dde_dddeYe\e]geYgd dde_ddde\e]geZgd d de_d
d de[gg dd de_d
dde[ge[e[gddde_dddg e\gddde_ddde\e]gg ddde_dde6g e[gddde_ddeg e[gddde_d d eg e[gdd de_d d e6g g dd!de_d"d#eg g dd$de_d%d&eg g dd'de_d(d)de[ge[gd d*de_d+d,eg e[gdd-de_d.d/eg e[gdd0de_d1d2eg e[gdd3de_d4d5e!g e[gdd6de_d7d8deOeOge[gd d9de_d:d;de[e[ge[gddde[e[ge[gdd?de_d@dAe!e\e]ge[gddBde_dCdDde\e[e]ge[gddEde_dFdGde[e[ge[gddHde_dIdJde[e[e[ge[gd dKde_dLdMeg g ddNde_dOdPde[gg ddQde_dRdSeg g d dTde_dUdVe g e[gddWde_dXdYde[ge[gddZdgAZ`[_i Zai Zbxzece`D ]n\ZdZeeejfeak
regd[eejfeaeejf edf eejhebk
regd\eejhebeejh edf edeaeejf< edebeejh<
qxW [a[b[d[ei Zixe`D ]Zeeeeieejh<
qW [edd^d_Zjej [jdd`daZkdbd Zldcd ZmddddZnG dedf dfZodgZpdhZqepeqdiZrdjdk Zsetdlk
rfddluZueujvdmdnZwewjxdoeuydpdqdrds ewjxdtduejzeuydvdwdx ewjxdydzd{d|d} ewjxd~dd eBddx ewjxddd{dd} ewjxddddd ewjxddd{dd} ewjxdd{dd} ew{ Z|e|j} res ne|j~ rʐdndZ~e|j sew nee|jdk
r ene|jd e|jde|je~ nZe|j
ri ndZxHe|jD ]>Ze|jjejfdZe|jed enee|jee|je~
q$W dS ( ar "Executable documentation" for the pickle module.
Extensive comments about the pickle protocols and pickle-machine opcodes
can be found here. Some functions meant for external use:
genops(pickle)
Generate all the opcodes in a pickle, as (opcode, arg, position) triples.
dis(pickle, out=None, memo=None, indentlevel=4)
Print a symbolic disassembly of a pickle.
Ndisgenopsoptimizec @ s e Zd ZdZdd ZdS )ArgumentDescriptor)namenreaderdocc C s` t |tst|| _t |tr8|dks<|ttttt fks>> import io
>>> read_uint1(io.BytesIO(b'\xff'))
255
r z'not enough data in stream to read uint1N)read
ValueError)fdatar r r
read_uint1 s
r& uint1r! zOne-byte unsigned integer.)r r r
r c C s0 | d}t|dkr$td|d S tddS )z
>>> import io
>>> read_uint2(io.BytesIO(b'\xff\x00'))
255
>>> read_uint2(io.BytesIO(b'\xff\xff'))
65535
z>> import io
>>> read_int4(io.BytesIO(b'\xff\x00\x00\x00'))
255
>>> read_int4(io.BytesIO(b'\x00\x00\x00\x80')) == -(2**31)
True
z>> import io
>>> read_uint4(io.BytesIO(b'\xff\x00\x00\x00'))
255
>>> read_uint4(io.BytesIO(b'\x00\x00\x00\x80')) == 2**31
True
r- z>> import io
>>> read_uint8(io.BytesIO(b'\xff\x00\x00\x00\x00\x00\x00\x00'))
255
>>> read_uint8(io.BytesIO(b'\xff' * 8)) == 2**64-1
True
z>> import io
>>> read_stringnl(io.BytesIO(b"'abcd'\nefg\n"))
'abcd'
>>> read_stringnl(io.BytesIO(b"\n"))
Traceback (most recent call last):
...
ValueError: no string quotes around b''
>>> read_stringnl(io.BytesIO(b"\n"), stripquotes=False)
''
>>> read_stringnl(io.BytesIO(b"''\n"))
''
>>> read_stringnl(io.BytesIO(b'"abcd"'))
Traceback (most recent call last):
...
ValueError: no newline found when trying to read stringnl
Embedded escapes are undone in the result.
>>> read_stringnl(io.BytesIO(br"'a\n\\b\x00c\td'" + b"\n'e'"))
'a\n\\b\x00c\td'
z-no newline found when trying to read stringnlNr ) " 'z,strinq quote %r not found at both ends of %rr! zno string quotes around %rr ascii)readlineendswithr#
startswithcodecs
escape_decodedecode)r$ r> stripquotesr% qr r r
read_stringnl; s
rA stringnlzA newline-terminated string.
This is a repr-style string, with embedded escapes, and
bracketing quotes.
c C s t | ddS )NF)r? )rA )r$ r r r read_stringnl_noescapet s rC stringnl_noescapea A newline-terminated string.
This is a str-style string, without embedded escapes,
or bracketing quotes. It should consist solely of
printable ASCII characters.
c C s dt | t | f S )zp
>>> import io
>>> read_stringnl_noescape_pair(io.BytesIO(b"Queue\nEmpty\njunk"))
'Queue Empty'
z%s %s)rC )r$ r r r read_stringnl_noescape_pair s rE stringnl_noescape_paira A pair of newline-terminated strings.
These are str-style strings, without embedded
escapes, or bracketing quotes. They should
consist solely of printable ASCII characters.
The pair is returned as a single string, with
a single blank separating the two strings.
c C sL t | }|dkst| |}t||kr4|dS td|t|f dS )z
>>> import io
>>> read_string1(io.BytesIO(b"\x00"))
''
>>> read_string1(io.BytesIO(b"\x03abcdef"))
'abc'
r zlatin-1z2expected %d bytes in a string1, but only %d remainN)r& r r" r) r> r# )r$ r r% r r r read_string1 s
rG string1zA counted string.
The first argument is a 1-byte unsigned int giving the number
of bytes in the string, and the second argument is that many
bytes.
c C sT t | }|dk r td| | |}t||kr<|dS td|t|f dS )aP
>>> import io
>>> read_string4(io.BytesIO(b"\x00\x00\x00\x00abc"))
''
>>> read_string4(io.BytesIO(b"\x03\x00\x00\x00abcdef"))
'abc'
>>> read_string4(io.BytesIO(b"\x00\x00\x00\x03abcdef"))
Traceback (most recent call last):
...
ValueError: expected 50331648 bytes in a string4, but only 6 remain
r zstring4 byte count < 0: %dzlatin-1z2expected %d bytes in a string4, but only %d remainN)r. r# r" r) r> )r$ r r% r r r read_string4 s
rI string4zA counted string.
The first argument is a 4-byte little-endian signed int giving
the number of bytes in the string, and the second argument is
that many bytes.
c C sF t | }|dkst| |}t||kr.|S td|t|f dS )z
>>> import io
>>> read_bytes1(io.BytesIO(b"\x00"))
b''
>>> read_bytes1(io.BytesIO(b"\x03abcdef"))
b'abc'
r z1expected %d bytes in a bytes1, but only %d remainN)r& r r" r) r# )r$ r r% r r r read_bytes1 s
rK bytes1zA counted bytes string.
The first argument is a 1-byte unsigned int giving the number
of bytes, and the second argument is that many bytes.
c C s\ t | }|dkst|tjkr*td| | |}t||krD|S td|t|f dS )aN
>>> import io
>>> read_bytes4(io.BytesIO(b"\x00\x00\x00\x00abc"))
b''
>>> read_bytes4(io.BytesIO(b"\x03\x00\x00\x00abcdef"))
b'abc'
>>> read_bytes4(io.BytesIO(b"\x00\x00\x00\x03abcdef"))
Traceback (most recent call last):
...
ValueError: expected 50331648 bytes in a bytes4, but only 6 remain
r z#bytes4 byte count > sys.maxsize: %dz1expected %d bytes in a bytes4, but only %d remainN)r0 r sysmaxsizer# r" r) )r$ r r% r r r read_bytes4 s
rO bytes4zA counted bytes string.
The first argument is a 4-byte little-endian unsigned int giving
the number of bytes, and the second argument is that many bytes.
c C s\ t | }|dkst|tjkr*td| | |}t||krD|S td|t|f dS )a
>>> import io, struct, sys
>>> read_bytes8(io.BytesIO(b"\x00\x00\x00\x00\x00\x00\x00\x00abc"))
b''
>>> read_bytes8(io.BytesIO(b"\x03\x00\x00\x00\x00\x00\x00\x00abcdef"))
b'abc'
>>> bigsize8 = struct.pack(">> read_bytes8(io.BytesIO(bigsize8 + b"abcdef")) #doctest: +ELLIPSIS
Traceback (most recent call last):
...
ValueError: expected ... bytes in a bytes8, but only 6 remain
r z#bytes8 byte count > sys.maxsize: %dz1expected %d bytes in a bytes8, but only %d remainN)r3 r rM rN r# r" r) )r$ r r% r r r read_bytes8 s
rQ bytes8zA counted bytes string.
The first argument is an 8-byte little-endian unsigned int giving
the number of bytes, and the second argument is that many bytes.
c C s0 | }|dstd|dd }t|dS )zm
>>> import io
>>> read_unicodestringnl(io.BytesIO(b"abc\\uabcd\njunk")) == 'abc\uabcd'
True
r5 z4no newline found when trying to read unicodestringnlNr zraw-unicode-escape)r9 r: r# r )r$ r% r r r read_unicodestringnl8 s
rS unicodestringnlzA newline-terminated Unicode string.
This is raw-unicode-escape encoded, so consists of
printable ASCII characters, and may contain embedded
escape sequences.
c C sN t | }|dkst| |}t||kr6t|ddS td|t|f dS )a
>>> import io
>>> s = 'abcd\uabcd'
>>> enc = s.encode('utf-8')
>>> enc
b'abcd\xea\xaf\x8d'
>>> n = bytes([len(enc)]) # little-endian 1-byte length
>>> t = read_unicodestring1(io.BytesIO(n + enc + b'junk'))
>>> s == t
True
>>> read_unicodestring1(io.BytesIO(n + enc[:-1]))
Traceback (most recent call last):
...
ValueError: expected 7 bytes in a unicodestring1, but only 6 remain
r zutf-8
surrogatepassz9expected %d bytes in a unicodestring1, but only %d remainN)r& r r" r) r r# )r$ r r% r r r read_unicodestring1R s
rV unicodestring1aA A counted Unicode string.
The first argument is a 1-byte little-endian signed int
giving the number of bytes in the string, and the second
argument-- the UTF-8 encoding of the Unicode string --
contains that many bytes.
c C sd t | }|dkst|tjkr*td| | |}t||krLt|ddS td|t|f dS )a
>>> import io
>>> s = 'abcd\uabcd'
>>> enc = s.encode('utf-8')
>>> enc
b'abcd\xea\xaf\x8d'
>>> n = bytes([len(enc), 0, 0, 0]) # little-endian 4-byte length
>>> t = read_unicodestring4(io.BytesIO(n + enc + b'junk'))
>>> s == t
True
>>> read_unicodestring4(io.BytesIO(n + enc[:-1]))
Traceback (most recent call last):
...
ValueError: expected 7 bytes in a unicodestring4, but only 6 remain
r z+unicodestring4 byte count > sys.maxsize: %dzutf-8rU z9expected %d bytes in a unicodestring4, but only %d remainN)r0 r rM rN r# r" r) r )r$ r r% r r r read_unicodestring4y s
rX unicodestring4aA A counted Unicode string.
The first argument is a 4-byte little-endian signed int
giving the number of bytes in the string, and the second
argument-- the UTF-8 encoding of the Unicode string --
contains that many bytes.
c C sd t | }|dkst|tjkr*td| | |}t||krLt|ddS td|t|f dS )a
>>> import io
>>> s = 'abcd\uabcd'
>>> enc = s.encode('utf-8')
>>> enc
b'abcd\xea\xaf\x8d'
>>> n = bytes([len(enc)]) + b'\0' * 7 # little-endian 8-byte length
>>> t = read_unicodestring8(io.BytesIO(n + enc + b'junk'))
>>> s == t
True
>>> read_unicodestring8(io.BytesIO(n + enc[:-1]))
Traceback (most recent call last):
...
ValueError: expected 7 bytes in a unicodestring8, but only 6 remain
r z+unicodestring8 byte count > sys.maxsize: %dzutf-8rU z9expected %d bytes in a unicodestring8, but only %d remainN)r3 r rM rN r# r" r) r )r$ r r% r r r read_unicodestring8 s
rZ unicodestring8aB A counted Unicode string.
The first argument is an 8-byte little-endian signed int
giving the number of bytes in the string, and the second
argument-- the UTF-8 encoding of the Unicode string --
contains that many bytes.
c C s. t | ddd}|dkrdS |dkr&dS t|S )z
>>> import io
>>> read_decimalnl_short(io.BytesIO(b"1234\n56"))
1234
>>> read_decimalnl_short(io.BytesIO(b"1234L\n56"))
Traceback (most recent call last):
...
ValueError: invalid literal for int() with base 10: b'1234L'
F)r> r? s 00s 01T)rA r )r$ sr r r read_decimalnl_short s r] c C s2 t | ddd}|dd dkr*|dd }t|S )z
>>> import io
>>> read_decimalnl_long(io.BytesIO(b"1234L\n56"))
1234
>>> read_decimalnl_long(io.BytesIO(b"123456789012345678901234L\n6"))
123456789012345678901234
F)r> r? r N L)rA r )r$ r\ r r r read_decimalnl_long s r_ decimalnl_shorta A newline-terminated decimal integer literal.
This never has a trailing 'L', and the integer fit
in a short Python int on the box where the pickle
was written -- but there's no guarantee it will fit
in a short Python int on the box where the pickle
is read.
decimalnl_longzA newline-terminated decimal integer literal.
This has a trailing 'L', and can represent integers
of any size.
c C s t | ddd}t|S )zO
>>> import io
>>> read_floatnl(io.BytesIO(b"-1.25\n6"))
-1.25
F)r> r? )rA float)r$ r\ r r r read_floatnl
s rc floatnla A newline-terminated decimal floating literal.
In general this requires 17 significant digits for roundtrip
identity, and pickling then unpickling infinities, NaNs, and
minus zero doesn't work across boxes, or on some boxes even
on itself (e.g., Windows can't read the strings it produces
for infinities or NaNs).
c C s0 | d}t|dkr$td|d S tddS )z
>>> import io, struct
>>> raw = struct.pack(">d", -1.25)
>>> raw
b'\xbf\xf4\x00\x00\x00\x00\x00\x00'
>>> read_float8(io.BytesIO(raw + b"\n"))
-1.25
r2 z>dr z(not enough data in stream to read float8N)r" r) r* r# )r$ r% r r r read_float8 s
re float8a An 8-byte binary representation of a float, big-endian.
The format is unique to Python, and shared with the struct
module (format string '>d') "in theory" (the struct and pickle
implementations don't share the code -- they should). It's
strongly related to the IEEE-754 double format, and, in normal
cases, is in fact identical to the big-endian 754 double format.
On other boxes the dynamic range is limited to that of a 754
double, and "add a half and chop" rounding is used to reduce
the precision to 53 bits. However, even on a 754 box,
infinities, NaNs, and minus zero may not be handled correctly
(may not survive roundtrip pickling intact).
) decode_longc C s. t | }| |}t||kr&tdt|S )a+
>>> import io
>>> read_long1(io.BytesIO(b"\x00"))
0
>>> read_long1(io.BytesIO(b"\x02\xff\x00"))
255
>>> read_long1(io.BytesIO(b"\x02\xff\x7f"))
32767
>>> read_long1(io.BytesIO(b"\x02\x00\xff"))
-256
>>> read_long1(io.BytesIO(b"\x02\x00\x80"))
-32768
z'not enough data in stream to read long1)r& r" r) r# rg )r$ r r% r r r
read_long1F s
rh long1a A binary long, little-endian, using 1-byte size.
This first reads one byte as an unsigned size, then reads that
many bytes and interprets them as a little-endian 2's-complement long.
If the size is 0, that's taken as a shortcut for the long 0L.
c C sB t | }|dk r td| | |}t||kr:tdt|S )ag
>>> import io
>>> read_long4(io.BytesIO(b"\x02\x00\x00\x00\xff\x00"))
255
>>> read_long4(io.BytesIO(b"\x02\x00\x00\x00\xff\x7f"))
32767
>>> read_long4(io.BytesIO(b"\x02\x00\x00\x00\x00\xff"))
-256
>>> read_long4(io.BytesIO(b"\x02\x00\x00\x00\x00\x80"))
-32768
>>> read_long1(io.BytesIO(b"\x00\x00\x00\x00"))
0
r zlong4 byte count < 0: %dz'not enough data in stream to read long4)r. r# r" r) rg )r$ r r% r r r
read_long4f s
rj long4a A binary representation of a long, little-endian.
This first reads four bytes as a signed size (but requires the
size to be >= 0), then reads that many bytes and interprets them
as a little-endian 2's-complement long. If the size is 0, that's taken
as a shortcut for the int 0, although LONG1 should really be used
then instead (and in any case where # of bytes < 256).
c @ s e Zd ZdZdd Zdd ZdS ) StackObject)r obtyper c C sp t |tst|| _t |ts,t |ts,tt |trRx|D ]}t |tsbinary
conversion). Proto 2 added linear-time (in C; still quadratic-time
in Python) LONG1 and LONG4 opcodes.
ZLONG1 z|Long integer using one-byte length.
A more efficient encoding of a Python long; the long1 encoding
says it all.ZLONG4 z~Long integer using found-byte length.
A more efficient encoding of a Python long; the long4 encoding
says it all.STRINGSa Push a Python string object.
The argument is a repr-style string, with bracketing quote characters,
and perhaps embedded escapes. The argument extends until the next
newline character. These are usually decoded into a str instance
using the encoding given to the Unpickler constructor. or the default,
'ASCII'. If the encoding given was 'bytes' however, they will be
decoded as bytes object instead.
Z BINSTRINGTa Push a Python string object.
There are two arguments: the first is a 4-byte little-endian
signed int giving the number of bytes in the string, and the
second is that many bytes, which are taken literally as the string
content. These are usually decoded into a str instance using the
encoding given to the Unpickler constructor. or the default,
'ASCII'. If the encoding given was 'bytes' however, they will be
decoded as bytes object instead.
ZSHORT_BINSTRINGUa Push a Python string object.
There are two arguments: the first is a 1-byte unsigned int giving
the number of bytes in the string, and the second is that many
bytes, which are taken literally as the string content. These are
usually decoded into a str instance using the encoding given to
the Unpickler constructor. or the default, 'ASCII'. If the
encoding given was 'bytes' however, they will be decoded as bytes
object instead.
ZBINBYTESB zPush a Python bytes object.
There are two arguments: the first is a 4-byte little-endian unsigned int
giving the number of bytes, and the second is that many bytes, which are
taken literally as the bytes content.
ZSHORT_BINBYTESCzPush a Python bytes object.
There are two arguments: the first is a 1-byte unsigned int giving
the number of bytes, and the second is that many bytes, which are taken
literally as the string content.
Z BINBYTES8 zPush a Python bytes object.
There are two arguments: the first is an 8-byte unsigned int giving
the number of bytes in the string, and the second is that many bytes,
which are taken literally as the string content.
ZNONENzPush None on the stack.ZNEWTRUE zPush True onto the stack.ZNEWFALSE zPush False onto the stack.UNICODEVzPush a Python Unicode string object.
The argument is a raw-unicode-escape encoding of a Unicode string,
and so may contain embedded escape sequences. The argument extends
until the next newline character.
ZSHORT_BINUNICODE a Push a Python Unicode string object.
There are two arguments: the first is a 1-byte little-endian signed int
giving the number of bytes in the string. The second is that many
bytes, and is the UTF-8 encoding of the Unicode string.
Z
BINUNICODEXa Push a Python Unicode string object.
There are two arguments: the first is a 4-byte little-endian unsigned int
giving the number of bytes in the string. The second is that many
bytes, and is the UTF-8 encoding of the Unicode string.
Z BINUNICODE8 a Push a Python Unicode string object.
There are two arguments: the first is an 8-byte little-endian signed int
giving the number of bytes in the string. The second is that many
bytes, and is the UTF-8 encoding of the Unicode string.
ZFLOATFa Newline-terminated decimal float literal.
The argument is repr(a_float), and in general requires 17 significant
digits for roundtrip conversion to be an identity (this is so for
IEEE-754 double precision values, which is what Python float maps to
on most boxes).
In general, FLOAT cannot be used to transport infinities, NaNs, or
minus zero across boxes (or even on a single box, if the platform C
library can't read the strings it produces for such things -- Windows
is like that), but may do less damage than BINFLOAT on boxes with
greater precision or dynamic range than IEEE-754 double.
ZBINFLOATGa Float stored in binary form, with 8 bytes of data.
This generally requires less than half the space of FLOAT encoding.
In general, BINFLOAT cannot be used to transport infinities, NaNs, or
minus zero, raises an exception if the exponent exceeds the range of
an IEEE-754 double, and retains no more than 53 bits of precision (if
there are more than that, "add a half and chop" rounding is used to
cut it back to 53 significant bits).
Z
EMPTY_LIST]zPush an empty list.ZAPPENDazAppend an object to a list.
Stack before: ... pylist anyobject
Stack after: ... pylist+[anyobject]
although pylist is really extended in-place.
ZAPPENDSezExtend a list by a slice of stack objects.
Stack before: ... pylist markobject stackslice
Stack after: ... pylist+stackslice
although pylist is really extended in-place.
ZLISTlas Build a list out of the topmost stack slice, after markobject.
All the stack entries following the topmost markobject are placed into
a single Python list, which single list object replaces all of the
stack from the topmost markobject onward. For example,
Stack before: ... markobject 1 2 3 'abc'
Stack after: ... [1, 2, 3, 'abc']
Z EMPTY_TUPLE)zPush an empty tuple.ZTUPLEtav Build a tuple out of the topmost stack slice, after markobject.
All the stack entries following the topmost markobject are placed into
a single Python tuple, which single tuple object replaces all of the
stack from the topmost markobject onward. For example,
Stack before: ... markobject 1 2 3 'abc'
Stack after: ... (1, 2, 3, 'abc')
ZTUPLE1 zBuild a one-tuple out of the topmost item on the stack.
This code pops one value off the stack and pushes a tuple of
length 1 whose one item is that value back onto it. In other
words:
stack[-1] = tuple(stack[-1:])
ZTUPLE2 a Build a two-tuple out of the top two items on the stack.
This code pops two values off the stack and pushes a tuple of
length 2 whose items are those values back onto it. In other
words:
stack[-2:] = [tuple(stack[-2:])]
ZTUPLE3 a Build a three-tuple out of the top three items on the stack.
This code pops three values off the stack and pushes a tuple of
length 3 whose items are those values back onto it. In other
words:
stack[-3:] = [tuple(stack[-3:])]
Z
EMPTY_DICT}zPush an empty dict.ZDICTda Build a dict out of the topmost stack slice, after markobject.
All the stack entries following the topmost markobject are placed into
a single Python dict, which single dict object replaces all of the
stack from the topmost markobject onward. The stack slice alternates
key, value, key, value, .... For example,
Stack before: ... markobject 1 2 3 'abc'
Stack after: ... {1: 2, 3: 'abc'}
ZSETITEMr\ zAdd a key+value pair to an existing dict.
Stack before: ... pydict key value
Stack after: ... pydict
where pydict has been modified via pydict[key] = value.
ZSETITEMSua\ Add an arbitrary number of key+value pairs to an existing dict.
The slice of the stack following the topmost markobject is taken as
an alternating sequence of keys and values, added to the dict
immediately under the topmost markobject. Everything at and after the
topmost markobject is popped, leaving the mutated dict at the top
of the stack.
Stack before: ... pydict markobject key_1 value_1 ... key_n value_n
Stack after: ... pydict
where pydict has been modified via pydict[key_i] = value_i for i in
1, 2, ..., n, and in that order.
Z EMPTY_SET zPush an empty set.ZADDITEMS a$ Add an arbitrary number of items to an existing set.
The slice of the stack following the topmost markobject is taken as
a sequence of items, added to the set immediately under the topmost
markobject. Everything at and after the topmost markobject is popped,
leaving the mutated set at the top of the stack.
Stack before: ... pyset markobject item_1 ... item_n
Stack after: ... pyset
where pyset has been modified via pyset.add(item_i) = item_i for i in
1, 2, ..., n, and in that order.
Z FROZENSET az Build a frozenset out of the topmost slice, after markobject.
All the stack entries following the topmost markobject are placed into
a single Python frozenset, which single frozenset object replaces all
of the stack from the topmost markobject onward. For example,
Stack before: ... markobject 1 2 3
Stack after: ... frozenset({1, 2, 3})
POP0z.
ZBUILDba Finish building an object, via __setstate__ or dict update.
Stack before: ... anyobject argument
Stack after: ... anyobject
where anyobject may have been mutated, as follows:
If the object has a __setstate__ method,
anyobject.__setstate__(argument)
is called.
Else the argument must be a dict, the object must have a __dict__, and
the object is updated via
anyobject.__dict__.update(argument)
ZINSTiaq Build a class instance.
This is the protocol 0 version of protocol 1's OBJ opcode.
INST is followed by two newline-terminated strings, giving a
module and class name, just as for the GLOBAL opcode (and see
GLOBAL for more details about that). self.find_class(module, name)
is used to get a class object.
In addition, all the objects on the stack following the topmost
markobject are gathered into a tuple and popped (along with the
topmost markobject), just as for the TUPLE opcode.
Now it gets complicated. If all of these are true:
+ The argtuple is empty (markobject was at the top of the stack
at the start).
+ The class object does not have a __getinitargs__ attribute.
then we want to create an old-style class instance without invoking
its __init__() method (pickle has waffled on this over the years; not
calling __init__() is current wisdom). In this case, an instance of
an old-style dummy class is created, and then we try to rebind its
__class__ attribute to the desired class object. If this succeeds,
the new instance object is pushed on the stack, and we're done.
Else (the argtuple is not empty, it's not an old-style class object,
or the class object does have a __getinitargs__ attribute), the code
first insists that the class object have a __safe_for_unpickling__
attribute. Unlike as for the __safe_for_unpickling__ check in REDUCE,
it doesn't matter whether this attribute has a true or false value, it
only matters whether it exists (XXX this is a bug). If
__safe_for_unpickling__ doesn't exist, UnpicklingError is raised.
Else (the class object does have a __safe_for_unpickling__ attr),
the class object obtained from INST's arguments is applied to the
argtuple obtained from the stack, and the resulting instance object
is pushed on the stack.
NOTE: checks for __safe_for_unpickling__ went away in Python 2.3.
NOTE: the distinction between old-style and new-style classes does
not make sense in Python 3.
ZOBJoa Build a class instance.
This is the protocol 1 version of protocol 0's INST opcode, and is
very much like it. The major difference is that the class object
is taken off the stack, allowing it to be retrieved from the memo
repeatedly if several instances of the same class are created. This
can be much more efficient (in both time and space) than repeatedly
embedding the module and class names in INST opcodes.
Unlike INST, OBJ takes no arguments from the opcode stream. Instead
the class object is taken off the stack, immediately above the
topmost markobject:
Stack before: ... markobject classobject stackslice
Stack after: ... new_instance_object
As for INST, the remainder of the stack above the markobject is
gathered into an argument tuple, and then the logic seems identical,
except that no __safe_for_unpickling__ check is done (XXX this is
a bug). See INST for the gory details.
NOTE: In Python 2.3, INST and OBJ are identical except for how they
get the class object. That was always the intent; the implementations
had diverged for accidental reasons.
ZNEWOBJ aL Build an object instance.
The stack before should be thought of as containing a class
object followed by an argument tuple (the tuple being the stack
top). Call these cls and args. They are popped off the stack,
and the value returned by cls.__new__(cls, *args) is pushed back
onto the stack.
Z NEWOBJ_EX au Build an object instance.
The stack before should be thought of as containing a class
object followed by an argument tuple and by a keyword argument dict
(the dict being the stack top). Call these cls and args. They are
popped off the stack, and the value returned by
cls.__new__(cls, *args, *kwargs) is pushed back onto the stack.
PROTO zProtocol version indicator.
For protocol 2 and above, a pickle must start with this opcode.
The argument is the protocol version, an int in range(2, 256).
STOP.zStop the unpickling machine.
Every pickle ends with this opcode. The object at the top of the stack
is popped, and that's the result of unpickling. The stack should be
empty then.
FRAME zIndicate the beginning of a new frame.
The unpickler may use this opcode to safely prefetch data from its
underlying stream.
ZPERSIDPa Push an object identified by a persistent ID.
The pickle module doesn't define what a persistent ID means. PERSID's
argument is a newline-terminated str-style (no embedded escapes, no
bracketing quote characters) string, which *is* "the persistent ID".
The unpickler passes this string to self.persistent_load(). Whatever
object that returns is pushed on the stack. There is no implementation
of persistent_load() in Python's unpickler: it must be supplied by an
unpickler subclass.
Z BINPERSIDQaX Push an object identified by a persistent ID.
Like PERSID, except the persistent ID is popped off the stack (instead
of being a string embedded in the opcode bytestream). The persistent
ID is passed to self.persistent_load(), and whatever object that
returns is pushed on the stack. See PERSID for more detail.
z%repeated name %r at indices %d and %dz%repeated code %r at indices %d and %dFc C s t }xtjD ]}td|s2| rtd| qtt|}t|t rRt
|dkrh| rtd||f q| d}||kr| rtd||f || }|j |krt
d|||j f ||= qt
d||f qW |rd g}x(| D ] \}}|d
|j |f qW t
d |d S ) Nz[A-Z][A-Z0-9_]+$z0skipping %r: it doesn't look like an opcode namer! z5skipping %r: value %r doesn't look like a pickle codezlatin-1z+checking name %r w/ code %r for consistencyzBfor pickle code %r, pickle.py uses name %r but we're using name %rzPpickle.py appears to have a pickle opcode with name %r and code %r, but we don'tz=we appear to have pickle opcodes that pickle.py doesn't have:z name %r with code %r
)code2opcopyr __all__rematchprintgetattrr rr r) r> r r# itemsappendjoin)verboser r Z
picklecoder msgr{ r r r assure_pickle_consistencyb s>
r c c s t | trt| } t| dr&| j}ndd }x| }| d}t| d}|d kr|dkrjt
dn t
d|d krzd n||f |j d krd }n |j | }|r|||| fV n |||fV |d
kr0|j
d kstP q0W d S ) Ntellc S s d S )Nr r r r r z_genops..r! zlatin-1r z#pickle exhausted before seeing STOPz!at position %s, opcode %r unknownz .r )r bytes_typesioBytesIOhasattrr r" r getr> r# r| r
r r )r%
yield_end_posZgetposposr{ opcoder| r r r _genops s0
r c C s t | S )ax Generate all the opcodes in a pickle.
'pickle' is a file-like object, or string, containing the pickle.
Each opcode in the pickle is generated, from the current pickle position,
stopping after a STOP opcode is delivered. A triple is generated for
each opcode:
opcode, arg, pos
opcode is an OpcodeInfo record, describing the current opcode.
If the opcode has an argument embedded in the pickle, arg is its decoded
value, as a Python object. If the opcode doesn't have an argument, arg
is None.
If the pickle has a tell() method, pos was the value of pickle.tell()
before reading the current opcode. If the pickle is a bytes object,
it's wrapped in a BytesIO object, and the latter's tell() result is
used. Else (the pickle doesn't have a tell(), and it's not obvious how
to query its current position) pos is None.
)r )r r r r r s c C s d}d}t }i }g }d}d}xt| ddD ]\}} }
} d|jkr\|| ||| f q,|jdkrt|} || ||| f q,d|jkrq,d|jkr|j|kr|j}d || < ||| f q,|jd
kr| |kr| }|
dkr| |
| }n||
| f q,||
| f q,W ~t }
|
| t
|
|}|d krJ|j
d} x|D ]\}} d }||kr| |krxqT|| }| || < | d
7 } n6||kr||| }n | || }t||j jk}|j j|d |r|j | n
| | qTW |j |
S )z7Optimize a pickle string by removing unused PUT opcodesr r r r T)r r r Nr r- Fr! )Zforce)rv r r addr r) r r r writer Z_PicklerZframerZ
start_framingputr Z_FRAME_SIZE_TARGETZ commit_frameZ
file_writeZ end_framinggetvalue)r r r ZoldidsZnewidsopcodesr Z protoheaderr r| r Zend_posidxoutZpickleropZ framelessr% r r r r sl
c C sx g }|dkri }d}g }d| }d} |}
x(t | D ]\} } }
|
dk rZtd|
d|d dt| jdd |t| | jf }t|| j}| j}| j }t|}d}t
|ks| jdkr|r|d t
krt
|kst t
|kr|d t kst |rx|
}|dkrd }nd
| }x |d t
k r2|
qW |
y|t
}W n* tk
rt | jdkslt d }Y nX nd } }| jd
kr| jdkrt|}d| }n| dk st | }||krd| } n,|sd} n |d t
krd} n |d ||< n<| jdkr:| |kr2t|dks&t || g}nd| } | dk sJ|r|ddt| j 7 }| dk rz|dt| 7 }|r|d| 7 }|r|d|
t| 7 }t|}
|
dkr|}
|d| jddd 7 }t||d | rt| t||k rtd|t|f |r$|| d= t
|krFt
|ks>> import pickle
>>> x = [1, 2, (3, 4), {b'abc': "def"}]
>>> pkl0 = pickle.dumps(x, 0)
>>> dis(pkl0)
0: ( MARK
1: l LIST (MARK at 0)
2: p PUT 0
5: I INT 1
8: a APPEND
9: I INT 2
12: a APPEND
13: ( MARK
14: I INT 3
17: I INT 4
20: t TUPLE (MARK at 13)
21: p PUT 1
24: a APPEND
25: ( MARK
26: d DICT (MARK at 25)
27: p PUT 2
30: c GLOBAL '_codecs encode'
46: p PUT 3
49: ( MARK
50: V UNICODE 'abc'
55: p PUT 4
58: V UNICODE 'latin1'
66: p PUT 5
69: t TUPLE (MARK at 49)
70: p PUT 6
73: R REDUCE
74: p PUT 7
77: V UNICODE 'def'
82: p PUT 8
85: s SETITEM
86: a APPEND
87: . STOP
highest protocol among opcodes = 0
Try again with a "binary" pickle.
>>> pkl1 = pickle.dumps(x, 1)
>>> dis(pkl1)
0: ] EMPTY_LIST
1: q BINPUT 0
3: ( MARK
4: K BININT1 1
6: K BININT1 2
8: ( MARK
9: K BININT1 3
11: K BININT1 4
13: t TUPLE (MARK at 8)
14: q BINPUT 1
16: } EMPTY_DICT
17: q BINPUT 2
19: c GLOBAL '_codecs encode'
35: q BINPUT 3
37: ( MARK
38: X BINUNICODE 'abc'
46: q BINPUT 4
48: X BINUNICODE 'latin1'
59: q BINPUT 5
61: t TUPLE (MARK at 37)
62: q BINPUT 6
64: R REDUCE
65: q BINPUT 7
67: X BINUNICODE 'def'
75: q BINPUT 8
77: s SETITEM
78: e APPENDS (MARK at 3)
79: . STOP
highest protocol among opcodes = 1
Exercise the INST/OBJ/BUILD family.
>>> import pickletools
>>> dis(pickle.dumps(pickletools.dis, 0))
0: c GLOBAL 'pickletools dis'
17: p PUT 0
20: . STOP
highest protocol among opcodes = 0
>>> from pickletools import _Example
>>> x = [_Example(42)] * 2
>>> dis(pickle.dumps(x, 0))
0: ( MARK
1: l LIST (MARK at 0)
2: p PUT 0
5: c GLOBAL 'copy_reg _reconstructor'
30: p PUT 1
33: ( MARK
34: c GLOBAL 'pickletools _Example'
56: p PUT 2
59: c GLOBAL '__builtin__ object'
79: p PUT 3
82: N NONE
83: t TUPLE (MARK at 33)
84: p PUT 4
87: R REDUCE
88: p PUT 5
91: ( MARK
92: d DICT (MARK at 91)
93: p PUT 6
96: V UNICODE 'value'
103: p PUT 7
106: I INT 42
110: s SETITEM
111: b BUILD
112: a APPEND
113: g GET 5
116: a APPEND
117: . STOP
highest protocol among opcodes = 0
>>> dis(pickle.dumps(x, 1))
0: ] EMPTY_LIST
1: q BINPUT 0
3: ( MARK
4: c GLOBAL 'copy_reg _reconstructor'
29: q BINPUT 1
31: ( MARK
32: c GLOBAL 'pickletools _Example'
54: q BINPUT 2
56: c GLOBAL '__builtin__ object'
76: q BINPUT 3
78: N NONE
79: t TUPLE (MARK at 31)
80: q BINPUT 4
82: R REDUCE
83: q BINPUT 5
85: } EMPTY_DICT
86: q BINPUT 6
88: X BINUNICODE 'value'
98: q BINPUT 7
100: K BININT1 42
102: s SETITEM
103: b BUILD
104: h BINGET 5
106: e APPENDS (MARK at 3)
107: . STOP
highest protocol among opcodes = 1
Try "the canonical" recursive-object test.
>>> L = []
>>> T = L,
>>> L.append(T)
>>> L[0] is T
True
>>> T[0] is L
True
>>> L[0][0] is L
True
>>> T[0][0] is T
True
>>> dis(pickle.dumps(L, 0))
0: ( MARK
1: l LIST (MARK at 0)
2: p PUT 0
5: ( MARK
6: g GET 0
9: t TUPLE (MARK at 5)
10: p PUT 1
13: a APPEND
14: . STOP
highest protocol among opcodes = 0
>>> dis(pickle.dumps(L, 1))
0: ] EMPTY_LIST
1: q BINPUT 0
3: ( MARK
4: h BINGET 0
6: t TUPLE (MARK at 3)
7: q BINPUT 1
9: a APPEND
10: . STOP
highest protocol among opcodes = 1
Note that, in the protocol 0 pickle of the recursive tuple, the disassembler
has to emulate the stack in order to realize that the POP opcode at 16 gets
rid of the MARK at 0.
>>> dis(pickle.dumps(T, 0))
0: ( MARK
1: ( MARK
2: l LIST (MARK at 1)
3: p PUT 0
6: ( MARK
7: g GET 0
10: t TUPLE (MARK at 6)
11: p PUT 1
14: a APPEND
15: 0 POP
16: 0 POP (MARK at 0)
17: g GET 1
20: . STOP
highest protocol among opcodes = 0
>>> dis(pickle.dumps(T, 1))
0: ( MARK
1: ] EMPTY_LIST
2: q BINPUT 0
4: ( MARK
5: h BINGET 0
7: t TUPLE (MARK at 4)
8: q BINPUT 1
10: a APPEND
11: 1 POP_MARK (MARK at 0)
12: h BINGET 1
14: . STOP
highest protocol among opcodes = 1
Try protocol 2.
>>> dis(pickle.dumps(L, 2))
0: \x80 PROTO 2
2: ] EMPTY_LIST
3: q BINPUT 0
5: h BINGET 0
7: \x85 TUPLE1
8: q BINPUT 1
10: a APPEND
11: . STOP
highest protocol among opcodes = 2
>>> dis(pickle.dumps(T, 2))
0: \x80 PROTO 2
2: ] EMPTY_LIST
3: q BINPUT 0
5: h BINGET 0
7: \x85 TUPLE1
8: q BINPUT 1
10: a APPEND
11: 0 POP
12: h BINGET 1
14: . STOP
highest protocol among opcodes = 2
Try protocol 3 with annotations:
>>> dis(pickle.dumps(T, 3), annotate=1)
0: \x80 PROTO 3 Protocol version indicator.
2: ] EMPTY_LIST Push an empty list.
3: q BINPUT 0 Store the stack top into the memo. The stack is not popped.
5: h BINGET 0 Read an object from the memo and push it on the stack.
7: \x85 TUPLE1 Build a one-tuple out of the topmost item on the stack.
8: q BINPUT 1 Store the stack top into the memo. The stack is not popped.
10: a APPEND Append an object to a list.
11: 0 POP Discard the top stack item, shrinking the stack by one item.
12: h BINGET 1 Read an object from the memo and push it on the stack.
14: . STOP Stop the unpickling machine.
highest protocol among opcodes = 2
a=
>>> import pickle
>>> import io
>>> f = io.BytesIO()
>>> p = pickle.Pickler(f, 2)
>>> x = [1, 2, 3]
>>> p.dump(x)
>>> p.dump(x)
>>> f.seek(0)
0
>>> memo = {}
>>> dis(f, memo=memo)
0: \x80 PROTO 2
2: ] EMPTY_LIST
3: q BINPUT 0
5: ( MARK
6: K BININT1 1
8: K BININT1 2
10: K BININT1 3
12: e APPENDS (MARK at 5)
13: . STOP
highest protocol among opcodes = 2
>>> dis(f, memo=memo)
14: \x80 PROTO 2
16: h BINGET 0
18: . STOP
highest protocol among opcodes = 2
)Zdisassembler_testZdisassembler_memo_testc C s dd l } | S )Nr )doctestZtestmod)r r r r _test
s r __main__z$disassemble one or more pickle files)Z description pickle_filebr*zthe pickle file)rn nargshelpz-oz--outputwz+the file where the output should be written)defaultrn r
z-mz--memo
store_truez#preserve memo between disassemblies)actionr
z-lz
--indentlevelz8the number of blanks by which to indent a new MARK levelz-az
--annotatez2annotate each line with a short opcode descriptionz-pz
--preamblez==> {name} <==zMif more than one pickle file is specified, print this before each disassembly)r r
z-tz--testzrun self-test suitez-vz)run verbosely; only affects self-test run )r r )TT)F)F)NNr- r )__doc__r< r r r rM r r r r r r r objectr
Zstructr r* r& r' r+ r, r. r/ r0 r1 r3 r4 rA rB rC rD rE rF rG rH rI rJ rK rL rO rP rQ rR rS rT rV rW rX rY rZ r[ r] r_ r` ra rc rd re rf rg rh ri rj rk rl r ZpyintZpylongrq Zpyinteger_or_boolZpyboolrb Zpyfloatrr r Zpybytes_or_strZpystringZpybytesZ pyunicodern Zpynonero Zpytuplert Zpylistru Zpydictrv ZpysetZ pyfrozensetZ anyobjectr ry rz r r Zname2iZcode2i enumerater r r r# r{ r r r r r r r Z _dis_testZ
_memo_testZ__test__r r argparseArgumentParserparser add_argumentZFileTypestdout
parse_argsargsZtestr r Z
print_helpr) outputr r r$ Zpreambleformatr r r r r s
$ /
;
+
&
C