3 \fI@s dZddlZddlZddlZddlZddlZdddgZejZdZdZ dZ dZ dZ Gd d d e Zdd lmZddZeddeddZddZeddeddZddZedd eddZddZedd eddZdd Zed!d"ed#dZdd%d&Zed'eed(dZd)d*Zed+eed,dZd-d.Z ed/ee d0dZ!d1d2Z"ed3e e"d4dZ#d5d6Z$ed7e e$d8dZ%d9d:Z&ed;e e&ddZ'd?d@Z(edAe e(dBdZ)dCdDZ*edEe e*dFdZ+dGdHZ,edIee,dJdZ-dKdLZ.edMe e.dNdZ/dOdPZ0edQe e0dRdZ1dSdTZ2edUe e2dVdZ3dWdXZ4dYdZZ5ed[ee4d\dZ6ed]ee5d^dZ7d_d`Z8edaee8dbdZ9dcddZ:eded"e:dfdZ;ddglmdldmZ?edne e?dodZ@Gdpdqdqe ZAeAdreBdsdtZCZDeAdueBeEfdvdtZFeAdweEdxdtZGeAdyeHdzdtZIeAd{eJeKfd|dtZLZMeAd}eJd~dtZNeAdeKddtZOeAdePdddtZQeAdeRddtZSeAdeTddtZUeAdeVddtZWeAdeXddtZYeAdeXddtZZeAde ddtZ[eAdeAddtZ\eAdeAddtZ]Gddde Z^e^Z_e_dde6geFgddde_ddegeCgddde_ddegeCgddde_ddegeCgddde_dde7geCgddde_dde>geCgddde_dde@geCgddde_ddegeLgddde_dde%geLgddde_dde#geLgddde_dde)geNgddde_dde'geNgddde_dde+geNgd dde_dddgeQgddde_dddgeGgddde_dddgeGgddde_dde-geOgddde_dde/geOgd dde_dde1geOgddde_dde3geOgd dde_dde9geIgddde_dde;geIgddde_dddgeUgddde_dddeUe[geUgddde_dddeUe\e]geUgddde_ddde\e]geUgddde_dddgeSgddde_ddde\e]geSgddde_ddde[geSgddde_ddde[e[geSgddde_ddde[e[e[geSgddde_dddgeWgddde_ddde\e]geWgddde_dddeWe[e[geWgddde_dddeWe\e]geWgddde_dddgeYgd dde_dddeYe\e]geYgd dde_ddde\e]geZgd dde_dd de[ggdd de_d d de[ge[e[gdd de_dddge\gddde_ddde\e]ggddde_dde6ge[gddde_ddege[gddde_ddege[gddde_dde6ggddde_d d!eggdd"de_d#d$eggdd%de_d&d'de[ge[gd d(de_d)d*ege[gdd+de_d,d-ege[gdd.de_d/d0ege[gdd1de_d2d3e!ge[gdd4de_d5d6deOeOge[gd d7de_d8d9de[e[ge[gdd:de_d;dd?e!e\e]ge[gdd@de_dAdBde\e[e]ge[gddCde_dDdEde[e[ge[gddFde_dGdHde[e[e[ge[gd dIde_dJdKeggddLde_dMdNde[ggddOde_dPdQeggd dRde_dSdTege[gddUde_dVdWde[ge[gddXdgAZ`[_iZaiZbxzece`D]n\ZdZeeejfeak regdYeejfeaeejfedfeejhebk regdZeejhebeejhedfedeaeejf<edebeejh< qW[a[b[d[eiZixe`D]Zeeeeieejh< qW[edd\d]Zjej[jdd^d_Zkd`dZldadZmddbdZnGdcddddZodeZpdfZqepeqdgZrdhdiZsetdjk rddluZueujvdkdlZwewjxdmeujydndodpdqewjxdrdsejzeujydtdudvewjxdwdxdydzd{ewjxd|d}d eBd~dvewjxdddydd{ewjxdddddewjxdddydd{ewjxddydd{ewj{Z|e|j} resne|j~ rdndZ~e|j sewjnee|jdk r*ene|jde|jde|je~nZe|j r6indZxHe|jD]>Ze|jjejfdZe|jjedenee|jee|je~ qBWdS(ar"Executable documentation" for the pickle module. Extensive comments about the pickle protocols and pickle-machine opcodes can be found here. Some functions meant for external use: genops(pickle) Generate all the opcodes in a pickle, as (opcode, arg, position) triples. dis(pickle, out=None, memo=None, indentlevel=4) Print a symbolic disassembly of a pickle. Ndisgenopsoptimizec@seZdZdZddZdS) ArgumentDescriptornamenreaderdoccCs`t|tst||_t|tr8|dks<|ttttt fks>> import io >>> read_uint1(io.BytesIO(b'\xff')) 255 rrz'not enough data in stream to read uint1N)read ValueError)fdatarrr read_uint1s r%uint1zOne-byte unsigned integer.)r r r rcCs0|jd}t|dkr$td|dStddS)z >>> import io >>> read_uint2(io.BytesIO(b'\xff\x00')) 255 >>> read_uint2(io.BytesIO(b'\xff\xff')) 65535 rz>> import io >>> read_int4(io.BytesIO(b'\xff\x00\x00\x00')) 255 >>> read_int4(io.BytesIO(b'\x00\x00\x00\x80')) == -(2**31) True rz>> import io >>> read_uint4(io.BytesIO(b'\xff\x00\x00\x00')) 255 >>> read_uint4(io.BytesIO(b'\x00\x00\x00\x80')) == 2**31 True rz>> 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 %rzno string quotes around %rrascii)r3r4r6)readlineendswithr" startswithcodecs escape_decodedecode)r#r< stripquotesr$qrrr read_stringnl;s         r?stringnlzA newline-terminated string. This is a repr-style string, with embedded escapes, and bracketing quotes. cCs t|ddS)NF)r=)r?)r#rrrread_stringnl_noescapetsrAstringnl_noescapeaA newline-terminated string. This is a str-style string, without embedded escapes, or bracketing quotes. It should consist solely of printable ASCII characters. cCsdt|t|fS)zp >>> import io >>> read_stringnl_noescape_pair(io.BytesIO(b"Queue\nEmpty\njunk")) 'Queue Empty' z%s %s)rA)r#rrrread_stringnl_noescape_pairsrCstringnl_noescape_pairaA pair of newline-terminated strings. These are str-style strings, without embedded escapes, or bracketing quotes. They should consist solely of printable ASCII characters. The pair is returned as a single string, with a single blank separating the two strings. cCsLt|}|dkst|j|}t||kr4|jdStd|t|fdS)z >>> import io >>> read_string1(io.BytesIO(b"\x00")) '' >>> read_string1(io.BytesIO(b"\x03abcdef")) 'abc' rzlatin-1z2expected %d bytes in a string1, but only %d remainN)r%rr!r'r<r")r#r r$rrr read_string1s     rEstring1zA 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. cCsTt|}|dkrtd||j|}t||kr<|jdStd|t|fdS)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 rzstring4 byte count < 0: %dzlatin-1z2expected %d bytes in a string4, but only %d remainN)r+r"r!r'r<)r#r r$rrr read_string4s     rGstring4zA counted string. The first argument is a 4-byte little-endian signed int giving the number of bytes in the string, and the second argument is that many bytes. cCsFt|}|dkst|j|}t||kr.|Std|t|fdS)z >>> import io >>> read_bytes1(io.BytesIO(b"\x00")) b'' >>> read_bytes1(io.BytesIO(b"\x03abcdef")) b'abc' rz1expected %d bytes in a bytes1, but only %d remainN)r%rr!r'r")r#r r$rrr read_bytes1s    rIbytes1zA counted bytes 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. cCsFt|}|dkst|j|}t||kr.|Std|t|fdS)z >>> import io >>> read_bytes1(io.BytesIO(b"\x00")) b'' >>> read_bytes1(io.BytesIO(b"\x03abcdef")) b'abc' rz1expected %d bytes in a bytes1, but only %d remainN)r%rr!r'r")r#r r$rrrrIs    zA counted bytes string. The first argument is a 1-byte unsigned int giving the number of bytes, and the second argument is that many bytes. cCs\t|}|dkst|tjkr*td||j|}t||krD|Std|t|fdS)aN >>> import io >>> read_bytes4(io.BytesIO(b"\x00\x00\x00\x00abc")) b'' >>> read_bytes4(io.BytesIO(b"\x03\x00\x00\x00abcdef")) b'abc' >>> read_bytes4(io.BytesIO(b"\x00\x00\x00\x03abcdef")) Traceback (most recent call last): ... ValueError: expected 50331648 bytes in a bytes4, but only 6 remain rz#bytes4 byte count > sys.maxsize: %dz1expected %d bytes in a bytes4, but only %d remainN)r-rsysmaxsizer"r!r')r#r r$rrr read_bytes4s      rMbytes4zA 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. cCs\t|}|dkst|tjkr*td||j|}t||krD|Std|t|fdS)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 rz#bytes8 byte count > sys.maxsize: %dz1expected %d bytes in a bytes8, but only %d remainN)r0rrKrLr"r!r')r#r r$rrr read_bytes83s     rObytes8zA counted bytes string. The first argument is an 8-byte little-endian unsigned int giving the number of bytes, and the second argument is that many bytes. cCs0|j}|jdstd|dd}t|dS)zm >>> import io >>> read_unicodestringnl(io.BytesIO(b"abc\\uabcd\njunk")) == 'abc\uabcd' True r2z4no newline found when trying to read unicodestringnlNrzraw-unicode-escaper6)r7r8r"r)r#r$rrrread_unicodestringnlUs   rQunicodestringnlzA newline-terminated Unicode string. This is raw-unicode-escape encoded, so consists of printable ASCII characters, and may contain embedded escape sequences. cCsNt|}|dkst|j|}t||kr6t|ddStd|t|fdS)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 rzutf-8 surrogatepassz9expected %d bytes in a unicodestring1, but only %d remainN)r%rr!r'rr")r#r r$rrrread_unicodestring1os    rTunicodestring1aAA counted Unicode string. The first argument is a 1-byte little-endian signed int giving the number of bytes in the string, and the second argument-- the UTF-8 encoding of the Unicode string -- contains that many bytes. cCsdt|}|dkst|tjkr*td||j|}t||krLt|ddStd|t|fdS)a >>> import io >>> s = 'abcd\uabcd' >>> enc = s.encode('utf-8') >>> enc b'abcd\xea\xaf\x8d' >>> n = bytes([len(enc), 0, 0, 0]) # little-endian 4-byte length >>> t = read_unicodestring4(io.BytesIO(n + enc + b'junk')) >>> s == t True >>> read_unicodestring4(io.BytesIO(n + enc[:-1])) Traceback (most recent call last): ... ValueError: expected 7 bytes in a unicodestring4, but only 6 remain rz+unicodestring4 byte count > sys.maxsize: %dzutf-8rSz9expected %d bytes in a unicodestring4, but only %d remainN)r-rrKrLr"r!r'r)r#r r$rrrread_unicodestring4s      rVunicodestring4aAA 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. cCsdt|}|dkst|tjkr*td||j|}t||krLt|ddStd|t|fdS)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 rz+unicodestring8 byte count > sys.maxsize: %dzutf-8rSz9expected %d bytes in a unicodestring8, but only %d remainN)r0rrKrLr"r!r'r)r#r r$rrrread_unicodestring8s      rXunicodestring8aBA counted Unicode string. The first argument is an 8-byte little-endian signed int giving the number of bytes in the string, and the second argument-- the UTF-8 encoding of the Unicode string -- contains that many bytes. cCs.t|ddd}|dkrdS|dkr&dSt|S)z >>> import io >>> read_decimalnl_short(io.BytesIO(b"1234\n56")) 1234 >>> read_decimalnl_short(io.BytesIO(b"1234L\n56")) Traceback (most recent call last): ... ValueError: invalid literal for int() with base 10: b'1234L' F)r<r=s00s01T)r?r)r#srrrread_decimalnl_shorts r[cCs2t|ddd}|dddkr*|dd}t|S)z >>> import io >>> read_decimalnl_long(io.BytesIO(b"1234L\n56")) 1234 >>> read_decimalnl_long(io.BytesIO(b"123456789012345678901234L\n6")) 123456789012345678901234 F)r<r=rNLr6r6)r?r)r#rZrrrread_decimalnl_longs  r]decimalnl_shortaA newline-terminated decimal integer literal. This never has a trailing 'L', and the integer fit in a short Python int on the box where the pickle was written -- but there's no guarantee it will fit in a short Python int on the box where the pickle is read. decimalnl_longzA newline-terminated decimal integer literal. This has a trailing 'L', and can represent integers of any size. cCst|ddd}t|S)zO >>> import io >>> read_floatnl(io.BytesIO(b"-1.25\n6")) -1.25 F)r<r=)r?float)r#rZrrr read_floatnl'srafloatnlaA newline-terminated decimal floating literal. In general this requires 17 significant digits for roundtrip identity, and pickling then unpickling infinities, NaNs, and minus zero doesn't work across boxes, or on some boxes even on itself (e.g., Windows can't read the strings it produces for infinities or NaNs). cCs0|jd}t|dkr$td|dStddS)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 r/z>drz(not enough data in stream to read float8N)r!r'r(r")r#r$rrr read_float8=s  rcfloat8aAn 8-byte binary representation of a float, big-endian. The format is unique to Python, and shared with the struct module (format string '>d') "in theory" (the struct and pickle implementations don't share the code -- they should). It's strongly related to the IEEE-754 double format, and, in normal cases, is in fact identical to the big-endian 754 double format. On other boxes the dynamic range is limited to that of a 754 double, and "add a half and chop" rounding is used to reduce the precision to 53 bits. However, even on a 754 box, infinities, NaNs, and minus zero may not be handled correctly (may not survive roundtrip pickling intact). ) decode_longcCs.t|}|j|}t||kr&tdt|S)a+ >>> import io >>> read_long1(io.BytesIO(b"\x00")) 0 >>> read_long1(io.BytesIO(b"\x02\xff\x00")) 255 >>> read_long1(io.BytesIO(b"\x02\xff\x7f")) 32767 >>> read_long1(io.BytesIO(b"\x02\x00\xff")) -256 >>> read_long1(io.BytesIO(b"\x02\x00\x80")) -32768 z'not enough data in stream to read long1)r%r!r'r"re)r#r r$rrr read_long1cs   rflong1aA 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. cCsBt|}|dkrtd||j|}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 rzlong4 byte count < 0: %dz'not enough data in stream to read long4)r+r"r!r're)r#r r$rrr read_long4s   rhlong4aA 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 eZdZd ZddZddZdS) StackObjectr obtypercCspt|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. ZLONG1Šz|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.STRINGSaPush 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 BINSTRINGTaPush 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_BINSTRINGUaPush 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. ZBINBYTESBzPush 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 BINBYTES8ŽzPush 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.ZNEWTRUEˆzPush 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_BINUNICODEŒaPush 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 BINUNICODEXaPush 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aPush 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. ZFLOATFaNewline-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. ZBINFLOATGaFloat 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. ZLISTlasBuild 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.ZTUPLEtavBuild 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†aBuild 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‡aBuild 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.ZDICTdaBuild 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'} ZSETITEMrZzAdd 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‘azBuild 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}) POP0zzStore the stack top into the memo. The stack is not popped. The index of the memo location to write into is given by the 1-byte unsigned integer following. LONG_BINPUTrzStore the stack top into the memo. The stack is not popped. The index of the memo location to write into is given by the 4-byte unsigned little-endian integer following. MEMOIZE”zStore the stack top into the memo. The stack is not popped. The index of the memo location to write is the number of elements currently present in the memo. ZEXT1‚aExtension code. This code and the similar EXT2 and EXT4 allow using a registry of popular objects that are pickled by name, typically classes. It is envisioned that through a global negotiation and registration process, third parties can set up a mapping between ints and object names. In order to guarantee pickle interchangeability, the extension code registry ought to be global, although a range of codes may be reserved for private use. EXT1 has a 1-byte integer argument. This is used to index into the extension registry, and the object at that index is pushed on the stack. ZEXT2ƒzNExtension code. See EXT1. EXT2 has a two-byte integer argument. ZEXT4„zOExtension code. See EXT1. EXT4 has a four-byte integer argument. ZGLOBALcaPush a global object (module.attr) on the stack. Two newline-terminated strings follow the GLOBAL opcode. The first is taken as a module name, and the second as a class name. The class object module.class is pushed on the stack. More accurately, the object returned by self.find_class(module, class) is pushed on the stack, so unpickling subclasses can override this form of lookup. Z STACK_GLOBAL“z7Push a global object (module.attr) on the stack. ZREDUCERaLPush an object built from a callable and an argument tuple. The opcode is named to remind of the __reduce__() method. Stack before: ... callable pytuple Stack after: ... callable(*pytuple) The callable and the argument tuple are the first two items returned by a __reduce__ method. Applying the callable to the argtuple is supposed to reproduce the original object, or at least get it started. If the __reduce__ method returns a 3-tuple, the last component is an argument to be passed to the object's __setstate__, and then the REDUCE opcode is followed by code to create setstate's argument, and then a BUILD opcode to apply __setstate__ to that argument. If not isinstance(callable, type), REDUCE complains unless the callable has been registered with the copyreg module's safe_constructors dict, or the callable has a magic '__safe_for_unpickling__' attribute with a true value. I'm not sure why it does this, but I've sure seen this complaint often enough when I didn't want to . ZBUILDbaFinish 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) ZINSTiaqBuild 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. ZOBJoaBuild 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aLBuild 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’auBuild 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. ZPERSIDPaPush 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 BINPERSIDQaXPush 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 %dFcCstj}xtjD]}tjd|s2|rtd|qtt|}t|t  sTt |dkrj|rtd||fq|j d}||kr|rtd||f||}|j |krt d|||j f||=qt d||fqW|rd g}x(|jD]\}}|jd |j |fqWt d j|dS) Nz[A-Z][A-Z0-9_]+$z0skipping %r: it doesn't look like an opcode namerz5skipping %r: value %r doesn't look like a pickle codezlatin-1z+checking name %r w/ code %r for consistencyzBfor pickle code %r, pickle.py uses name %r but we're using name %rzPpickle.py appears to have a pickle opcode with name %r and code %r, but we don'tz=we appear to have pickle opcodes that pickle.py doesn't have:z name %r with code %r )code2opcopyr~__all__rematchprintgetattrrrpr'r<r r"itemsappendjoin)verboserr Z picklecodermsgryrrrassure_pickle_consistencys>         rccst|trtj|}t|dr&|j}ndd}x|}|jd}tj|j d}|dkr|dkrjt dnt d|dkrzd n||f|j dkrd}n |j j |}|r||||fVn |||fV|d kr0|j d kstPq0WdS) NtellcSsdS)Nrrrrrsz_genops..rzlatin-1z#pickle exhausted before seeing STOPz!at position %s, opcode %r unknownz .r)r bytes_typesioBytesIOhasattrrr!rgetr<r"rzr r r)r$ yield_end_posZgetposposryopcoderzrrr_genopss0         rcCst|S)axGenerate 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~rrrrscCsd}d}t}i}g}d}d}xt|ddD]\}} } } d|jkr\|j| |j|| fq,|jdkrt|} |j| |j|| fq,d|jkrq,d|jkr|j|kr|j}d || <|j|| fq,|jd kr| |kr| }| dkr|| | }n|j| | fq,|j| | fq,W~tj} | j |t j | |}|d krJ|j j d} x|D]x\}} ||kr| |krtqT|j| }| || <| d 7} n&||kr|j|| }n ||| }|j j|j |qTW|j j| jS) z7Optimize a pickle string by removing unused PUT opcodesrrrrT)rrrNrrr)rtrr addrr'r}rrwriter~Z_PicklerZframerZ start_framingputrZ commit_frameZ end_framinggetvalue)rrrZoldidsZnewidsopcodesr}Z protoheaderrrzrZend_posidxoutZpickleropr$rrrrsd                    cCstg}|dkri}d}g}d|}d} |} x$t|D]\} } } | dk rZtd| d|ddt| jdd |t|| jf}t|| j}| j}| j }t|}d}t |ks| jdko|o|d!t kr|t |kst t |kr|d"t kst |rt|j }|dkr d}nd |}x|d#t k r.|j qW|j y|jt }Wn*tk rp| jdksht d }YnXnd } }| jd$kr| jdkrt|}d|}n| dk st | }||krd| } n,|sd} n |d%t krd} n |d&||<n<| jd'kr6| |kr.t|dks"t || g}nd| } | dk sF|r|ddt| j7}| dk rv|dt| 7}|r|d|7}|r|d| t|7}t|} | dkr|} |d| jjddd 7}t||d| rt| t||krtd|t|f|r || d=t |krBt |ks8t |j| |j|q4Wtd||d|rptd|dS)(aKProduce a symbolic disassembly of a pickle. 'pickle' is a file-like object, or string, containing a (at least one) pickle. The pickle is disassembled from the current position, through the first STOP opcode encountered. Optional arg 'out' is a file-like object to which the disassembly is printed. It defaults to sys.stdout. Optional arg 'memo' is a Python dict, used as the pickle's memo. It may be mutated by dis(), if the pickle contains PUT or BINPUT opcodes. Passing the same memo object to another dis() call then allows disassembly to proceed across multiple pickles that were all created by the same pickler with the same memo. Ordinarily you don't need to worry about this. Optional arg 'indentlevel' is the number of blanks by which to indent a new MARK level. It defaults to 4. Optional arg 'annotate' if nonzero instructs dis() to add short description of the opcode on each line of disassembled output. The value given to 'annotate' must be an integer and is used as a hint for the column where annotation should start. The default value is 0, meaning no annotations. In addition to printing the disassembly, some sanity checks are made: + All embedded opcode arguments "make sense". + Explicit and implicit pop operations have enough items on the stack. + When an opcode implicitly refers to a markobject, a markobject is actually on the stack. + A memo entry isn't referenced before it's defined. + The markobject isn't stored in the memo. + A memo entry isn't redefined. Nr z%5d:)endfilez %-4s %s%srz(MARK at unknown opcode offset)z (MARK at %d)rzno MARK exists on stackrrrrz(as %d)zmemo key %r already definedz'stack is empty -- can't store into memoz"can't store markobject in the memorrrz&memo key %r has never been stored into 2r)rz3tries to pop %d items from stack with only %d itemsz highest protocol among opcodes =zstack not empty after STOP: %rr6r6r6r6r6)rrrrr6r6)rrr)rrreprryr'r maxr}r{r| markobjectrrwpopindexr"rsplitrextend)r~rmemo indentlevelannotatestackZmaxprotoZ markstackZ indentchunkZerrormsgZannocolrrzrlineZbeforeafterZnumtopopZmarkmsgZmarkposZmemo_idxrrrr% s-                     c@seZdZddZdS)_ExamplecCs ||_dS)N)value)rrrrrr sz_Example.__init__N)rrrrrrrrr sra >>> 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: L LONG 1 9: a APPEND 10: L LONG 2 14: a APPEND 15: ( MARK 16: L LONG 3 20: L LONG 4 24: t TUPLE (MARK at 15) 25: p PUT 1 28: a APPEND 29: ( MARK 30: d DICT (MARK at 29) 31: p PUT 2 34: c GLOBAL '_codecs encode' 50: p PUT 3 53: ( MARK 54: V UNICODE 'abc' 59: p PUT 4 62: V UNICODE 'latin1' 70: p PUT 5 73: t TUPLE (MARK at 53) 74: p PUT 6 77: R REDUCE 78: p PUT 7 81: V UNICODE 'def' 86: p PUT 8 89: s SETITEM 90: a APPEND 91: . 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: L LONG 42 111: s SETITEM 112: b BUILD 113: a APPEND 114: g GET 5 117: a APPEND 118: . 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_testcCsddl}|jS)Nr)doctestZtestmod)rrrr_test sr__main__z$disassemble one or more pickle files)Z description pickle_filebr*zthe pickle file)rlnargshelpz-oz--outputwz+the file where the output should be written)defaultrlr 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 rr6)TT)F)F)NNrr)__doc__r:rr~rrKrrrrrrrobjectr Zstructr r(r%r&r)r*r+r,r-r.r0r1r?r@rArBrCrDrErFrGrHrIrJrMrNrOrPrQrRrTrUrVrWrXrYr[r]r^r_rarbrcrdrerfrgrhrirjrZpyintZpylongroZpyinteger_or_boolZpyboolr`ZpyfloatrprZpybytes_or_strZpystringZpybytesZ pyunicoderlZpynonermZpytuplerrZpylistrsZpydictrtZpysetZ pyfrozensetZ anyobjectrrwrxrrZname2iZcode2i enumeraterrr r"ryrrrrrrrZ _dis_testZ _memo_testZ__test__rrargparseArgumentParserparser add_argumentZFileTypestdout parse_argsargsZtestrrZ print_helpr'outputrrr#Zpreambleformatrrrrr s $   /        ;           +      &  >