unpickle 0.0.1
Sign up to get free protection for your applications and to get access to all the features.
- data/README.md +37 -0
- data/Rakefile +8 -0
- data/lib/unpickle.rb +213 -0
- data/test/test_unpickle.rb +99 -0
- metadata +49 -0
data/README.md
ADDED
@@ -0,0 +1,37 @@
|
|
1
|
+
# unpickle
|
2
|
+
|
3
|
+
This is a very limited tool to unpickle python 'pickle' objects.
|
4
|
+
|
5
|
+
# Usage
|
6
|
+
|
7
|
+
require 'unpickle'
|
8
|
+
|
9
|
+
fh = File.open('some.pickle')
|
10
|
+
o = unpickle(fh.read)
|
11
|
+
|
12
|
+
unpickle() will raise Unpickle::UnpickleException if it doesn't support an
|
13
|
+
opcode in the picklestream, or encounters some kind of problem or invalid
|
14
|
+
sequence.
|
15
|
+
|
16
|
+
# Limitations
|
17
|
+
|
18
|
+
Currently unpickle only supports protocol 0.
|
19
|
+
|
20
|
+
It only supports Integers (not Longs), Strings (non-unicode),
|
21
|
+
booleans, dictionaries, lists, tuples and None.
|
22
|
+
|
23
|
+
Tuples will be returned as a frozen array.
|
24
|
+
|
25
|
+
None will be returned as nil.
|
26
|
+
|
27
|
+
# TODO
|
28
|
+
|
29
|
+
* Support more of protocol 0.
|
30
|
+
* Support newer protocols.
|
31
|
+
* Test object identity mapping.
|
32
|
+
* Package as a gem
|
33
|
+
* fix namespace.
|
34
|
+
|
35
|
+
# Author
|
36
|
+
|
37
|
+
Chris Collins <chris.collins@anchor.net.au>
|
data/Rakefile
ADDED
data/lib/unpickle.rb
ADDED
@@ -0,0 +1,213 @@
|
|
1
|
+
# unpickle.rb
|
2
|
+
#
|
3
|
+
# baby-unpickler to handle limited protocol 0 crap.
|
4
|
+
#
|
5
|
+
# vim:et sts=4 sw=4 ts=8:
|
6
|
+
#
|
7
|
+
module Unpickle
|
8
|
+
class UnpickleException < RuntimeError
|
9
|
+
end
|
10
|
+
|
11
|
+
class Mark
|
12
|
+
end
|
13
|
+
|
14
|
+
class PickleMachine
|
15
|
+
def initialize(input)
|
16
|
+
@stack = []
|
17
|
+
@memo = {}
|
18
|
+
@input = input
|
19
|
+
@idx = 0
|
20
|
+
end
|
21
|
+
|
22
|
+
def at_end?
|
23
|
+
@idx >= @input.length
|
24
|
+
end
|
25
|
+
|
26
|
+
def next_char
|
27
|
+
rv = @input[@idx..@idx]
|
28
|
+
@idx += 1
|
29
|
+
return rv
|
30
|
+
end
|
31
|
+
|
32
|
+
def peek_char
|
33
|
+
@input[@idx..@idx]
|
34
|
+
end
|
35
|
+
|
36
|
+
def read_int
|
37
|
+
strout = ""
|
38
|
+
while peek_char != "\n"
|
39
|
+
strout += next_char
|
40
|
+
end
|
41
|
+
next_char
|
42
|
+
case strout
|
43
|
+
when '00'
|
44
|
+
return false
|
45
|
+
when '01'
|
46
|
+
return true
|
47
|
+
else
|
48
|
+
return strout.to_i
|
49
|
+
end
|
50
|
+
end
|
51
|
+
|
52
|
+
def marker
|
53
|
+
idx = @stack.length-1
|
54
|
+
while idx >= 0
|
55
|
+
if @stack[idx].is_a?(Mark)
|
56
|
+
return idx
|
57
|
+
end
|
58
|
+
idx -= 1
|
59
|
+
end
|
60
|
+
raise UnpickleException, "Couldn't find Mark"
|
61
|
+
end
|
62
|
+
|
63
|
+
# read from the input stream to read the python string.
|
64
|
+
#
|
65
|
+
# returns the value.
|
66
|
+
def read_string
|
67
|
+
strout = ''
|
68
|
+
if next_char != '\''
|
69
|
+
raise UnpickleException, "Couldn't find leading quote for string"
|
70
|
+
end
|
71
|
+
while not at_end?
|
72
|
+
c = next_char
|
73
|
+
case c
|
74
|
+
when "\\"
|
75
|
+
opt = next_char
|
76
|
+
case opt
|
77
|
+
when 'x'
|
78
|
+
num = ''
|
79
|
+
while peek_char.match(/[\dA-Fa-f]/)
|
80
|
+
num += next_char
|
81
|
+
if num.length >= 2
|
82
|
+
break
|
83
|
+
end
|
84
|
+
end
|
85
|
+
unless (1..2).include?(num.length)
|
86
|
+
raise UnpickleException, "Bad \\x sequence in string"
|
87
|
+
end
|
88
|
+
strout += num.to_i(16).chr
|
89
|
+
when '0'
|
90
|
+
num = ''
|
91
|
+
while peek_char.match(/[0-7]/)
|
92
|
+
num += next_char
|
93
|
+
if num.length >= 3
|
94
|
+
break
|
95
|
+
end
|
96
|
+
end
|
97
|
+
unless (1..3).include?(num.length)
|
98
|
+
raise UnpickleException, "Bad \\0 sequence in string"
|
99
|
+
end
|
100
|
+
strout += num.to_i(8).chr
|
101
|
+
when 'n'
|
102
|
+
strout += "\n"
|
103
|
+
when "\\"
|
104
|
+
strout += "\\"
|
105
|
+
when 't'
|
106
|
+
strout += "\t"
|
107
|
+
when "'"
|
108
|
+
strout += "'"
|
109
|
+
else
|
110
|
+
raise UnpickleException, "Unexpected \\ escape: \\#{opt}"
|
111
|
+
end
|
112
|
+
when "'"
|
113
|
+
# valid end of string...
|
114
|
+
break
|
115
|
+
else
|
116
|
+
strout += c
|
117
|
+
end
|
118
|
+
end
|
119
|
+
if next_char != "\n"
|
120
|
+
raise UnpickleException, "Expected \\n after string"
|
121
|
+
end
|
122
|
+
return strout
|
123
|
+
end
|
124
|
+
|
125
|
+
def unpickle
|
126
|
+
while not at_end?
|
127
|
+
op = next_char
|
128
|
+
case op
|
129
|
+
when '(' # MARK
|
130
|
+
@stack.push(Mark.new)
|
131
|
+
when 'd' # DICT
|
132
|
+
newdict = {}
|
133
|
+
while true
|
134
|
+
if @stack.empty?
|
135
|
+
raise UnpickleException, "Stack empty during 'd'"
|
136
|
+
end
|
137
|
+
v = @stack.pop
|
138
|
+
if v.is_a?(Mark)
|
139
|
+
break
|
140
|
+
end
|
141
|
+
if @stack.empty?
|
142
|
+
raise UnpickleException, "Stack empty during 'd'"
|
143
|
+
end
|
144
|
+
k = stack.pop
|
145
|
+
if k.is_a?(Mark)
|
146
|
+
raise UnpickleException, "Odd number of elements during 'd' stack walk"
|
147
|
+
end
|
148
|
+
newdict[k] = v
|
149
|
+
end
|
150
|
+
@stack.push(newdict)
|
151
|
+
when 'S' # STRING
|
152
|
+
newstr = read_string
|
153
|
+
@stack.push(newstr)
|
154
|
+
when 'p' # PUT (string)
|
155
|
+
index = read_int
|
156
|
+
@memo[index] = @stack[-1]
|
157
|
+
when 'g' # GET (string)
|
158
|
+
index = read_int
|
159
|
+
@stack.push(@memo[index])
|
160
|
+
when 'I' # INT
|
161
|
+
intarg = read_int
|
162
|
+
@stack.push(intarg)
|
163
|
+
when 's' # SETITEM
|
164
|
+
value = @stack.pop
|
165
|
+
key = @stack.pop
|
166
|
+
dict = @stack[-1]
|
167
|
+
dict[key] = value
|
168
|
+
when 't' # TUPLE
|
169
|
+
midx = marker
|
170
|
+
tuple = @stack[midx+1..-1]
|
171
|
+
@stack = @stack[0...midx]
|
172
|
+
tuple.freeze
|
173
|
+
@stack.push(tuple)
|
174
|
+
when 'l' # LIST
|
175
|
+
midx = marker
|
176
|
+
list = @stack[midx+1..-1]
|
177
|
+
@stack = @stack[0...midx]
|
178
|
+
@stack.push(list)
|
179
|
+
when 'N' # NONE
|
180
|
+
@stack.push(nil)
|
181
|
+
when 'a' # APPEND
|
182
|
+
e = @stack.pop
|
183
|
+
@stack[-1].push(e)
|
184
|
+
when '0' # POP
|
185
|
+
@stack.pop
|
186
|
+
when '2' # DUP
|
187
|
+
@stack.push(@stack[-1])
|
188
|
+
when '.' # STOP
|
189
|
+
return @stack.pop
|
190
|
+
else
|
191
|
+
raise UnpickleException, "Unsupported unpickle operation '#{op}'"
|
192
|
+
end
|
193
|
+
end
|
194
|
+
raise UnpickleException, "Hit end of input stream"
|
195
|
+
end
|
196
|
+
end
|
197
|
+
|
198
|
+
# Unpickle the python object pickled into str.
|
199
|
+
#
|
200
|
+
# At this time, this ONLY works with a limited set of constructs
|
201
|
+
# (dicts, lists, tuples, strings, ints, bools, None) and only with
|
202
|
+
# protocol 0.
|
203
|
+
#
|
204
|
+
# Object uniqueness should obey the python semantics but is largely
|
205
|
+
# untested.
|
206
|
+
#
|
207
|
+
# Raises an UnpickleException if anything goes wrong.
|
208
|
+
def Unpickle.loads(str)
|
209
|
+
p = Unpickle::PickleMachine.new(str)
|
210
|
+
return p.unpickle
|
211
|
+
end
|
212
|
+
end
|
213
|
+
|
@@ -0,0 +1,99 @@
|
|
1
|
+
# tests for Unpickle.loads.rb
|
2
|
+
#
|
3
|
+
# vim:et sts=4 sw=4 ts=8:
|
4
|
+
require 'test/unit'
|
5
|
+
require 'unpickle'
|
6
|
+
|
7
|
+
class UnpickleTests < Test::Unit::TestCase
|
8
|
+
# all pickle_str's were generated using python 2.5
|
9
|
+
# and pickle.dumps(...., 0)
|
10
|
+
|
11
|
+
def test_simple_string
|
12
|
+
# >>> pickle_obj = 'abcdefg'
|
13
|
+
pickle_str = "S'abcdefg'\np0\n."
|
14
|
+
|
15
|
+
o = Unpickle.loads(pickle_str)
|
16
|
+
|
17
|
+
assert_equal('abcdefg', o)
|
18
|
+
end
|
19
|
+
|
20
|
+
def test_simple_bool_true
|
21
|
+
# >>> pickle_obj = True
|
22
|
+
pickle_str = "I01\n."
|
23
|
+
|
24
|
+
o = Unpickle.loads(pickle_str)
|
25
|
+
|
26
|
+
assert_equal(true, o)
|
27
|
+
end
|
28
|
+
|
29
|
+
def test_simple_bool_false
|
30
|
+
# >>> pickle_obj = False
|
31
|
+
pickle_str = "I00\n."
|
32
|
+
|
33
|
+
o = Unpickle.loads(pickle_str)
|
34
|
+
|
35
|
+
assert_equal(false, o)
|
36
|
+
end
|
37
|
+
|
38
|
+
def test_simple_list
|
39
|
+
# >>> pickle_obj = [1,2,3]
|
40
|
+
pickle_str = "(lp0\nI1\naI2\naI3\na."
|
41
|
+
|
42
|
+
o = Unpickle.loads(pickle_str)
|
43
|
+
|
44
|
+
assert_equal([1,2,3], o)
|
45
|
+
end
|
46
|
+
|
47
|
+
def test_simple_dict
|
48
|
+
# >>> pickle_obj = {'a': 1, 'b': 2, 'c': 3}
|
49
|
+
pickle_str = "(dp0\nS'a'\np1\nI1\nsS'c'\np2\nI3\nsS'b'\np3\nI2\ns."
|
50
|
+
|
51
|
+
o = Unpickle.loads(pickle_str)
|
52
|
+
|
53
|
+
assert_equal({'a' => 1, 'b' => 2, 'c' => 3}, o)
|
54
|
+
end
|
55
|
+
|
56
|
+
def test_simple_int_zero
|
57
|
+
# >>> pickle_obj = 0
|
58
|
+
pickle_str = "I0\n."
|
59
|
+
|
60
|
+
o = Unpickle.loads(pickle_str)
|
61
|
+
|
62
|
+
assert_equal(0, o)
|
63
|
+
end
|
64
|
+
|
65
|
+
def test_none
|
66
|
+
# >>> pickle_obj = None
|
67
|
+
pickle_str = 'N.'
|
68
|
+
|
69
|
+
o = Unpickle.loads(pickle_str)
|
70
|
+
|
71
|
+
assert_equal(nil, o)
|
72
|
+
end
|
73
|
+
|
74
|
+
def test_mixed
|
75
|
+
# >>> pickle_obj = {'a': [1,2,3,4], 'b': (1,2,3), 'c': None, 'd': 'abcd'}
|
76
|
+
pickle_str = "(dp0\nS'a'\np1\n(lp2\nI1\naI2\naI3\naI4\nasS'c'\np3\nNsS'b'\np4\n(I1\nI2\nI3\ntp5\nsS'd'\np6\nS'abcd'\np7\ns."
|
77
|
+
|
78
|
+
o = Unpickle.loads(pickle_str)
|
79
|
+
|
80
|
+
assert_equal({'a' => [1,2,3,4], 'b' => [1,2,3], 'c' => nil, 'd' => 'abcd'}, o)
|
81
|
+
end
|
82
|
+
|
83
|
+
def test_recursive
|
84
|
+
# >>> aobj = {}
|
85
|
+
# >>> bobj = {'a': aobj}
|
86
|
+
# >>> aobj['b'] = bobj
|
87
|
+
# >>> pickle.dumps(aobj, 0)
|
88
|
+
pickle_str = "(dp0\nS'b'\np1\n(dp2\nS'a'\np3\ng0\nss."
|
89
|
+
|
90
|
+
o = Unpickle.loads(pickle_str)
|
91
|
+
|
92
|
+
assert(o.is_a?(Hash))
|
93
|
+
assert(o.include?('b'))
|
94
|
+
assert(o['b'].is_a?(Hash))
|
95
|
+
assert(o['b'].include?('a'))
|
96
|
+
# check that the object identity is correctly respected.
|
97
|
+
assert(o.object_id == o['b']['a'].object_id)
|
98
|
+
end
|
99
|
+
end
|
metadata
ADDED
@@ -0,0 +1,49 @@
|
|
1
|
+
--- !ruby/object:Gem::Specification
|
2
|
+
name: unpickle
|
3
|
+
version: !ruby/object:Gem::Version
|
4
|
+
version: 0.0.1
|
5
|
+
prerelease:
|
6
|
+
platform: ruby
|
7
|
+
authors:
|
8
|
+
- Chris Collins
|
9
|
+
autorequire:
|
10
|
+
bindir: bin
|
11
|
+
cert_chain: []
|
12
|
+
date: 2012-06-01 00:00:00.000000000 Z
|
13
|
+
dependencies: []
|
14
|
+
description: A library to unpickle simple python objects directly into ruby
|
15
|
+
email: kuroneko-rubygems@sysadninjas.net
|
16
|
+
executables: []
|
17
|
+
extensions: []
|
18
|
+
extra_rdoc_files: []
|
19
|
+
files:
|
20
|
+
- lib/unpickle.rb
|
21
|
+
- README.md
|
22
|
+
- Rakefile
|
23
|
+
- test/test_unpickle.rb
|
24
|
+
homepage: http://github.com/kuroneko/unpickle-rb
|
25
|
+
licenses: []
|
26
|
+
post_install_message:
|
27
|
+
rdoc_options: []
|
28
|
+
require_paths:
|
29
|
+
- lib
|
30
|
+
required_ruby_version: !ruby/object:Gem::Requirement
|
31
|
+
none: false
|
32
|
+
requirements:
|
33
|
+
- - ! '>='
|
34
|
+
- !ruby/object:Gem::Version
|
35
|
+
version: '0'
|
36
|
+
required_rubygems_version: !ruby/object:Gem::Requirement
|
37
|
+
none: false
|
38
|
+
requirements:
|
39
|
+
- - ! '>='
|
40
|
+
- !ruby/object:Gem::Version
|
41
|
+
version: '0'
|
42
|
+
requirements: []
|
43
|
+
rubyforge_project:
|
44
|
+
rubygems_version: 1.8.15
|
45
|
+
signing_key:
|
46
|
+
specification_version: 3
|
47
|
+
summary: unpickle python objects in ruby
|
48
|
+
test_files:
|
49
|
+
- test/test_unpickle.rb
|