1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
|
import struct
import zlib
import re
class GzipDecompressor(object):
"""An interface to gzip which is similar to bz2.BZ2Decompressor and
lzma.LZMADecompressor."""
def __init__(self):
self.sawheader = False
self.inbuffer = b""
self.decompressor = None
self.crc = 0
self.size = 0
def decompress(self, data):
"""
@raises ValueError: if no gzip magic is found
@raises zlib.error: from zlib invocations
"""
while True:
if self.decompressor:
data = self.decompressor.decompress(data)
self.crc = zlib.crc32(data, self.crc)
self.size += len(data)
unused_data = self.decompressor.unused_data
if not unused_data:
return data
self.decompressor = None
return data + self.decompress(unused_data)
self.inbuffer += data
skip = 10
if len(self.inbuffer) < skip:
return b""
if not self.inbuffer.startswith(b"\037\213\010"):
raise ValueError("gzip magic not found")
flag = ord(self.inbuffer[3])
if flag & 4:
if len(self.inbuffer) < skip + 2:
return b""
length, = struct.unpack("<H", self.inbuffer[skip:skip+2])
skip += 2 + length
for field in (8, 16):
if flag & field:
length = self.inbuffer.find(b"\0", skip)
if length < 0:
return b""
skip = length + 1
if flag & 2:
skip += 2
if len(self.inbuffer) < skip:
return b""
data = self.inbuffer[skip:]
self.inbuffer = b""
self.sawheader = True
self.decompressor = zlib.decompressobj(-zlib.MAX_WBITS)
@property
def unused_data(self):
if self.decompressor:
return self.decompressor.unused_data
elif not self.sawheader:
return self.inbuffer
else:
expect = struct.pack("<ll", self.crc, self.size)
if self.inbuffer.startswith(expect) and \
self.inbuffer[len(expect):].replace("\0", "") == "":
return b""
return self.inbuffer
def flush(self):
"""
@raises zlib.error: from zlib invocations
"""
if not self.decompressor:
return b""
return self.decompressor.flush()
def copy(self):
new = GzipDecompressor()
new.inbuffer = self.inbuffer
if self.decompressor:
new.decompressor = self.decompressor.copy()
new.sawheader = self.sawheader
new.crc = self.crc
new.size = self.size
return new
class DecompressedStream(object):
"""Turn a readable file-like into a decompressed file-like. Te only part
of being file-like consists of the read(size) method in both cases."""
blocksize = 65536
def __init__(self, fileobj, decompressor):
"""
@param fileobj: a file-like object providing read(size)
@param decompressor: a bz2.BZ2Decompressor or lzma.LZMADecompressor
like object providing methods decompress and flush and an
attribute unused_data
"""
self.fileobj = fileobj
self.decompressor = decompressor
self.buff = b""
def read(self, length=None):
data = True
while True:
if length is not None and len(self.buff) >= length:
ret = self.buff[:length]
self.buff = self.buff[length:]
return ret
elif not data: # read EOF in last iteration
ret = self.buff
self.buff = b""
return ret
data = self.fileobj.read(self.blocksize)
if data:
self.buff += self.decompressor.decompress(data)
else:
self.buff += self.decompressor.flush()
class SpaceCompressor(object):
"""Not actually a compresor. It squeezes spaces."""
spacerc = re.compile(r"\s+")
unused_data = ""
def __init__(self):
self.lastspace = False
def decompress(self, data):
data = self.spacerc.sub(" ", data)
newlastspace = data[-1:] == " "
if self.lastspace and data[0:1] == " ":
data = data[1:]
self.lastspace = newlastspace
return data
def flush(self):
return ""
def copy(self):
new = SpaceCompressor()
new.lastspace = self.lastspace
return new
|
