diff options
Diffstat (limited to 'froofle/protobuf/internal/wire_format.py')
-rw-r--r-- | froofle/protobuf/internal/wire_format.py | 236 |
1 files changed, 236 insertions, 0 deletions
diff --git a/froofle/protobuf/internal/wire_format.py b/froofle/protobuf/internal/wire_format.py new file mode 100644 index 00000000..4d823c8d --- /dev/null +++ b/froofle/protobuf/internal/wire_format.py | |||
@@ -0,0 +1,236 @@ | |||
1 | # Protocol Buffers - Google's data interchange format | ||
2 | # Copyright 2008 Google Inc. All rights reserved. | ||
3 | # http://code.google.com/p/protobuf/ | ||
4 | # | ||
5 | # Redistribution and use in source and binary forms, with or without | ||
6 | # modification, are permitted provided that the following conditions are | ||
7 | # met: | ||
8 | # | ||
9 | # * Redistributions of source code must retain the above copyright | ||
10 | # notice, this list of conditions and the following disclaimer. | ||
11 | # * Redistributions in binary form must reproduce the above | ||
12 | # copyright notice, this list of conditions and the following disclaimer | ||
13 | # in the documentation and/or other materials provided with the | ||
14 | # distribution. | ||
15 | # * Neither the name of Google Inc. nor the names of its | ||
16 | # contributors may be used to endorse or promote products derived from | ||
17 | # this software without specific prior written permission. | ||
18 | # | ||
19 | # THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS | ||
20 | # "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT | ||
21 | # LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR | ||
22 | # A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT | ||
23 | # OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, | ||
24 | # SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT | ||
25 | # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, | ||
26 | # DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY | ||
27 | # THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT | ||
28 | # (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE | ||
29 | # OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. | ||
30 | |||
31 | """Constants and static functions to support protocol buffer wire format.""" | ||
32 | |||
33 | __author__ = 'robinson@google.com (Will Robinson)' | ||
34 | |||
35 | import struct | ||
36 | from froofle.protobuf import message | ||
37 | |||
38 | |||
39 | TAG_TYPE_BITS = 3 # Number of bits used to hold type info in a proto tag. | ||
40 | _TAG_TYPE_MASK = (1 << TAG_TYPE_BITS) - 1 # 0x7 | ||
41 | |||
42 | # These numbers identify the wire type of a protocol buffer value. | ||
43 | # We use the least-significant TAG_TYPE_BITS bits of the varint-encoded | ||
44 | # tag-and-type to store one of these WIRETYPE_* constants. | ||
45 | # These values must match WireType enum in //net/proto2/public/wire_format.h. | ||
46 | WIRETYPE_VARINT = 0 | ||
47 | WIRETYPE_FIXED64 = 1 | ||
48 | WIRETYPE_LENGTH_DELIMITED = 2 | ||
49 | WIRETYPE_START_GROUP = 3 | ||
50 | WIRETYPE_END_GROUP = 4 | ||
51 | WIRETYPE_FIXED32 = 5 | ||
52 | _WIRETYPE_MAX = 5 | ||
53 | |||
54 | |||
55 | # Bounds for various integer types. | ||
56 | INT32_MAX = int((1 << 31) - 1) | ||
57 | INT32_MIN = int(-(1 << 31)) | ||
58 | UINT32_MAX = (1 << 32) - 1 | ||
59 | |||
60 | INT64_MAX = (1 << 63) - 1 | ||
61 | INT64_MIN = -(1 << 63) | ||
62 | UINT64_MAX = (1 << 64) - 1 | ||
63 | |||
64 | # "struct" format strings that will encode/decode the specified formats. | ||
65 | FORMAT_UINT32_LITTLE_ENDIAN = '<I' | ||
66 | FORMAT_UINT64_LITTLE_ENDIAN = '<Q' | ||
67 | |||
68 | |||
69 | # We'll have to provide alternate implementations of AppendLittleEndian*() on | ||
70 | # any architectures where these checks fail. | ||
71 | if struct.calcsize(FORMAT_UINT32_LITTLE_ENDIAN) != 4: | ||
72 | raise AssertionError('Format "I" is not a 32-bit number.') | ||
73 | if struct.calcsize(FORMAT_UINT64_LITTLE_ENDIAN) != 8: | ||
74 | raise AssertionError('Format "Q" is not a 64-bit number.') | ||
75 | |||
76 | |||
77 | def PackTag(field_number, wire_type): | ||
78 | """Returns an unsigned 32-bit integer that encodes the field number and | ||
79 | wire type information in standard protocol message wire format. | ||
80 | |||
81 | Args: | ||
82 | field_number: Expected to be an integer in the range [1, 1 << 29) | ||
83 | wire_type: One of the WIRETYPE_* constants. | ||
84 | """ | ||
85 | if not 0 <= wire_type <= _WIRETYPE_MAX: | ||
86 | raise message.EncodeError('Unknown wire type: %d' % wire_type) | ||
87 | return (field_number << TAG_TYPE_BITS) | wire_type | ||
88 | |||
89 | |||
90 | def UnpackTag(tag): | ||
91 | """The inverse of PackTag(). Given an unsigned 32-bit number, | ||
92 | returns a (field_number, wire_type) tuple. | ||
93 | """ | ||
94 | return (tag >> TAG_TYPE_BITS), (tag & _TAG_TYPE_MASK) | ||
95 | |||
96 | |||
97 | def ZigZagEncode(value): | ||
98 | """ZigZag Transform: Encodes signed integers so that they can be | ||
99 | effectively used with varint encoding. See wire_format.h for | ||
100 | more details. | ||
101 | """ | ||
102 | if value >= 0: | ||
103 | return value << 1 | ||
104 | return (value << 1) ^ (~0) | ||
105 | |||
106 | |||
107 | def ZigZagDecode(value): | ||
108 | """Inverse of ZigZagEncode().""" | ||
109 | if not value & 0x1: | ||
110 | return value >> 1 | ||
111 | return (value >> 1) ^ (~0) | ||
112 | |||
113 | |||
114 | |||
115 | # The *ByteSize() functions below return the number of bytes required to | ||
116 | # serialize "field number + type" information and then serialize the value. | ||
117 | |||
118 | |||
119 | def Int32ByteSize(field_number, int32): | ||
120 | return Int64ByteSize(field_number, int32) | ||
121 | |||
122 | |||
123 | def Int64ByteSize(field_number, int64): | ||
124 | # Have to convert to uint before calling UInt64ByteSize(). | ||
125 | return UInt64ByteSize(field_number, 0xffffffffffffffff & int64) | ||
126 | |||
127 | |||
128 | def UInt32ByteSize(field_number, uint32): | ||
129 | return UInt64ByteSize(field_number, uint32) | ||
130 | |||
131 | |||
132 | def UInt64ByteSize(field_number, uint64): | ||
133 | return _TagByteSize(field_number) + _VarUInt64ByteSizeNoTag(uint64) | ||
134 | |||
135 | |||
136 | def SInt32ByteSize(field_number, int32): | ||
137 | return UInt32ByteSize(field_number, ZigZagEncode(int32)) | ||
138 | |||
139 | |||
140 | def SInt64ByteSize(field_number, int64): | ||
141 | return UInt64ByteSize(field_number, ZigZagEncode(int64)) | ||
142 | |||
143 | |||
144 | def Fixed32ByteSize(field_number, fixed32): | ||
145 | return _TagByteSize(field_number) + 4 | ||
146 | |||
147 | |||
148 | def Fixed64ByteSize(field_number, fixed64): | ||
149 | return _TagByteSize(field_number) + 8 | ||
150 | |||
151 | |||
152 | def SFixed32ByteSize(field_number, sfixed32): | ||
153 | return _TagByteSize(field_number) + 4 | ||
154 | |||
155 | |||
156 | def SFixed64ByteSize(field_number, sfixed64): | ||
157 | return _TagByteSize(field_number) + 8 | ||
158 | |||
159 | |||
160 | def FloatByteSize(field_number, flt): | ||
161 | return _TagByteSize(field_number) + 4 | ||
162 | |||
163 | |||
164 | def DoubleByteSize(field_number, double): | ||
165 | return _TagByteSize(field_number) + 8 | ||
166 | |||
167 | |||
168 | def BoolByteSize(field_number, b): | ||
169 | return _TagByteSize(field_number) + 1 | ||
170 | |||
171 | |||
172 | def EnumByteSize(field_number, enum): | ||
173 | return UInt32ByteSize(field_number, enum) | ||
174 | |||
175 | |||
176 | def StringByteSize(field_number, string): | ||
177 | return BytesByteSize(field_number, string.encode('utf-8')) | ||
178 | |||
179 | |||
180 | def BytesByteSize(field_number, b): | ||
181 | return (_TagByteSize(field_number) | ||
182 | + _VarUInt64ByteSizeNoTag(len(b)) | ||
183 | + len(b)) | ||
184 | |||
185 | |||
186 | def GroupByteSize(field_number, message): | ||
187 | return (2 * _TagByteSize(field_number) # START and END group. | ||
188 | + message.ByteSize()) | ||
189 | |||
190 | |||
191 | def MessageByteSize(field_number, message): | ||
192 | return (_TagByteSize(field_number) | ||
193 | + _VarUInt64ByteSizeNoTag(message.ByteSize()) | ||
194 | + message.ByteSize()) | ||
195 | |||
196 | |||
197 | def MessageSetItemByteSize(field_number, msg): | ||
198 | # First compute the sizes of the tags. | ||
199 | # There are 2 tags for the beginning and ending of the repeated group, that | ||
200 | # is field number 1, one with field number 2 (type_id) and one with field | ||
201 | # number 3 (message). | ||
202 | total_size = (2 * _TagByteSize(1) + _TagByteSize(2) + _TagByteSize(3)) | ||
203 | |||
204 | # Add the number of bytes for type_id. | ||
205 | total_size += _VarUInt64ByteSizeNoTag(field_number) | ||
206 | |||
207 | message_size = msg.ByteSize() | ||
208 | |||
209 | # The number of bytes for encoding the length of the message. | ||
210 | total_size += _VarUInt64ByteSizeNoTag(message_size) | ||
211 | |||
212 | # The size of the message. | ||
213 | total_size += message_size | ||
214 | return total_size | ||
215 | |||
216 | |||
217 | # Private helper functions for the *ByteSize() functions above. | ||
218 | |||
219 | |||
220 | def _TagByteSize(field_number): | ||
221 | """Returns the bytes required to serialize a tag with this field number.""" | ||
222 | # Just pass in type 0, since the type won't affect the tag+type size. | ||
223 | return _VarUInt64ByteSizeNoTag(PackTag(field_number, 0)) | ||
224 | |||
225 | |||
226 | def _VarUInt64ByteSizeNoTag(uint64): | ||
227 | """Returns the bytes required to serialize a single varint. | ||
228 | uint64 must be unsigned. | ||
229 | """ | ||
230 | if uint64 > UINT64_MAX: | ||
231 | raise message.EncodeError('Value out of range: %d' % uint64) | ||
232 | bytes = 1 | ||
233 | while uint64 > 0x7f: | ||
234 | bytes += 1 | ||
235 | uint64 >>= 7 | ||
236 | return bytes | ||