The Android Open Source Project | cf31fe9 | 2008-10-21 07:00:00 -0700 | [diff] [blame^] | 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 | """InputStream is the primitive interface for reading bits from the wire. |
| 32 | |
| 33 | All protocol buffer deserialization can be expressed in terms of |
| 34 | the InputStream primitives provided here. |
| 35 | """ |
| 36 | |
| 37 | __author__ = 'robinson@google.com (Will Robinson)' |
| 38 | |
| 39 | import struct |
| 40 | from array import array |
| 41 | from froofle.protobuf import message |
| 42 | from froofle.protobuf.internal import wire_format |
| 43 | |
| 44 | |
| 45 | # Note that much of this code is ported from //net/proto/ProtocolBuffer, and |
| 46 | # that the interface is strongly inspired by CodedInputStream from the C++ |
| 47 | # proto2 implementation. |
| 48 | |
| 49 | |
| 50 | class InputStreamBuffer(object): |
| 51 | |
| 52 | """Contains all logic for reading bits, and dealing with stream position. |
| 53 | |
| 54 | If an InputStream method ever raises an exception, the stream is left |
| 55 | in an indeterminate state and is not safe for further use. |
| 56 | """ |
| 57 | |
| 58 | def __init__(self, s): |
| 59 | # What we really want is something like array('B', s), where elements we |
| 60 | # read from the array are already given to us as one-byte integers. BUT |
| 61 | # using array() instead of buffer() would force full string copies to result |
| 62 | # from each GetSubBuffer() call. |
| 63 | # |
| 64 | # So, if the N serialized bytes of a single protocol buffer object are |
| 65 | # split evenly between 2 child messages, and so on recursively, using |
| 66 | # array('B', s) instead of buffer() would incur an additional N*logN bytes |
| 67 | # copied during deserialization. |
| 68 | # |
| 69 | # The higher constant overhead of having to ord() for every byte we read |
| 70 | # from the buffer in _ReadVarintHelper() could definitely lead to worse |
| 71 | # performance in many real-world scenarios, even if the asymptotic |
| 72 | # complexity is better. However, our real answer is that the mythical |
| 73 | # Python/C extension module output mode for the protocol compiler will |
| 74 | # be blazing-fast and will eliminate most use of this class anyway. |
| 75 | self._buffer = buffer(s) |
| 76 | self._pos = 0 |
| 77 | |
| 78 | def EndOfStream(self): |
| 79 | """Returns true iff we're at the end of the stream. |
| 80 | If this returns true, then a call to any other InputStream method |
| 81 | will raise an exception. |
| 82 | """ |
| 83 | return self._pos >= len(self._buffer) |
| 84 | |
| 85 | def Position(self): |
| 86 | """Returns the current position in the stream, or equivalently, the |
| 87 | number of bytes read so far. |
| 88 | """ |
| 89 | return self._pos |
| 90 | |
| 91 | def GetSubBuffer(self, size=None): |
| 92 | """Returns a sequence-like object that represents a portion of our |
| 93 | underlying sequence. |
| 94 | |
| 95 | Position 0 in the returned object corresponds to self.Position() |
| 96 | in this stream. |
| 97 | |
| 98 | If size is specified, then the returned object ends after the |
| 99 | next "size" bytes in this stream. If size is not specified, |
| 100 | then the returned object ends at the end of this stream. |
| 101 | |
| 102 | We guarantee that the returned object R supports the Python buffer |
| 103 | interface (and thus that the call buffer(R) will work). |
| 104 | |
| 105 | Note that the returned buffer is read-only. |
| 106 | |
| 107 | The intended use for this method is for nested-message and nested-group |
| 108 | deserialization, where we want to make a recursive MergeFromString() |
| 109 | call on the portion of the original sequence that contains the serialized |
| 110 | nested message. (And we'd like to do so without making unnecessary string |
| 111 | copies). |
| 112 | |
| 113 | REQUIRES: size is nonnegative. |
| 114 | """ |
| 115 | # Note that buffer() doesn't perform any actual string copy. |
| 116 | if size is None: |
| 117 | return buffer(self._buffer, self._pos) |
| 118 | else: |
| 119 | if size < 0: |
| 120 | raise message.DecodeError('Negative size %d' % size) |
| 121 | return buffer(self._buffer, self._pos, size) |
| 122 | |
| 123 | def SkipBytes(self, num_bytes): |
| 124 | """Skip num_bytes bytes ahead, or go to the end of the stream, whichever |
| 125 | comes first. |
| 126 | |
| 127 | REQUIRES: num_bytes is nonnegative. |
| 128 | """ |
| 129 | if num_bytes < 0: |
| 130 | raise message.DecodeError('Negative num_bytes %d' % num_bytes) |
| 131 | self._pos += num_bytes |
| 132 | self._pos = min(self._pos, len(self._buffer)) |
| 133 | |
| 134 | def ReadBytes(self, size): |
| 135 | """Reads up to 'size' bytes from the stream, stopping early |
| 136 | only if we reach the end of the stream. Returns the bytes read |
| 137 | as a string. |
| 138 | """ |
| 139 | if size < 0: |
| 140 | raise message.DecodeError('Negative size %d' % size) |
| 141 | s = (self._buffer[self._pos : self._pos + size]) |
| 142 | self._pos += len(s) # Only advance by the number of bytes actually read. |
| 143 | return s |
| 144 | |
| 145 | def ReadLittleEndian32(self): |
| 146 | """Interprets the next 4 bytes of the stream as a little-endian |
| 147 | encoded, unsiged 32-bit integer, and returns that integer. |
| 148 | """ |
| 149 | try: |
| 150 | i = struct.unpack(wire_format.FORMAT_UINT32_LITTLE_ENDIAN, |
| 151 | self._buffer[self._pos : self._pos + 4]) |
| 152 | self._pos += 4 |
| 153 | return i[0] # unpack() result is a 1-element tuple. |
| 154 | except struct.error, e: |
| 155 | raise message.DecodeError(e) |
| 156 | |
| 157 | def ReadLittleEndian64(self): |
| 158 | """Interprets the next 8 bytes of the stream as a little-endian |
| 159 | encoded, unsiged 64-bit integer, and returns that integer. |
| 160 | """ |
| 161 | try: |
| 162 | i = struct.unpack(wire_format.FORMAT_UINT64_LITTLE_ENDIAN, |
| 163 | self._buffer[self._pos : self._pos + 8]) |
| 164 | self._pos += 8 |
| 165 | return i[0] # unpack() result is a 1-element tuple. |
| 166 | except struct.error, e: |
| 167 | raise message.DecodeError(e) |
| 168 | |
| 169 | def ReadVarint32(self): |
| 170 | """Reads a varint from the stream, interprets this varint |
| 171 | as a signed, 32-bit integer, and returns the integer. |
| 172 | """ |
| 173 | i = self.ReadVarint64() |
| 174 | if not wire_format.INT32_MIN <= i <= wire_format.INT32_MAX: |
| 175 | raise message.DecodeError('Value out of range for int32: %d' % i) |
| 176 | return int(i) |
| 177 | |
| 178 | def ReadVarUInt32(self): |
| 179 | """Reads a varint from the stream, interprets this varint |
| 180 | as an unsigned, 32-bit integer, and returns the integer. |
| 181 | """ |
| 182 | i = self.ReadVarUInt64() |
| 183 | if i > wire_format.UINT32_MAX: |
| 184 | raise message.DecodeError('Value out of range for uint32: %d' % i) |
| 185 | return i |
| 186 | |
| 187 | def ReadVarint64(self): |
| 188 | """Reads a varint from the stream, interprets this varint |
| 189 | as a signed, 64-bit integer, and returns the integer. |
| 190 | """ |
| 191 | i = self.ReadVarUInt64() |
| 192 | if i > wire_format.INT64_MAX: |
| 193 | i -= (1 << 64) |
| 194 | return i |
| 195 | |
| 196 | def ReadVarUInt64(self): |
| 197 | """Reads a varint from the stream, interprets this varint |
| 198 | as an unsigned, 64-bit integer, and returns the integer. |
| 199 | """ |
| 200 | i = self._ReadVarintHelper() |
| 201 | if not 0 <= i <= wire_format.UINT64_MAX: |
| 202 | raise message.DecodeError('Value out of range for uint64: %d' % i) |
| 203 | return i |
| 204 | |
| 205 | def _ReadVarintHelper(self): |
| 206 | """Helper for the various varint-reading methods above. |
| 207 | Reads an unsigned, varint-encoded integer from the stream and |
| 208 | returns this integer. |
| 209 | |
| 210 | Does no bounds checking except to ensure that we read at most as many bytes |
| 211 | as could possibly be present in a varint-encoded 64-bit number. |
| 212 | """ |
| 213 | result = 0 |
| 214 | shift = 0 |
| 215 | while 1: |
| 216 | if shift >= 64: |
| 217 | raise message.DecodeError('Too many bytes when decoding varint.') |
| 218 | try: |
| 219 | b = ord(self._buffer[self._pos]) |
| 220 | except IndexError: |
| 221 | raise message.DecodeError('Truncated varint.') |
| 222 | self._pos += 1 |
| 223 | result |= ((b & 0x7f) << shift) |
| 224 | shift += 7 |
| 225 | if not (b & 0x80): |
| 226 | return result |
| 227 | |
| 228 | class InputStreamArray(object): |
| 229 | def __init__(self, s): |
| 230 | self._buffer = array('B', s) |
| 231 | self._pos = 0 |
| 232 | |
| 233 | def EndOfStream(self): |
| 234 | return self._pos >= len(self._buffer) |
| 235 | |
| 236 | def Position(self): |
| 237 | return self._pos |
| 238 | |
| 239 | def GetSubBuffer(self, size=None): |
| 240 | if size is None: |
| 241 | return self._buffer[self._pos : ].tostring() |
| 242 | else: |
| 243 | if size < 0: |
| 244 | raise message.DecodeError('Negative size %d' % size) |
| 245 | return self._buffer[self._pos : self._pos + size].tostring() |
| 246 | |
| 247 | def SkipBytes(self, num_bytes): |
| 248 | if num_bytes < 0: |
| 249 | raise message.DecodeError('Negative num_bytes %d' % num_bytes) |
| 250 | self._pos += num_bytes |
| 251 | self._pos = min(self._pos, len(self._buffer)) |
| 252 | |
| 253 | def ReadBytes(self, size): |
| 254 | if size < 0: |
| 255 | raise message.DecodeError('Negative size %d' % size) |
| 256 | s = self._buffer[self._pos : self._pos + size].tostring() |
| 257 | self._pos += len(s) # Only advance by the number of bytes actually read. |
| 258 | return s |
| 259 | |
| 260 | def ReadLittleEndian32(self): |
| 261 | try: |
| 262 | i = struct.unpack(wire_format.FORMAT_UINT32_LITTLE_ENDIAN, |
| 263 | self._buffer[self._pos : self._pos + 4]) |
| 264 | self._pos += 4 |
| 265 | return i[0] # unpack() result is a 1-element tuple. |
| 266 | except struct.error, e: |
| 267 | raise message.DecodeError(e) |
| 268 | |
| 269 | def ReadLittleEndian64(self): |
| 270 | try: |
| 271 | i = struct.unpack(wire_format.FORMAT_UINT64_LITTLE_ENDIAN, |
| 272 | self._buffer[self._pos : self._pos + 8]) |
| 273 | self._pos += 8 |
| 274 | return i[0] # unpack() result is a 1-element tuple. |
| 275 | except struct.error, e: |
| 276 | raise message.DecodeError(e) |
| 277 | |
| 278 | def ReadVarint32(self): |
| 279 | i = self.ReadVarint64() |
| 280 | if not wire_format.INT32_MIN <= i <= wire_format.INT32_MAX: |
| 281 | raise message.DecodeError('Value out of range for int32: %d' % i) |
| 282 | return int(i) |
| 283 | |
| 284 | def ReadVarUInt32(self): |
| 285 | i = self.ReadVarUInt64() |
| 286 | if i > wire_format.UINT32_MAX: |
| 287 | raise message.DecodeError('Value out of range for uint32: %d' % i) |
| 288 | return i |
| 289 | |
| 290 | def ReadVarint64(self): |
| 291 | i = self.ReadVarUInt64() |
| 292 | if i > wire_format.INT64_MAX: |
| 293 | i -= (1 << 64) |
| 294 | return i |
| 295 | |
| 296 | def ReadVarUInt64(self): |
| 297 | i = self._ReadVarintHelper() |
| 298 | if not 0 <= i <= wire_format.UINT64_MAX: |
| 299 | raise message.DecodeError('Value out of range for uint64: %d' % i) |
| 300 | return i |
| 301 | |
| 302 | def _ReadVarintHelper(self): |
| 303 | result = 0 |
| 304 | shift = 0 |
| 305 | while 1: |
| 306 | if shift >= 64: |
| 307 | raise message.DecodeError('Too many bytes when decoding varint.') |
| 308 | try: |
| 309 | b = self._buffer[self._pos] |
| 310 | except IndexError: |
| 311 | raise message.DecodeError('Truncated varint.') |
| 312 | self._pos += 1 |
| 313 | result |= ((b & 0x7f) << shift) |
| 314 | shift += 7 |
| 315 | if not (b & 0x80): |
| 316 | return result |
| 317 | |
| 318 | try: |
| 319 | buffer("") |
| 320 | InputStream = InputStreamBuffer |
| 321 | except NotImplementedError: |
| 322 | # Google App Engine: dev_appserver.py |
| 323 | InputStream = InputStreamArray |
| 324 | except RuntimeError: |
| 325 | # Google App Engine: production |
| 326 | InputStream = InputStreamArray |