temporal | 40ee551 | 2008-07-10 02:12:20 +0000 | [diff] [blame] | 1 | # Protocol Buffers - Google's data interchange format |
kenton@google.com | 24bf56f | 2008-09-24 20:31:01 +0000 | [diff] [blame] | 2 | # Copyright 2008 Google Inc. All rights reserved. |
temporal | 40ee551 | 2008-07-10 02:12:20 +0000 | [diff] [blame] | 3 | # http://code.google.com/p/protobuf/ |
| 4 | # |
kenton@google.com | 24bf56f | 2008-09-24 20:31:01 +0000 | [diff] [blame] | 5 | # Redistribution and use in source and binary forms, with or without |
| 6 | # modification, are permitted provided that the following conditions are |
| 7 | # met: |
temporal | 40ee551 | 2008-07-10 02:12:20 +0000 | [diff] [blame] | 8 | # |
kenton@google.com | 24bf56f | 2008-09-24 20:31:01 +0000 | [diff] [blame] | 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. |
temporal | 40ee551 | 2008-07-10 02:12:20 +0000 | [diff] [blame] | 18 | # |
kenton@google.com | 24bf56f | 2008-09-24 20:31:01 +0000 | [diff] [blame] | 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. |
temporal | 40ee551 | 2008-07-10 02:12:20 +0000 | [diff] [blame] | 30 | |
| 31 | """Contains routines for printing protocol messages in text format.""" |
| 32 | |
| 33 | __author__ = 'kenton@google.com (Kenton Varda)' |
| 34 | |
| 35 | import cStringIO |
kenton@google.com | 80b1d62 | 2009-07-29 01:13:20 +0000 | [diff] [blame] | 36 | import re |
temporal | 40ee551 | 2008-07-10 02:12:20 +0000 | [diff] [blame] | 37 | |
kenton@google.com | 80b1d62 | 2009-07-29 01:13:20 +0000 | [diff] [blame] | 38 | from collections import deque |
| 39 | from google.protobuf.internal import type_checkers |
temporal | 40ee551 | 2008-07-10 02:12:20 +0000 | [diff] [blame] | 40 | from google.protobuf import descriptor |
| 41 | |
kenton@google.com | 80b1d62 | 2009-07-29 01:13:20 +0000 | [diff] [blame] | 42 | __all__ = [ 'MessageToString', 'PrintMessage', 'PrintField', |
| 43 | 'PrintFieldValue', 'Merge' ] |
| 44 | |
| 45 | |
kenton@google.com | d0047c4 | 2009-12-23 02:01:01 +0000 | [diff] [blame] | 46 | # Infinity and NaN are not explicitly supported by Python pre-2.6, and |
| 47 | # float('inf') does not work on Windows (pre-2.6). |
kenton@google.com | 46ed74e | 2009-12-23 02:08:05 +0000 | [diff] [blame] | 48 | _INFINITY = 1e10000 # overflows, thus will actually be infinity. |
kenton@google.com | d0047c4 | 2009-12-23 02:01:01 +0000 | [diff] [blame] | 49 | _NAN = _INFINITY * 0 |
| 50 | |
| 51 | |
kenton@google.com | 80b1d62 | 2009-07-29 01:13:20 +0000 | [diff] [blame] | 52 | class ParseError(Exception): |
| 53 | """Thrown in case of ASCII parsing error.""" |
| 54 | |
temporal | 40ee551 | 2008-07-10 02:12:20 +0000 | [diff] [blame] | 55 | |
liujisi@google.com | 33165fe | 2010-11-02 13:14:58 +0000 | [diff] [blame^] | 56 | def MessageToString(message, as_utf8=False, as_one_line=False): |
temporal | 40ee551 | 2008-07-10 02:12:20 +0000 | [diff] [blame] | 57 | out = cStringIO.StringIO() |
liujisi@google.com | 33165fe | 2010-11-02 13:14:58 +0000 | [diff] [blame^] | 58 | PrintMessage(message, out, as_utf8=as_utf8, as_one_line=as_one_line) |
temporal | 40ee551 | 2008-07-10 02:12:20 +0000 | [diff] [blame] | 59 | result = out.getvalue() |
| 60 | out.close() |
liujisi@google.com | 33165fe | 2010-11-02 13:14:58 +0000 | [diff] [blame^] | 61 | if as_one_line: |
| 62 | return result.rstrip() |
temporal | 40ee551 | 2008-07-10 02:12:20 +0000 | [diff] [blame] | 63 | return result |
| 64 | |
kenton@google.com | 80b1d62 | 2009-07-29 01:13:20 +0000 | [diff] [blame] | 65 | |
liujisi@google.com | 33165fe | 2010-11-02 13:14:58 +0000 | [diff] [blame^] | 66 | def PrintMessage(message, out, indent=0, as_utf8=False, as_one_line=False): |
temporal | 40ee551 | 2008-07-10 02:12:20 +0000 | [diff] [blame] | 67 | for field, value in message.ListFields(): |
| 68 | if field.label == descriptor.FieldDescriptor.LABEL_REPEATED: |
| 69 | for element in value: |
liujisi@google.com | 33165fe | 2010-11-02 13:14:58 +0000 | [diff] [blame^] | 70 | PrintField(field, element, out, indent, as_utf8, as_one_line) |
temporal | 40ee551 | 2008-07-10 02:12:20 +0000 | [diff] [blame] | 71 | else: |
liujisi@google.com | 33165fe | 2010-11-02 13:14:58 +0000 | [diff] [blame^] | 72 | PrintField(field, value, out, indent, as_utf8, as_one_line) |
temporal | 40ee551 | 2008-07-10 02:12:20 +0000 | [diff] [blame] | 73 | |
kenton@google.com | 80b1d62 | 2009-07-29 01:13:20 +0000 | [diff] [blame] | 74 | |
liujisi@google.com | 33165fe | 2010-11-02 13:14:58 +0000 | [diff] [blame^] | 75 | def PrintField(field, value, out, indent=0, as_utf8=False, as_one_line=False): |
temporal | 40ee551 | 2008-07-10 02:12:20 +0000 | [diff] [blame] | 76 | """Print a single field name/value pair. For repeated fields, the value |
| 77 | should be a single element.""" |
| 78 | |
| 79 | out.write(' ' * indent); |
| 80 | if field.is_extension: |
| 81 | out.write('[') |
| 82 | if (field.containing_type.GetOptions().message_set_wire_format and |
| 83 | field.type == descriptor.FieldDescriptor.TYPE_MESSAGE and |
| 84 | field.message_type == field.extension_scope and |
| 85 | field.label == descriptor.FieldDescriptor.LABEL_OPTIONAL): |
| 86 | out.write(field.message_type.full_name) |
| 87 | else: |
| 88 | out.write(field.full_name) |
| 89 | out.write(']') |
| 90 | elif field.type == descriptor.FieldDescriptor.TYPE_GROUP: |
| 91 | # For groups, use the capitalized name. |
| 92 | out.write(field.message_type.name) |
| 93 | else: |
| 94 | out.write(field.name) |
| 95 | |
| 96 | if field.cpp_type != descriptor.FieldDescriptor.CPPTYPE_MESSAGE: |
| 97 | # The colon is optional in this case, but our cross-language golden files |
| 98 | # don't include it. |
| 99 | out.write(': ') |
| 100 | |
liujisi@google.com | 33165fe | 2010-11-02 13:14:58 +0000 | [diff] [blame^] | 101 | PrintFieldValue(field, value, out, indent, as_utf8, as_one_line) |
| 102 | if as_one_line: |
| 103 | out.write(' ') |
| 104 | else: |
| 105 | out.write('\n') |
temporal | 40ee551 | 2008-07-10 02:12:20 +0000 | [diff] [blame] | 106 | |
kenton@google.com | 80b1d62 | 2009-07-29 01:13:20 +0000 | [diff] [blame] | 107 | |
liujisi@google.com | 33165fe | 2010-11-02 13:14:58 +0000 | [diff] [blame^] | 108 | def PrintFieldValue(field, value, out, indent=0, |
| 109 | as_utf8=False, as_one_line=False): |
temporal | 40ee551 | 2008-07-10 02:12:20 +0000 | [diff] [blame] | 110 | """Print a single field value (not including name). For repeated fields, |
| 111 | the value should be a single element.""" |
| 112 | |
| 113 | if field.cpp_type == descriptor.FieldDescriptor.CPPTYPE_MESSAGE: |
liujisi@google.com | 33165fe | 2010-11-02 13:14:58 +0000 | [diff] [blame^] | 114 | if as_one_line: |
| 115 | out.write(' { ') |
| 116 | PrintMessage(value, out, indent, as_utf8, as_one_line) |
| 117 | out.write('}') |
| 118 | else: |
| 119 | out.write(' {\n') |
| 120 | PrintMessage(value, out, indent + 2, as_utf8, as_one_line) |
| 121 | out.write(' ' * indent + '}') |
temporal | 40ee551 | 2008-07-10 02:12:20 +0000 | [diff] [blame] | 122 | elif field.cpp_type == descriptor.FieldDescriptor.CPPTYPE_ENUM: |
| 123 | out.write(field.enum_type.values_by_number[value].name) |
| 124 | elif field.cpp_type == descriptor.FieldDescriptor.CPPTYPE_STRING: |
| 125 | out.write('\"') |
liujisi@google.com | 33165fe | 2010-11-02 13:14:58 +0000 | [diff] [blame^] | 126 | if type(value) is unicode: |
| 127 | out.write(_CEscape(value.encode('utf-8'), as_utf8)) |
| 128 | else: |
| 129 | out.write(_CEscape(value, as_utf8)) |
temporal | 40ee551 | 2008-07-10 02:12:20 +0000 | [diff] [blame] | 130 | out.write('\"') |
| 131 | elif field.cpp_type == descriptor.FieldDescriptor.CPPTYPE_BOOL: |
| 132 | if value: |
| 133 | out.write("true") |
| 134 | else: |
| 135 | out.write("false") |
| 136 | else: |
| 137 | out.write(str(value)) |
| 138 | |
kenton@google.com | 80b1d62 | 2009-07-29 01:13:20 +0000 | [diff] [blame] | 139 | |
| 140 | def Merge(text, message): |
| 141 | """Merges an ASCII representation of a protocol message into a message. |
| 142 | |
| 143 | Args: |
| 144 | text: Message ASCII representation. |
| 145 | message: A protocol buffer message to merge into. |
| 146 | |
| 147 | Raises: |
| 148 | ParseError: On ASCII parsing problems. |
| 149 | """ |
| 150 | tokenizer = _Tokenizer(text) |
| 151 | while not tokenizer.AtEnd(): |
| 152 | _MergeField(tokenizer, message) |
| 153 | |
| 154 | |
| 155 | def _MergeField(tokenizer, message): |
| 156 | """Merges a single protocol message field into a message. |
| 157 | |
| 158 | Args: |
| 159 | tokenizer: A tokenizer to parse the field name and values. |
| 160 | message: A protocol message to record the data. |
| 161 | |
| 162 | Raises: |
| 163 | ParseError: In case of ASCII parsing problems. |
| 164 | """ |
| 165 | message_descriptor = message.DESCRIPTOR |
| 166 | if tokenizer.TryConsume('['): |
| 167 | name = [tokenizer.ConsumeIdentifier()] |
| 168 | while tokenizer.TryConsume('.'): |
| 169 | name.append(tokenizer.ConsumeIdentifier()) |
| 170 | name = '.'.join(name) |
| 171 | |
kenton@google.com | fccb146 | 2009-12-18 02:11:36 +0000 | [diff] [blame] | 172 | if not message_descriptor.is_extendable: |
| 173 | raise tokenizer.ParseErrorPreviousToken( |
| 174 | 'Message type "%s" does not have extensions.' % |
| 175 | message_descriptor.full_name) |
kenton@google.com | 80b1d62 | 2009-07-29 01:13:20 +0000 | [diff] [blame] | 176 | field = message.Extensions._FindExtensionByName(name) |
| 177 | if not field: |
| 178 | raise tokenizer.ParseErrorPreviousToken( |
| 179 | 'Extension "%s" not registered.' % name) |
| 180 | elif message_descriptor != field.containing_type: |
| 181 | raise tokenizer.ParseErrorPreviousToken( |
| 182 | 'Extension "%s" does not extend message type "%s".' % ( |
| 183 | name, message_descriptor.full_name)) |
| 184 | tokenizer.Consume(']') |
| 185 | else: |
| 186 | name = tokenizer.ConsumeIdentifier() |
| 187 | field = message_descriptor.fields_by_name.get(name, None) |
| 188 | |
| 189 | # Group names are expected to be capitalized as they appear in the |
| 190 | # .proto file, which actually matches their type names, not their field |
| 191 | # names. |
| 192 | if not field: |
| 193 | field = message_descriptor.fields_by_name.get(name.lower(), None) |
| 194 | if field and field.type != descriptor.FieldDescriptor.TYPE_GROUP: |
| 195 | field = None |
| 196 | |
| 197 | if (field and field.type == descriptor.FieldDescriptor.TYPE_GROUP and |
| 198 | field.message_type.name != name): |
| 199 | field = None |
| 200 | |
| 201 | if not field: |
| 202 | raise tokenizer.ParseErrorPreviousToken( |
| 203 | 'Message type "%s" has no field named "%s".' % ( |
| 204 | message_descriptor.full_name, name)) |
| 205 | |
| 206 | if field.cpp_type == descriptor.FieldDescriptor.CPPTYPE_MESSAGE: |
| 207 | tokenizer.TryConsume(':') |
| 208 | |
| 209 | if tokenizer.TryConsume('<'): |
| 210 | end_token = '>' |
| 211 | else: |
| 212 | tokenizer.Consume('{') |
| 213 | end_token = '}' |
| 214 | |
| 215 | if field.label == descriptor.FieldDescriptor.LABEL_REPEATED: |
| 216 | if field.is_extension: |
| 217 | sub_message = message.Extensions[field].add() |
| 218 | else: |
| 219 | sub_message = getattr(message, field.name).add() |
| 220 | else: |
| 221 | if field.is_extension: |
| 222 | sub_message = message.Extensions[field] |
| 223 | else: |
| 224 | sub_message = getattr(message, field.name) |
kenton@google.com | fccb146 | 2009-12-18 02:11:36 +0000 | [diff] [blame] | 225 | sub_message.SetInParent() |
kenton@google.com | 80b1d62 | 2009-07-29 01:13:20 +0000 | [diff] [blame] | 226 | |
| 227 | while not tokenizer.TryConsume(end_token): |
| 228 | if tokenizer.AtEnd(): |
| 229 | raise tokenizer.ParseErrorPreviousToken('Expected "%s".' % (end_token)) |
| 230 | _MergeField(tokenizer, sub_message) |
| 231 | else: |
| 232 | _MergeScalarField(tokenizer, message, field) |
| 233 | |
| 234 | |
| 235 | def _MergeScalarField(tokenizer, message, field): |
| 236 | """Merges a single protocol message scalar field into a message. |
| 237 | |
| 238 | Args: |
| 239 | tokenizer: A tokenizer to parse the field value. |
| 240 | message: A protocol message to record the data. |
| 241 | field: The descriptor of the field to be merged. |
| 242 | |
| 243 | Raises: |
| 244 | ParseError: In case of ASCII parsing problems. |
| 245 | RuntimeError: On runtime errors. |
| 246 | """ |
| 247 | tokenizer.Consume(':') |
| 248 | value = None |
| 249 | |
| 250 | if field.type in (descriptor.FieldDescriptor.TYPE_INT32, |
| 251 | descriptor.FieldDescriptor.TYPE_SINT32, |
| 252 | descriptor.FieldDescriptor.TYPE_SFIXED32): |
| 253 | value = tokenizer.ConsumeInt32() |
| 254 | elif field.type in (descriptor.FieldDescriptor.TYPE_INT64, |
| 255 | descriptor.FieldDescriptor.TYPE_SINT64, |
| 256 | descriptor.FieldDescriptor.TYPE_SFIXED64): |
| 257 | value = tokenizer.ConsumeInt64() |
| 258 | elif field.type in (descriptor.FieldDescriptor.TYPE_UINT32, |
| 259 | descriptor.FieldDescriptor.TYPE_FIXED32): |
| 260 | value = tokenizer.ConsumeUint32() |
| 261 | elif field.type in (descriptor.FieldDescriptor.TYPE_UINT64, |
| 262 | descriptor.FieldDescriptor.TYPE_FIXED64): |
| 263 | value = tokenizer.ConsumeUint64() |
| 264 | elif field.type in (descriptor.FieldDescriptor.TYPE_FLOAT, |
| 265 | descriptor.FieldDescriptor.TYPE_DOUBLE): |
| 266 | value = tokenizer.ConsumeFloat() |
| 267 | elif field.type == descriptor.FieldDescriptor.TYPE_BOOL: |
| 268 | value = tokenizer.ConsumeBool() |
| 269 | elif field.type == descriptor.FieldDescriptor.TYPE_STRING: |
| 270 | value = tokenizer.ConsumeString() |
| 271 | elif field.type == descriptor.FieldDescriptor.TYPE_BYTES: |
| 272 | value = tokenizer.ConsumeByteString() |
| 273 | elif field.type == descriptor.FieldDescriptor.TYPE_ENUM: |
| 274 | # Enum can be specified by a number (the enum value), or by |
| 275 | # a string literal (the enum name). |
| 276 | enum_descriptor = field.enum_type |
| 277 | if tokenizer.LookingAtInteger(): |
| 278 | number = tokenizer.ConsumeInt32() |
| 279 | enum_value = enum_descriptor.values_by_number.get(number, None) |
| 280 | if enum_value is None: |
| 281 | raise tokenizer.ParseErrorPreviousToken( |
| 282 | 'Enum type "%s" has no value with number %d.' % ( |
| 283 | enum_descriptor.full_name, number)) |
| 284 | else: |
| 285 | identifier = tokenizer.ConsumeIdentifier() |
| 286 | enum_value = enum_descriptor.values_by_name.get(identifier, None) |
| 287 | if enum_value is None: |
| 288 | raise tokenizer.ParseErrorPreviousToken( |
| 289 | 'Enum type "%s" has no value named %s.' % ( |
| 290 | enum_descriptor.full_name, identifier)) |
| 291 | value = enum_value.number |
| 292 | else: |
| 293 | raise RuntimeError('Unknown field type %d' % field.type) |
| 294 | |
| 295 | if field.label == descriptor.FieldDescriptor.LABEL_REPEATED: |
| 296 | if field.is_extension: |
| 297 | message.Extensions[field].append(value) |
| 298 | else: |
| 299 | getattr(message, field.name).append(value) |
| 300 | else: |
| 301 | if field.is_extension: |
| 302 | message.Extensions[field] = value |
| 303 | else: |
| 304 | setattr(message, field.name, value) |
| 305 | |
| 306 | |
| 307 | class _Tokenizer(object): |
| 308 | """Protocol buffer ASCII representation tokenizer. |
| 309 | |
| 310 | This class handles the lower level string parsing by splitting it into |
| 311 | meaningful tokens. |
| 312 | |
| 313 | It was directly ported from the Java protocol buffer API. |
| 314 | """ |
| 315 | |
| 316 | _WHITESPACE = re.compile('(\\s|(#.*$))+', re.MULTILINE) |
| 317 | _TOKEN = re.compile( |
| 318 | '[a-zA-Z_][0-9a-zA-Z_+-]*|' # an identifier |
| 319 | '[0-9+-][0-9a-zA-Z_.+-]*|' # a number |
| 320 | '\"([^\"\n\\\\]|\\\\.)*(\"|\\\\?$)|' # a double-quoted string |
kenton@google.com | eef5f83 | 2009-12-23 01:32:45 +0000 | [diff] [blame] | 321 | '\'([^\'\n\\\\]|\\\\.)*(\'|\\\\?$)') # a single-quoted string |
kenton@google.com | 80b1d62 | 2009-07-29 01:13:20 +0000 | [diff] [blame] | 322 | _IDENTIFIER = re.compile('\w+') |
| 323 | _INTEGER_CHECKERS = [type_checkers.Uint32ValueChecker(), |
| 324 | type_checkers.Int32ValueChecker(), |
| 325 | type_checkers.Uint64ValueChecker(), |
| 326 | type_checkers.Int64ValueChecker()] |
| 327 | _FLOAT_INFINITY = re.compile('-?inf(inity)?f?', re.IGNORECASE) |
| 328 | _FLOAT_NAN = re.compile("nanf?", re.IGNORECASE) |
| 329 | |
| 330 | def __init__(self, text_message): |
| 331 | self._text_message = text_message |
| 332 | |
| 333 | self._position = 0 |
| 334 | self._line = -1 |
| 335 | self._column = 0 |
| 336 | self._token_start = None |
| 337 | self.token = '' |
| 338 | self._lines = deque(text_message.split('\n')) |
| 339 | self._current_line = '' |
| 340 | self._previous_line = 0 |
| 341 | self._previous_column = 0 |
| 342 | self._SkipWhitespace() |
| 343 | self.NextToken() |
| 344 | |
| 345 | def AtEnd(self): |
| 346 | """Checks the end of the text was reached. |
| 347 | |
| 348 | Returns: |
| 349 | True iff the end was reached. |
| 350 | """ |
liujisi@google.com | 33165fe | 2010-11-02 13:14:58 +0000 | [diff] [blame^] | 351 | return self.token == '' |
kenton@google.com | 80b1d62 | 2009-07-29 01:13:20 +0000 | [diff] [blame] | 352 | |
| 353 | def _PopLine(self): |
liujisi@google.com | 33165fe | 2010-11-02 13:14:58 +0000 | [diff] [blame^] | 354 | while len(self._current_line) <= self._column: |
kenton@google.com | 80b1d62 | 2009-07-29 01:13:20 +0000 | [diff] [blame] | 355 | if not self._lines: |
| 356 | self._current_line = '' |
| 357 | return |
| 358 | self._line += 1 |
| 359 | self._column = 0 |
| 360 | self._current_line = self._lines.popleft() |
| 361 | |
| 362 | def _SkipWhitespace(self): |
| 363 | while True: |
| 364 | self._PopLine() |
liujisi@google.com | 33165fe | 2010-11-02 13:14:58 +0000 | [diff] [blame^] | 365 | match = self._WHITESPACE.match(self._current_line, self._column) |
kenton@google.com | 80b1d62 | 2009-07-29 01:13:20 +0000 | [diff] [blame] | 366 | if not match: |
| 367 | break |
| 368 | length = len(match.group(0)) |
kenton@google.com | 80b1d62 | 2009-07-29 01:13:20 +0000 | [diff] [blame] | 369 | self._column += length |
| 370 | |
| 371 | def TryConsume(self, token): |
| 372 | """Tries to consume a given piece of text. |
| 373 | |
| 374 | Args: |
| 375 | token: Text to consume. |
| 376 | |
| 377 | Returns: |
| 378 | True iff the text was consumed. |
| 379 | """ |
| 380 | if self.token == token: |
| 381 | self.NextToken() |
| 382 | return True |
| 383 | return False |
| 384 | |
| 385 | def Consume(self, token): |
| 386 | """Consumes a piece of text. |
| 387 | |
| 388 | Args: |
| 389 | token: Text to consume. |
| 390 | |
| 391 | Raises: |
| 392 | ParseError: If the text couldn't be consumed. |
| 393 | """ |
| 394 | if not self.TryConsume(token): |
| 395 | raise self._ParseError('Expected "%s".' % token) |
| 396 | |
| 397 | def LookingAtInteger(self): |
| 398 | """Checks if the current token is an integer. |
| 399 | |
| 400 | Returns: |
| 401 | True iff the current token is an integer. |
| 402 | """ |
| 403 | if not self.token: |
| 404 | return False |
| 405 | c = self.token[0] |
| 406 | return (c >= '0' and c <= '9') or c == '-' or c == '+' |
| 407 | |
| 408 | def ConsumeIdentifier(self): |
| 409 | """Consumes protocol message field identifier. |
| 410 | |
| 411 | Returns: |
| 412 | Identifier string. |
| 413 | |
| 414 | Raises: |
| 415 | ParseError: If an identifier couldn't be consumed. |
| 416 | """ |
| 417 | result = self.token |
liujisi@google.com | 33165fe | 2010-11-02 13:14:58 +0000 | [diff] [blame^] | 418 | if not self._IDENTIFIER.match(result): |
kenton@google.com | 80b1d62 | 2009-07-29 01:13:20 +0000 | [diff] [blame] | 419 | raise self._ParseError('Expected identifier.') |
| 420 | self.NextToken() |
| 421 | return result |
| 422 | |
| 423 | def ConsumeInt32(self): |
| 424 | """Consumes a signed 32bit integer number. |
| 425 | |
| 426 | Returns: |
| 427 | The integer parsed. |
| 428 | |
| 429 | Raises: |
| 430 | ParseError: If a signed 32bit integer couldn't be consumed. |
| 431 | """ |
| 432 | try: |
| 433 | result = self._ParseInteger(self.token, is_signed=True, is_long=False) |
| 434 | except ValueError, e: |
| 435 | raise self._IntegerParseError(e) |
| 436 | self.NextToken() |
| 437 | return result |
| 438 | |
| 439 | def ConsumeUint32(self): |
| 440 | """Consumes an unsigned 32bit integer number. |
| 441 | |
| 442 | Returns: |
| 443 | The integer parsed. |
| 444 | |
| 445 | Raises: |
| 446 | ParseError: If an unsigned 32bit integer couldn't be consumed. |
| 447 | """ |
| 448 | try: |
| 449 | result = self._ParseInteger(self.token, is_signed=False, is_long=False) |
| 450 | except ValueError, e: |
| 451 | raise self._IntegerParseError(e) |
| 452 | self.NextToken() |
| 453 | return result |
| 454 | |
| 455 | def ConsumeInt64(self): |
| 456 | """Consumes a signed 64bit integer number. |
| 457 | |
| 458 | Returns: |
| 459 | The integer parsed. |
| 460 | |
| 461 | Raises: |
| 462 | ParseError: If a signed 64bit integer couldn't be consumed. |
| 463 | """ |
| 464 | try: |
| 465 | result = self._ParseInteger(self.token, is_signed=True, is_long=True) |
| 466 | except ValueError, e: |
| 467 | raise self._IntegerParseError(e) |
| 468 | self.NextToken() |
| 469 | return result |
| 470 | |
| 471 | def ConsumeUint64(self): |
| 472 | """Consumes an unsigned 64bit integer number. |
| 473 | |
| 474 | Returns: |
| 475 | The integer parsed. |
| 476 | |
| 477 | Raises: |
| 478 | ParseError: If an unsigned 64bit integer couldn't be consumed. |
| 479 | """ |
| 480 | try: |
| 481 | result = self._ParseInteger(self.token, is_signed=False, is_long=True) |
| 482 | except ValueError, e: |
| 483 | raise self._IntegerParseError(e) |
| 484 | self.NextToken() |
| 485 | return result |
| 486 | |
| 487 | def ConsumeFloat(self): |
| 488 | """Consumes an floating point number. |
| 489 | |
| 490 | Returns: |
| 491 | The number parsed. |
| 492 | |
| 493 | Raises: |
| 494 | ParseError: If a floating point number couldn't be consumed. |
| 495 | """ |
| 496 | text = self.token |
liujisi@google.com | 33165fe | 2010-11-02 13:14:58 +0000 | [diff] [blame^] | 497 | if self._FLOAT_INFINITY.match(text): |
kenton@google.com | 80b1d62 | 2009-07-29 01:13:20 +0000 | [diff] [blame] | 498 | self.NextToken() |
| 499 | if text.startswith('-'): |
kenton@google.com | d0047c4 | 2009-12-23 02:01:01 +0000 | [diff] [blame] | 500 | return -_INFINITY |
| 501 | return _INFINITY |
kenton@google.com | 80b1d62 | 2009-07-29 01:13:20 +0000 | [diff] [blame] | 502 | |
liujisi@google.com | 33165fe | 2010-11-02 13:14:58 +0000 | [diff] [blame^] | 503 | if self._FLOAT_NAN.match(text): |
kenton@google.com | 80b1d62 | 2009-07-29 01:13:20 +0000 | [diff] [blame] | 504 | self.NextToken() |
kenton@google.com | d0047c4 | 2009-12-23 02:01:01 +0000 | [diff] [blame] | 505 | return _NAN |
kenton@google.com | 80b1d62 | 2009-07-29 01:13:20 +0000 | [diff] [blame] | 506 | |
| 507 | try: |
| 508 | result = float(text) |
| 509 | except ValueError, e: |
| 510 | raise self._FloatParseError(e) |
| 511 | self.NextToken() |
| 512 | return result |
| 513 | |
| 514 | def ConsumeBool(self): |
| 515 | """Consumes a boolean value. |
| 516 | |
| 517 | Returns: |
| 518 | The bool parsed. |
| 519 | |
| 520 | Raises: |
| 521 | ParseError: If a boolean value couldn't be consumed. |
| 522 | """ |
liujisi@google.com | 33165fe | 2010-11-02 13:14:58 +0000 | [diff] [blame^] | 523 | if self.token in ('true', 't', '1'): |
kenton@google.com | 80b1d62 | 2009-07-29 01:13:20 +0000 | [diff] [blame] | 524 | self.NextToken() |
| 525 | return True |
liujisi@google.com | 33165fe | 2010-11-02 13:14:58 +0000 | [diff] [blame^] | 526 | elif self.token in ('false', 'f', '0'): |
kenton@google.com | 80b1d62 | 2009-07-29 01:13:20 +0000 | [diff] [blame] | 527 | self.NextToken() |
| 528 | return False |
| 529 | else: |
| 530 | raise self._ParseError('Expected "true" or "false".') |
| 531 | |
| 532 | def ConsumeString(self): |
| 533 | """Consumes a string value. |
| 534 | |
| 535 | Returns: |
| 536 | The string parsed. |
| 537 | |
| 538 | Raises: |
| 539 | ParseError: If a string value couldn't be consumed. |
| 540 | """ |
liujisi@google.com | 33165fe | 2010-11-02 13:14:58 +0000 | [diff] [blame^] | 541 | bytes = self.ConsumeByteString() |
| 542 | try: |
| 543 | return unicode(bytes, 'utf-8') |
| 544 | except UnicodeDecodeError, e: |
| 545 | raise self._StringParseError(e) |
kenton@google.com | 80b1d62 | 2009-07-29 01:13:20 +0000 | [diff] [blame] | 546 | |
| 547 | def ConsumeByteString(self): |
| 548 | """Consumes a byte array value. |
| 549 | |
| 550 | Returns: |
| 551 | The array parsed (as a string). |
| 552 | |
| 553 | Raises: |
| 554 | ParseError: If a byte array value couldn't be consumed. |
| 555 | """ |
kenton@google.com | 5353018 | 2010-01-07 02:08:03 +0000 | [diff] [blame] | 556 | list = [self._ConsumeSingleByteString()] |
kenton@google.com | eef5f83 | 2009-12-23 01:32:45 +0000 | [diff] [blame] | 557 | while len(self.token) > 0 and self.token[0] in ('\'', '"'): |
kenton@google.com | 5353018 | 2010-01-07 02:08:03 +0000 | [diff] [blame] | 558 | list.append(self._ConsumeSingleByteString()) |
kenton@google.com | eef5f83 | 2009-12-23 01:32:45 +0000 | [diff] [blame] | 559 | return "".join(list) |
| 560 | |
kenton@google.com | 5353018 | 2010-01-07 02:08:03 +0000 | [diff] [blame] | 561 | def _ConsumeSingleByteString(self): |
| 562 | """Consume one token of a string literal. |
| 563 | |
| 564 | String literals (whether bytes or text) can come in multiple adjacent |
| 565 | tokens which are automatically concatenated, like in C or Python. This |
| 566 | method only consumes one token. |
| 567 | """ |
kenton@google.com | 80b1d62 | 2009-07-29 01:13:20 +0000 | [diff] [blame] | 568 | text = self.token |
| 569 | if len(text) < 1 or text[0] not in ('\'', '"'): |
| 570 | raise self._ParseError('Exptected string.') |
| 571 | |
| 572 | if len(text) < 2 or text[-1] != text[0]: |
| 573 | raise self._ParseError('String missing ending quote.') |
| 574 | |
| 575 | try: |
| 576 | result = _CUnescape(text[1:-1]) |
| 577 | except ValueError, e: |
| 578 | raise self._ParseError(str(e)) |
| 579 | self.NextToken() |
| 580 | return result |
| 581 | |
| 582 | def _ParseInteger(self, text, is_signed=False, is_long=False): |
| 583 | """Parses an integer. |
| 584 | |
| 585 | Args: |
| 586 | text: The text to parse. |
| 587 | is_signed: True if a signed integer must be parsed. |
| 588 | is_long: True if a long integer must be parsed. |
| 589 | |
| 590 | Returns: |
| 591 | The integer value. |
| 592 | |
| 593 | Raises: |
| 594 | ValueError: Thrown Iff the text is not a valid integer. |
| 595 | """ |
| 596 | pos = 0 |
| 597 | if text.startswith('-'): |
| 598 | pos += 1 |
| 599 | |
| 600 | base = 10 |
| 601 | if text.startswith('0x', pos) or text.startswith('0X', pos): |
| 602 | base = 16 |
| 603 | elif text.startswith('0', pos): |
| 604 | base = 8 |
| 605 | |
| 606 | # Do the actual parsing. Exception handling is propagated to caller. |
| 607 | result = int(text, base) |
| 608 | |
| 609 | # Check if the integer is sane. Exceptions handled by callers. |
| 610 | checker = self._INTEGER_CHECKERS[2 * int(is_long) + int(is_signed)] |
| 611 | checker.CheckValue(result) |
| 612 | return result |
| 613 | |
| 614 | def ParseErrorPreviousToken(self, message): |
| 615 | """Creates and *returns* a ParseError for the previously read token. |
| 616 | |
| 617 | Args: |
| 618 | message: A message to set for the exception. |
| 619 | |
| 620 | Returns: |
| 621 | A ParseError instance. |
| 622 | """ |
| 623 | return ParseError('%d:%d : %s' % ( |
| 624 | self._previous_line + 1, self._previous_column + 1, message)) |
| 625 | |
| 626 | def _ParseError(self, message): |
| 627 | """Creates and *returns* a ParseError for the current token.""" |
| 628 | return ParseError('%d:%d : %s' % ( |
liujisi@google.com | 33165fe | 2010-11-02 13:14:58 +0000 | [diff] [blame^] | 629 | self._line + 1, self._column - len(self.token) + 1, message)) |
kenton@google.com | 80b1d62 | 2009-07-29 01:13:20 +0000 | [diff] [blame] | 630 | |
| 631 | def _IntegerParseError(self, e): |
| 632 | return self._ParseError('Couldn\'t parse integer: ' + str(e)) |
| 633 | |
| 634 | def _FloatParseError(self, e): |
| 635 | return self._ParseError('Couldn\'t parse number: ' + str(e)) |
| 636 | |
liujisi@google.com | 33165fe | 2010-11-02 13:14:58 +0000 | [diff] [blame^] | 637 | def _StringParseError(self, e): |
| 638 | return self._ParseError('Couldn\'t parse string: ' + str(e)) |
| 639 | |
kenton@google.com | 80b1d62 | 2009-07-29 01:13:20 +0000 | [diff] [blame] | 640 | def NextToken(self): |
| 641 | """Reads the next meaningful token.""" |
| 642 | self._previous_line = self._line |
| 643 | self._previous_column = self._column |
liujisi@google.com | 33165fe | 2010-11-02 13:14:58 +0000 | [diff] [blame^] | 644 | |
| 645 | self._column += len(self.token) |
| 646 | self._SkipWhitespace() |
| 647 | |
| 648 | if not self._lines and len(self._current_line) <= self._column: |
kenton@google.com | 80b1d62 | 2009-07-29 01:13:20 +0000 | [diff] [blame] | 649 | self.token = '' |
| 650 | return |
kenton@google.com | 80b1d62 | 2009-07-29 01:13:20 +0000 | [diff] [blame] | 651 | |
liujisi@google.com | 33165fe | 2010-11-02 13:14:58 +0000 | [diff] [blame^] | 652 | match = self._TOKEN.match(self._current_line, self._column) |
kenton@google.com | 80b1d62 | 2009-07-29 01:13:20 +0000 | [diff] [blame] | 653 | if match: |
| 654 | token = match.group(0) |
kenton@google.com | 80b1d62 | 2009-07-29 01:13:20 +0000 | [diff] [blame] | 655 | self.token = token |
| 656 | else: |
liujisi@google.com | 33165fe | 2010-11-02 13:14:58 +0000 | [diff] [blame^] | 657 | self.token = self._current_line[self._column] |
kenton@google.com | 80b1d62 | 2009-07-29 01:13:20 +0000 | [diff] [blame] | 658 | |
| 659 | |
temporal | 40ee551 | 2008-07-10 02:12:20 +0000 | [diff] [blame] | 660 | # text.encode('string_escape') does not seem to satisfy our needs as it |
| 661 | # encodes unprintable characters using two-digit hex escapes whereas our |
| 662 | # C++ unescaping function allows hex escapes to be any length. So, |
| 663 | # "\0011".encode('string_escape') ends up being "\\x011", which will be |
| 664 | # decoded in C++ as a single-character string with char code 0x11. |
liujisi@google.com | 33165fe | 2010-11-02 13:14:58 +0000 | [diff] [blame^] | 665 | def _CEscape(text, as_utf8): |
temporal | 40ee551 | 2008-07-10 02:12:20 +0000 | [diff] [blame] | 666 | def escape(c): |
| 667 | o = ord(c) |
| 668 | if o == 10: return r"\n" # optional escape |
| 669 | if o == 13: return r"\r" # optional escape |
| 670 | if o == 9: return r"\t" # optional escape |
| 671 | if o == 39: return r"\'" # optional escape |
| 672 | |
| 673 | if o == 34: return r'\"' # necessary escape |
| 674 | if o == 92: return r"\\" # necessary escape |
| 675 | |
liujisi@google.com | 33165fe | 2010-11-02 13:14:58 +0000 | [diff] [blame^] | 676 | # necessary escapes |
| 677 | if not as_utf8 and (o >= 127 or o < 32): return "\\%03o" % o |
temporal | 40ee551 | 2008-07-10 02:12:20 +0000 | [diff] [blame] | 678 | return c |
| 679 | return "".join([escape(c) for c in text]) |
kenton@google.com | 80b1d62 | 2009-07-29 01:13:20 +0000 | [diff] [blame] | 680 | |
| 681 | |
liujisi@google.com | 33165fe | 2010-11-02 13:14:58 +0000 | [diff] [blame^] | 682 | _CUNESCAPE_HEX = re.compile('\\\\x([0-9a-fA-F]{2}|[0-9a-fA-F])') |
kenton@google.com | 80b1d62 | 2009-07-29 01:13:20 +0000 | [diff] [blame] | 683 | |
| 684 | |
| 685 | def _CUnescape(text): |
| 686 | def ReplaceHex(m): |
| 687 | return chr(int(m.group(0)[2:], 16)) |
| 688 | # This is required because the 'string_escape' encoding doesn't |
| 689 | # allow single-digit hex escapes (like '\xf'). |
| 690 | result = _CUNESCAPE_HEX.sub(ReplaceHex, text) |
| 691 | return result.decode('string_escape') |