pyffi.object

124 """This class is the base class for all file formats. It implements 125 a number of useful functions such as walking over directory trees 126 (:meth:`walkData`) and a default attribute naming function 127 (:meth:`name_attribute`). 128 It also implements the base class for representing file data 129 (:class:`FileFormat.Data`). 130 """ 131 132 RE_FILENAME = None 133 """Override this with a regular expression (the result of a ``re.compile`` 134 call) for the file extension of the format you are implementing. 135 """ 136 137 ARCHIVE_CLASSES = [] 138 """Override this with a list of archive formats that may contain 139 files of the format. 140 """ 141 142 # precompiled regular expressions, used in name_parts 143 144 _RE_NAME_SEP = re.compile('[_\W]+') 145 """Matches seperators for splitting names.""" 146 147 _RE_NAME_DIGITS = re.compile('([0-9]+)|([a-zA-Z]+)') 148 """Matches digits or characters for splitting names.""" 149 150 _RE_NAME_CAMEL = re.compile('([A-Z][a-z]*)|([a-z]+)') 151 """Finds components of camelCase and CamelCase names.""" 152 153 _RE_NAME_LC = re.compile('[a-z]') 154 """Matches a lower case character.""" 155 156 _RE_NAME_UC = re.compile('[A-Z]') 157 """Matches an upper case character.""" 158 159 # override this with the data instance for this format

160 - class Data(pyffi.utils.graph.GlobalNode):

161 """Base class for representing data in a particular format. 162 Override this class to implement reading and writing. 163 """ 164 165 _byte_order = '<' 166 """Set to '<' for little-endian, and '>' for big-endian.""" 167 168 version = None 169 """Version of the data.""" 170 171 user_version = None 172 """User version (additional version field) of the data.""" 173

174 - def inspect(self, stream):

175 """Quickly checks whether the stream appears to contain 176 data of a particular format. Resets stream to original position. 177 Call this function if you simply wish to check that a file is 178 of a particular format without having to parse it completely. 179 180 Override this method. 181 182 :param stream: The file to inspect. 183 :type stream: file 184 :return: ``True`` if stream is of particular format, ``False`` 185 otherwise. 186 """ 187 raise NotImplementedError

188

189 - def read(self, stream):

190 """Read data of particular format from stream. 191 Override this method. 192 193 :param stream: The file to read from. 194 :type stream: ``file`` 195 """ 196 raise NotImplementedError

197

198 - def write(self, stream):

199 """Write data of particular format to stream. 200 Override this method. 201 202 :param stream: The file to write to. 203 :type stream: ``file`` 204 """ 205 raise NotImplementedError

206 207 @staticmethod

208 - def version_number(version_str):

209 """Converts version string into an integer. 210 This default implementation simply returns zero at all times, 211 and works for formats that are not versioned. 212 213 Override for versioned formats. 214 215 :param version_str: The version string. 216 :type version_str: ``str`` 217 :return: A version integer. 218 """ 219 return 0

220 221 @classmethod

222 - def name_parts(cls, name):

223 """Intelligently split a name into parts: 224 225 * first, split at non-alphanumeric characters 226 * next, seperate digits from characters 227 * finally, if some part has mixed case, it must be 228 camel case so split it further at upper case characters 229 230 >>> FileFormat.name_parts("hello_world") 231 ['hello', 'world'] 232 >>> FileFormat.name_parts("HELLO_WORLD") 233 ['HELLO', 'WORLD'] 234 >>> FileFormat.name_parts("HelloWorld") 235 ['Hello', 'World'] 236 >>> FileFormat.name_parts("helloWorld") 237 ['hello', 'World'] 238 >>> FileFormat.name_parts("xs:NMTOKEN") 239 ['xs', 'NMTOKEN'] 240 >>> FileFormat.name_parts("xs:NCName") 241 ['xs', 'N', 'C', 'Name'] 242 >>> FileFormat.name_parts('this IS a sillyNAME') 243 ['this', 'IS', 'a', 'silly', 'N', 'A', 'M', 'E'] 244 >>> FileFormat.name_parts('tHis is A Silly naME') 245 ['t', 'His', 'is', 'A', 'Silly', 'na', 'M', 'E'] 246 """ 247 # str(name) converts name to string in case it is a py2k 248 # unicode string 249 name = str(name) 250 # separate at symbols 251 parts = cls._RE_NAME_SEP.split(name) 252 # seperate digits 253 newparts = [] 254 for part in parts: 255 for part_groups in cls._RE_NAME_DIGITS.findall(part): 256 for group in part_groups: 257 if group: 258 newparts.append(group) 259 break 260 parts = newparts 261 # separate at upper case characters for CamelCase and camelCase words 262 newparts = [] 263 for part in parts: 264 if cls._RE_NAME_LC.search(part) and cls._RE_NAME_UC.search(part): 265 # find the camel bumps 266 for part_groups in cls._RE_NAME_CAMEL.findall(part): 267 for group in part_groups: 268 if group: 269 newparts.append(group) 270 break 271 else: 272 newparts.append(part) 273 parts = newparts 274 # return result 275 return parts

276 277 @classmethod

278 - def name_attribute(cls, name):

279 """Converts an attribute name, as in the description file, 280 into a name usable by python. 281 282 :param name: The attribute name. 283 :type name: ``str`` 284 :return: Reformatted attribute name, useable by python. 285 286 >>> FileFormat.name_attribute('tHis is A Silly naME') 287 't_his_is_a_silly_na_m_e' 288 >>> FileFormat.name_attribute('Test:Something') 289 'test_something' 290 >>> FileFormat.name_attribute('unknown?') 291 'unknown' 292 """ 293 return '_'.join(part.lower() for part in cls.name_parts(name))

294 295 @classmethod

296 - def name_class(cls, name):

297 """Converts a class name, as in the xsd file, into a name usable 298 by python. 299 300 :param name: The class name. 301 :type name: str 302 :return: Reformatted class name, useable by python. 303 304 >>> FileFormat.name_class('this IS a sillyNAME') 305 'ThisIsASillyNAME' 306 """ 307 return ''.join(part.capitalize() 308 for part in cls.name_parts(name))

309 310 @classmethod

311 - def walkData(cls, top, topdown=True, mode='rb'):

312 """A generator which yields the data of all files in 313 directory top whose filename matches the regular expression 314 :attr:`RE_FILENAME`. The argument top can also be a file instead of a 315 directory. Errors coming from os.walk are ignored. 316 317 Note that the caller is not responsible for closing the stream. 318 319 This function is for instance used by :mod:`pyffi.spells` to implement 320 modifying a file after reading and parsing. 321 322 :param top: The top folder. 323 :type top: ``str`` 324 :param topdown: Determines whether subdirectories should be iterated 325 over first. 326 :type topdown: ``bool`` 327 :param mode: The mode in which to open files. 328 :type mode: ``str`` 329 """ 330 # now walk over all these files in directory top 331 for filename in pyffi.utils.walk(top, topdown, onerror=None, 332 re_filename=cls.RE_FILENAME): 333 stream = open(filename, mode) 334 try: 335 # return data for the stream 336 # the caller can call data.read(stream), 337 # or data.inspect(stream), etc. 338 yield stream, cls.Data() 339 finally: 340 stream.close()

341 342 @classmethod

343 - def walk(cls, top, topdown=True, mode='rb'):

344 """A generator which yields all files in 345 directory top whose filename matches the regular expression 346 :attr:`RE_FILENAME`. The argument top can also be a file instead of a 347 directory. Errors coming from os.walk are ignored. 348 349 Note that the caller is not responsible for closing the stream. 350 351 This function is for instance used by :mod:`pyffi.spells` to implement 352 modifying a file after reading and parsing. 353 354 :param top: The top folder. 355 :type top: ``str`` 356 :param topdown: Determines whether subdirectories should be iterated 357 over first. 358 :type topdown: ``bool`` 359 :param mode: The mode in which to open files. 360 :type mode: ``str`` 361 """ 362 # now walk over all these files in directory top 363 for filename in pyffi.utils.walk(top, topdown, onerror=None, 364 re_filename=cls.RE_FILENAME): 365 stream = open(filename, mode) 366 try: 367 yield stream 368 finally: 369 stream.close()

Source Code for Package pyffi.object_models