1 #!/usr/bin/env python 2 3 """ 4 Request helper classes. 5 6 Copyright (C) 2004, 2005, 2006 Paul Boddie <paul@boddie.org.uk> 7 8 This library is free software; you can redistribute it and/or 9 modify it under the terms of the GNU Lesser General Public 10 License as published by the Free Software Foundation; either 11 version 2.1 of the License, or (at your option) any later version. 12 13 This library is distributed in the hope that it will be useful, 14 but WITHOUT ANY WARRANTY; without even the implied warranty of 15 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU 16 Lesser General Public License for more details. 17 18 You should have received a copy of the GNU Lesser General Public 19 License along with this library; if not, write to the Free Software 20 Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA 21 """ 22 23 class MessageBodyStream: 24 25 """ 26 A naive stream class, providing a non-blocking stream for transactions when 27 reading the message body. According to the HTTP standard, the following 28 things decide how long the message is: 29 30 * Use of the Content-Length header field (see 4.4 Message Length). 31 * Use of the Transfer-Coding header field (see 3.6 Transfer Codings), 32 particularly when the "chunked" coding is used. 33 34 NOTE: For now, we don't support the Transfer-Coding business. 35 """ 36 37 def __init__(self, stream, headers): 38 39 """ 40 Initialise the object with the given underlying 'stream'. The supplied 41 'headers' in a dictionary-style object are used to examine the nature of 42 the request. 43 """ 44 45 self.stream = stream 46 self.headers = headers 47 self.length = int(headers.get("Content-Length") or 0) 48 49 def read(self, limit=None): 50 51 "Reads all remaining data from the message body." 52 53 if limit is not None: 54 limit = min(limit, self.length) 55 else: 56 limit = self.length 57 data = self.stream.read(limit) 58 self.length = self.length - len(data) 59 return data 60 61 def readline(self, n=None): 62 63 "Reads a single line of data from the message body." 64 65 data = [] 66 while self.length > 0: 67 data.append(self.read(1)) 68 if data[-1] == "\n": 69 break 70 return "".join(data) 71 72 def readlines(self): 73 74 """ 75 Reads all remaining data from the message body, splitting it into lines 76 and returning the data as a list of lines. 77 """ 78 79 lines = self.read().split("\n") 80 for i in range(0, len(lines) - 1): 81 lines[i] = lines[i] + "\n" 82 return lines 83 84 def close(self): 85 86 "Closes the stream." 87 88 self.stream.close() 89 90 class HeaderValue: 91 92 "A container for header information." 93 94 def __init__(self, principal_value, **attributes): 95 96 """ 97 Initialise the container with the given 'principal_value' and optional 98 keyword attributes representing the key=value pairs which accompany the 99 'principal_value'. 100 """ 101 102 self.principal_value = principal_value 103 self.attributes = attributes 104 105 def __getattr__(self, name): 106 if self.attributes.has_key(name): 107 return self.attributes[name] 108 else: 109 raise AttributeError, name 110 111 def __str__(self): 112 113 """ 114 Format the header value object, producing a string suitable for the 115 response header field. 116 """ 117 118 l = [] 119 if self.principal_value: 120 l.append(self.principal_value) 121 for name, value in self.attributes.items(): 122 l.append("; ") 123 l.append("%s=%s" % (name, value)) 124 125 # Make sure that only ASCII is used. 126 127 return "".join(l).encode("US-ASCII") 128 129 class ContentType(HeaderValue): 130 131 "A container for content type information." 132 133 def __init__(self, media_type, charset=None, **attributes): 134 135 """ 136 Initialise the container with the given 'media_type', an optional 137 'charset', and optional keyword attributes representing the key=value 138 pairs which qualify content types. 139 """ 140 141 if charset is not None: 142 attributes["charset"] = charset 143 HeaderValue.__init__(self, media_type, **attributes) 144 145 def __getattr__(self, name): 146 if name == "media_type": 147 return self.principal_value 148 elif name == "charset": 149 return self.attributes.get("charset") 150 elif self.attributes.has_key(name): 151 return self.attributes[name] 152 else: 153 raise AttributeError, name 154 155 class Cookie: 156 157 """ 158 A simple cookie class for frameworks which do not return cookies in 159 structured form. Instances of this class contain the following attributes: 160 161 * name - the name associated with the cookie 162 * value - the value retained by the cookie 163 """ 164 165 def __init__(self, name, value): 166 self.name = name 167 self.value = value 168 169 class FileContent: 170 171 """ 172 A simple class representing uploaded file content. This is useful in holding 173 metadata as well as being an indicator of such content in environments such 174 as Jython where it is not trivial to differentiate between plain strings and 175 Unicode in a fashion also applicable to CPython. 176 177 Instances of this class contain the following attributes: 178 179 * stream - a stream object through which the content of an uploaded file 180 may be accessed 181 * content - a plain string containing the contents of the uploaded file 182 * headers - a dictionary containing the headers associated with the 183 uploaded file 184 """ 185 186 def __init__(self, stream, headers=None): 187 188 """ 189 Initialise the object with a 'stream' through which the file can be 190 read, along with optional 'headers' describing the content. 191 """ 192 193 self.stream = stream 194 self.headers = headers or {} 195 self.cache = None 196 197 def __getattr__(self, name): 198 if name != "content": 199 raise AttributeError, name 200 if self.cache is not None: 201 return self.cache 202 if hasattr(self.stream, "seek"): 203 self.stream.seek(0) 204 return self.stream.read() 205 else: 206 self.cache = self.stream.read() 207 return self.cache 208 209 def __str__(self): 210 return self.content 211 212 def parse_header_value(header_class, header_value_str): 213 214 """ 215 Create an object of the given 'header_class' by determining the details 216 of the given 'header_value_str' - a string containing the value of a 217 particular header. 218 """ 219 220 if header_value_str is None: 221 return header_class(None) 222 223 l = header_value_str.split(";") 224 attributes = {} 225 226 # Find the attributes. 227 228 principal_value, attributes_str = l[0].strip(), l[1:] 229 230 for attribute_str in attributes_str: 231 t = attribute_str.split("=") 232 if len(t) > 1: 233 name, value = t[0].strip(), t[1].strip() 234 attributes[name] = value 235 236 return header_class(principal_value, **attributes) 237 238 def parse_headers(headers): 239 240 """ 241 Parse the given 'headers' dictionary (containing names mapped to values), 242 returing a dictionary mapping names to HeaderValue objects. 243 """ 244 245 new_headers = {} 246 for name, value in headers.items(): 247 new_headers[name] = parse_header_value(HeaderValue, value) 248 return new_headers 249 250 def get_storage_items(storage_body): 251 252 """ 253 Return the items (2-tuples of the form key, values) from the 'storage_body'. 254 This is used in conjunction with FieldStorage objects. 255 """ 256 257 items = [] 258 for key in storage_body.keys(): 259 items.append((key, storage_body[key])) 260 return items 261 262 def get_body_fields(field_items, encoding): 263 264 """ 265 Returns a dictionary mapping field names to lists of field values for all 266 entries in the given 'field_items' (2-tuples of the form key, values) using 267 the given 'encoding'. 268 This is used in conjunction with FieldStorage objects. 269 """ 270 271 fields = {} 272 273 for field_name, field_values in field_items: 274 field_name = decode_value(field_name, encoding) 275 276 if type(field_values) == type([]): 277 fields[field_name] = [] 278 for field_value in field_values: 279 fields[field_name].append(get_body_field_or_file(field_value, encoding)) 280 else: 281 fields[field_name] = [get_body_field_or_file(field_values, encoding)] 282 283 return fields 284 285 def get_body_field_or_file(field_value, encoding): 286 287 """ 288 Returns the appropriate value for the given 'field_value' either for a 289 normal form field (thus employing the given 'encoding') or for a file 290 upload field (returning a plain string). 291 """ 292 293 if hasattr(field_value, "headers") and field_value.headers.has_key("content-type"): 294 295 # Detect stray FileUpload objects (eg. with Zope). 296 297 if hasattr(field_value, "read"): 298 return FileContent(field_value, parse_headers(field_value.headers)) 299 else: 300 return FileContent(field_value.file, parse_headers(field_value.headers)) 301 else: 302 return get_body_field(field_value, encoding) 303 304 def get_body_field(field_str, encoding): 305 306 """ 307 Returns the appropriate value for the given 'field_str' string using the 308 given 'encoding'. 309 """ 310 311 # Detect stray FieldStorage objects (eg. with Webware). 312 313 if hasattr(field_str, "value"): 314 return get_body_field(field_str.value, encoding) 315 else: 316 return decode_value(field_str, encoding) 317 318 def decode_value(s, encoding): 319 if encoding is not None: 320 try: 321 return unicode(s, encoding) 322 except UnicodeError: 323 pass 324 # NOTE: Hacks to permit graceful failure. 325 return unicode(s, "iso-8859-1") 326 327 def get_fields_from_query_string(query_string, decoder): 328 329 """ 330 Returns a dictionary mapping field names to lists of values for the data 331 encoded in the given 'query_string'. Use the given 'decoder' function or 332 method to process the URL-encoded values. 333 """ 334 335 fields = {} 336 337 for pair in query_string.split("&"): 338 t = pair.split("=") 339 name = decoder(t[0]) 340 341 if len(t) == 2: 342 value = decoder(t[1]) 343 else: 344 value = "" 345 346 # NOTE: Remove empty names. 347 348 if name: 349 if not fields.has_key(name): 350 fields[name] = [] 351 fields[name].append(value) 352 353 return fields 354 355 def filter_fields(all_fields, fields_from_path): 356 357 """ 358 Taking items from the 'all_fields' dictionary, produce a new dictionary 359 which does not contain items from the 'fields_from_path' dictionary. 360 Return a new dictionary. 361 """ 362 363 fields = {} 364 for field_name, field_values in all_fields.items(): 365 366 # Find the path values for this field (for filtering below). 367 368 if fields_from_path.has_key(field_name): 369 field_from_path_values = fields_from_path[field_name] 370 if type(field_from_path_values) != type([]): 371 field_from_path_values = [field_from_path_values] 372 else: 373 field_from_path_values = [] 374 375 fields[field_name] = [] 376 for field_value in field_values: 377 378 # Filter path values. 379 380 if field_value not in field_from_path_values: 381 fields[field_name].append(field_value) 382 383 # Remove filtered fields. 384 385 if fields[field_name] == []: 386 del fields[field_name] 387 388 return fields 389 390 # vim: tabstop=4 expandtab shiftwidth=4