1.1 --- /dev/null Thu Jan 01 00:00:00 1970 +0000
1.2 +++ b/testing/mochitest/pywebsocket/mod_pywebsocket/http_header_util.py Wed Dec 31 06:09:35 2014 +0100
1.3 @@ -0,0 +1,263 @@
1.4 +# Copyright 2011, Google Inc.
1.5 +# All rights reserved.
1.6 +#
1.7 +# Redistribution and use in source and binary forms, with or without
1.8 +# modification, are permitted provided that the following conditions are
1.9 +# met:
1.10 +#
1.11 +# * Redistributions of source code must retain the above copyright
1.12 +# notice, this list of conditions and the following disclaimer.
1.13 +# * Redistributions in binary form must reproduce the above
1.14 +# copyright notice, this list of conditions and the following disclaimer
1.15 +# in the documentation and/or other materials provided with the
1.16 +# distribution.
1.17 +# * Neither the name of Google Inc. nor the names of its
1.18 +# contributors may be used to endorse or promote products derived from
1.19 +# this software without specific prior written permission.
1.20 +#
1.21 +# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
1.22 +# "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
1.23 +# LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
1.24 +# A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
1.25 +# OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
1.26 +# SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
1.27 +# LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
1.28 +# DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
1.29 +# THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
1.30 +# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
1.31 +# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
1.32 +
1.33 +
1.34 +"""Utilities for parsing and formatting headers that follow the grammar defined
1.35 +in HTTP RFC http://www.ietf.org/rfc/rfc2616.txt.
1.36 +"""
1.37 +
1.38 +
1.39 +import urlparse
1.40 +
1.41 +
1.42 +_SEPARATORS = '()<>@,;:\\"/[]?={} \t'
1.43 +
1.44 +
1.45 +def _is_char(c):
1.46 + """Returns true iff c is in CHAR as specified in HTTP RFC."""
1.47 +
1.48 + return ord(c) <= 127
1.49 +
1.50 +
1.51 +def _is_ctl(c):
1.52 + """Returns true iff c is in CTL as specified in HTTP RFC."""
1.53 +
1.54 + return ord(c) <= 31 or ord(c) == 127
1.55 +
1.56 +
1.57 +class ParsingState(object):
1.58 +
1.59 + def __init__(self, data):
1.60 + self.data = data
1.61 + self.head = 0
1.62 +
1.63 +
1.64 +def peek(state, pos=0):
1.65 + """Peeks the character at pos from the head of data."""
1.66 +
1.67 + if state.head + pos >= len(state.data):
1.68 + return None
1.69 +
1.70 + return state.data[state.head + pos]
1.71 +
1.72 +
1.73 +def consume(state, amount=1):
1.74 + """Consumes specified amount of bytes from the head and returns the
1.75 + consumed bytes. If there's not enough bytes to consume, returns None.
1.76 + """
1.77 +
1.78 + if state.head + amount > len(state.data):
1.79 + return None
1.80 +
1.81 + result = state.data[state.head:state.head + amount]
1.82 + state.head = state.head + amount
1.83 + return result
1.84 +
1.85 +
1.86 +def consume_string(state, expected):
1.87 + """Given a parsing state and a expected string, consumes the string from
1.88 + the head. Returns True if consumed successfully. Otherwise, returns
1.89 + False.
1.90 + """
1.91 +
1.92 + pos = 0
1.93 +
1.94 + for c in expected:
1.95 + if c != peek(state, pos):
1.96 + return False
1.97 + pos += 1
1.98 +
1.99 + consume(state, pos)
1.100 + return True
1.101 +
1.102 +
1.103 +def consume_lws(state):
1.104 + """Consumes a LWS from the head. Returns True if any LWS is consumed.
1.105 + Otherwise, returns False.
1.106 +
1.107 + LWS = [CRLF] 1*( SP | HT )
1.108 + """
1.109 +
1.110 + original_head = state.head
1.111 +
1.112 + consume_string(state, '\r\n')
1.113 +
1.114 + pos = 0
1.115 +
1.116 + while True:
1.117 + c = peek(state, pos)
1.118 + if c == ' ' or c == '\t':
1.119 + pos += 1
1.120 + else:
1.121 + if pos == 0:
1.122 + state.head = original_head
1.123 + return False
1.124 + else:
1.125 + consume(state, pos)
1.126 + return True
1.127 +
1.128 +
1.129 +def consume_lwses(state):
1.130 + """Consumes *LWS from the head."""
1.131 +
1.132 + while consume_lws(state):
1.133 + pass
1.134 +
1.135 +
1.136 +def consume_token(state):
1.137 + """Consumes a token from the head. Returns the token or None if no token
1.138 + was found.
1.139 + """
1.140 +
1.141 + pos = 0
1.142 +
1.143 + while True:
1.144 + c = peek(state, pos)
1.145 + if c is None or c in _SEPARATORS or _is_ctl(c) or not _is_char(c):
1.146 + if pos == 0:
1.147 + return None
1.148 +
1.149 + return consume(state, pos)
1.150 + else:
1.151 + pos += 1
1.152 +
1.153 +
1.154 +def consume_token_or_quoted_string(state):
1.155 + """Consumes a token or a quoted-string, and returns the token or unquoted
1.156 + string. If no token or quoted-string was found, returns None.
1.157 + """
1.158 +
1.159 + original_head = state.head
1.160 +
1.161 + if not consume_string(state, '"'):
1.162 + return consume_token(state)
1.163 +
1.164 + result = []
1.165 +
1.166 + expect_quoted_pair = False
1.167 +
1.168 + while True:
1.169 + if not expect_quoted_pair and consume_lws(state):
1.170 + result.append(' ')
1.171 + continue
1.172 +
1.173 + c = consume(state)
1.174 + if c is None:
1.175 + # quoted-string is not enclosed with double quotation
1.176 + state.head = original_head
1.177 + return None
1.178 + elif expect_quoted_pair:
1.179 + expect_quoted_pair = False
1.180 + if _is_char(c):
1.181 + result.append(c)
1.182 + else:
1.183 + # Non CHAR character found in quoted-pair
1.184 + state.head = original_head
1.185 + return None
1.186 + elif c == '\\':
1.187 + expect_quoted_pair = True
1.188 + elif c == '"':
1.189 + return ''.join(result)
1.190 + elif _is_ctl(c):
1.191 + # Invalid character %r found in qdtext
1.192 + state.head = original_head
1.193 + return None
1.194 + else:
1.195 + result.append(c)
1.196 +
1.197 +
1.198 +def quote_if_necessary(s):
1.199 + """Quotes arbitrary string into quoted-string."""
1.200 +
1.201 + quote = False
1.202 + if s == '':
1.203 + return '""'
1.204 +
1.205 + result = []
1.206 + for c in s:
1.207 + if c == '"' or c in _SEPARATORS or _is_ctl(c) or not _is_char(c):
1.208 + quote = True
1.209 +
1.210 + if c == '"' or _is_ctl(c):
1.211 + result.append('\\' + c)
1.212 + else:
1.213 + result.append(c)
1.214 +
1.215 + if quote:
1.216 + return '"' + ''.join(result) + '"'
1.217 + else:
1.218 + return ''.join(result)
1.219 +
1.220 +
1.221 +def parse_uri(uri):
1.222 + """Parse absolute URI then return host, port and resource."""
1.223 +
1.224 + parsed = urlparse.urlsplit(uri)
1.225 + if parsed.scheme != 'wss' and parsed.scheme != 'ws':
1.226 + # |uri| must be a relative URI.
1.227 + # TODO(toyoshim): Should validate |uri|.
1.228 + return None, None, uri
1.229 +
1.230 + if parsed.hostname is None:
1.231 + return None, None, None
1.232 +
1.233 + port = None
1.234 + try:
1.235 + port = parsed.port
1.236 + except ValueError, e:
1.237 + # port property cause ValueError on invalid null port description like
1.238 + # 'ws://host:/path'.
1.239 + return None, None, None
1.240 +
1.241 + if port is None:
1.242 + if parsed.scheme == 'ws':
1.243 + port = 80
1.244 + else:
1.245 + port = 443
1.246 +
1.247 + path = parsed.path
1.248 + if not path:
1.249 + path += '/'
1.250 + if parsed.query:
1.251 + path += '?' + parsed.query
1.252 + if parsed.fragment:
1.253 + path += '#' + parsed.fragment
1.254 +
1.255 + return parsed.hostname, port, path
1.256 +
1.257 +
1.258 +try:
1.259 + urlparse.uses_netloc.index('ws')
1.260 +except ValueError, e:
1.261 + # urlparse in Python2.5.1 doesn't have 'ws' and 'wss' entries.
1.262 + urlparse.uses_netloc.append('ws')
1.263 + urlparse.uses_netloc.append('wss')
1.264 +
1.265 +
1.266 +# vi:sts=4 sw=4 et
