1 // src/parse_http_request.cc
2 // This file is part of libpbe; see http://anyterm.org/
3 // (C) 2005-2008 Philip Endecott
4
5 // This program is free software; you can redistribute it and/or modify
6 // it under the terms of the GNU General Public License as published by
7 // the Free Software Foundation; either version 2 of the License, or
8 // any later version.
9 //
10 // This program is distributed in the hope that it will be useful,
11 // but WITHOUT ANY WARRANTY; without even the implied warranty of
12 // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
13 // GNU General Public License for more details.
14 //
15 // You should have received a copy of the GNU General Public License
16 // along with this program; if not, write to the Free Software
17 // Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
18
19 #include "parse_http_request.hh"
20
21 #include <boost/spirit/include/classic.hpp>
22 #include <boost/spirit/home/classic/actor/insert_at_actor.hpp>
23 #include <boost/spirit/home/classic/dynamic/if.hpp>
24 #include <boost/spirit/home/classic/utility/loops.hpp>
25 #include <boost/spirit/home/classic/iterator/multi_pass.hpp>
26
27 #include <boost/lexical_cast.hpp>
28 #include <boost/scoped_array.hpp>
29
30 #include <map>
31
32
33 using namespace std;
34 using namespace boost::spirit::classic;
35
36
37 namespace pbe {
38
39
parse_http_request(istream & strm)40 HttpRequest parse_http_request(istream& strm)
41 {
42 HttpRequest request;
43
44 typedef multi_pass<istreambuf_iterator<char> > iterator_t;
45 typedef scanner<iterator_t> scanner_t;
46 typedef rule<scanner_t> rule_t;
47
48 string tmp_hn;
49 string tmp_hv;
50
51 // HTTP/1.1 request parsing, based on:
52 // RFC2616
53 // RFC2396
54 // HTTP/1.1 Errata (http://skrb.org/ietf/http_errata.html)
55
56 rule_t CRLF = str_p("\r\n");
57
58 rule_t extension_method = nothing_p;
59
60 rule_t method = str_p("OPTIONS") | "GET" | "HEAD" | "POST" | "PUT"
61 | "DELETE" | "TRACE" | "CONNECT" | extension_method;
62
63 rule_t mark = ch_p('-') | '_' | '.' | '!' | '~' | '*' | '\'' | '(' | ')';
64
65 rule_t unreserved = alnum_p | mark;
66
67 rule_t escaped = ch_p('%') >> xdigit_p >> xdigit_p;
68
69 rule_t reserved = ch_p(';') | '/' | '?' | ':' | '@' | '&' | '=' | '+' | '$' | ',';
70
71 rule_t pchar = unreserved | escaped | ':' | '@' | '&' | '=' | '+' | '$' | ',';
72
73 rule_t param = *pchar;
74
75 rule_t segment = *pchar >> *(';' >> param);
76
77 rule_t path_segments = segment >> *('/' >> segment);
78
79 rule_t abs_path = ( ch_p('/') >> path_segments )[assign_a(request.abs_path)];
80
81 rule_t scheme = alpha_p >> *(alpha_p | digit_p | '+' | '-' | '.' );
82
83 rule_t userinfo = *(unreserved | escaped | ';' | ':' | '&' | '=' | '+' | '$' | ',' );
84
85 rule_t domainlabel = alnum_p | alnum_p >> *(alnum_p | '-') >> alnum_p;
86
87 rule_t toplabel = alpha_p | alpha_p >> *(alnum_p | '-') >> alnum_p;
88
89 rule_t hostname = *(domainlabel >> '.') >> toplabel >> !ch_p('.');
90
91 uint_parser<unsigned,10,1,3> decimal_byte;
92
93 rule_t ipv4address = decimal_byte >> '.' >> decimal_byte >> '.' >>
94 decimal_byte >> '.' >> decimal_byte;
95
96 rule_t host = hostname | ipv4address;
97
98 rule_t port = uint_p;
99
100 rule_t hostport = host >> !(':' >> port);
101
102 rule_t server = !( !(userinfo >> '@') >> hostport );
103
104 rule_t reg_name = +(unreserved | escaped | '$' | ',' | ';' | ':' | '@' |
105 '&' | '=' | '+');
106
107 rule_t authority = server | reg_name;
108
109 rule_t net_path = str_p("//") >> authority >> !abs_path;
110
111 rule_t uric = reserved | unreserved | escaped;
112
113 rule_t query = (*uric) [assign_a(request.query)];
114
115 rule_t hier_part = (net_path | abs_path) >> !('?' >> query);
116
117 rule_t uric_no_slash = unreserved | escaped | ';' | '?' | ':' | '@' |
118 '&' | '=' | '+' | '$' | ',';
119
120 rule_t opaque_part = uric_no_slash >> *uric;
121
122 rule_t absolute_uri = scheme >> ':' >> (hier_part | opaque_part);
123
124 rule_t request_uri = ch_p('*')
125 | absolute_uri
126 | (abs_path >> !(ch_p('?') >> query))
127 | authority;
128
129 rule_t http_version = str_p("HTTP/") >> uint_p >> '.' >> uint_p;
130
131 rule_t request_line =
132 method [assign_a(request.method)]
133 >> ' ' >> request_uri
134 >> ' ' >> http_version [assign_a(request.http_version)]
135 >> CRLF ;
136
137 rule_t header_name = +(alnum_p|'-');
138
139 rule_t header_value = *(print_p|' '|'\t');
140
141 rule_t header = ( header_name [assign_a(tmp_hn)] >>
142 ':' >> *(ch_p(' ')) >> header_value [assign_a(tmp_hv)] )
143 [insert_at_a(request.headers,tmp_hn,tmp_hv)];
144
145 rule_t request_r =
146 request_line
147 >> *(header >> CRLF)
148 >> CRLF;
149
150 iterator_t first(make_multi_pass(std::istreambuf_iterator<char>(strm)));
151 iterator_t last(make_multi_pass(std::istreambuf_iterator<char>()));
152
153 scanner_t scanner(first,last);
154
155 if (!request_r.parse(scanner)) {
156 throw HttpRequestSyntaxError();
157 }
158
159 HttpRequest::headers_t::const_iterator i = request.headers.find("Content-Length");
160 if (i==request.headers.end()) {
161 return request;
162 }
163 size_t content_length = boost::lexical_cast<size_t>(i->second);
164 boost::scoped_array<char> buf(new char[content_length]);
165 strm.read(buf.get(),content_length);
166 request.body = string(buf.get(),content_length);
167
168 return request;
169 }
170
171
172 };
173
174