2 * Copyright 2014 Facebook, Inc.
4 * Licensed under the Apache License, Version 2.0 (the "License");
5 * you may not use this file except in compliance with the License.
6 * You may obtain a copy of the License at
8 * http://www.apache.org/licenses/LICENSE-2.0
10 * Unless required by applicable law or agreed to in writing, software
11 * distributed under the License is distributed on an "AS IS" BASIS,
12 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
13 * See the License for the specific language governing permissions and
14 * limitations under the License.
17 #include "folly/Uri.h"
20 #include <boost/regex.hpp>
26 fbstring submatch(const boost::cmatch& m, size_t idx) {
28 return fbstring(sub.first, sub.second);
31 template <class String>
32 void toLower(String& s) {
40 Uri::Uri(StringPiece str) : port_(0) {
41 static const boost::regex uriRegex(
42 "([a-zA-Z][a-zA-Z0-9+.-]*):" // scheme:
43 "([^?#]*)" // authority and path
44 "(?:\\?([^#]*))?" // ?query
45 "(?:#(.*))?"); // #fragment
46 static const boost::regex authorityAndPathRegex("//([^/]*)(/.*)?");
49 if (UNLIKELY(!boost::regex_match(str.begin(), str.end(), match, uriRegex))) {
50 throw std::invalid_argument(to<std::string>("invalid URI ", str));
53 scheme_ = submatch(match, 1);
56 StringPiece authorityAndPath(match[2].first, match[2].second);
57 boost::cmatch authorityAndPathMatch;
58 if (!boost::regex_match(authorityAndPath.begin(),
59 authorityAndPath.end(),
60 authorityAndPathMatch,
61 authorityAndPathRegex)) {
62 // Does not start with //, doesn't have authority
63 path_ = authorityAndPath.fbstr();
65 static const boost::regex authorityRegex(
66 "(?:([^@:]*)(?::([^@]*))?@)?" // username, password
67 "(\\[[^\\]]*\\]|[^\\[:]*)" // host (IP-literal (e.g. '['+IPv6+']',
68 // dotted-IPv4, or named host)
69 "(?::(\\d*))?"); // port
71 auto authority = authorityAndPathMatch[1];
72 boost::cmatch authorityMatch;
73 if (!boost::regex_match(authority.first,
77 throw std::invalid_argument(
78 to<std::string>("invalid URI authority ",
79 StringPiece(authority.first, authority.second)));
82 StringPiece port(authorityMatch[4].first, authorityMatch[4].second);
84 port_ = to<uint16_t>(port);
87 username_ = submatch(authorityMatch, 1);
88 password_ = submatch(authorityMatch, 2);
89 host_ = submatch(authorityMatch, 3);
90 path_ = submatch(authorityAndPathMatch, 2);
93 query_ = submatch(match, 3);
94 fragment_ = submatch(match, 4);
97 fbstring Uri::authority() const {
100 // Port is 5 characters max and we have up to 3 delimiters.
101 result.reserve(host().size() + username().size() + password().size() + 8);
103 if (!username().empty() || !password().empty()) {
104 result.append(username());
106 if (!password().empty()) {
107 result.push_back(':');
108 result.append(password());
111 result.push_back('@');
114 result.append(host());
117 result.push_back(':');
118 toAppend(port(), &result);
124 fbstring Uri::hostname() const {
125 if (host_.size() > 0 && host_[0] == '[') {
126 // If it starts with '[', then it should end with ']', this is ensured by
128 return host_.substr(1, host_.size() - 2);