1// Copyright 2007, Google Inc. 2// All rights reserved. 3// 4// Redistribution and use in source and binary forms, with or without 5// modification, are permitted provided that the following conditions are 6// met: 7// 8// * Redistributions of source code must retain the above copyright 9// notice, this list of conditions and the following disclaimer. 10// * Redistributions in binary form must reproduce the above 11// copyright notice, this list of conditions and the following disclaimer 12// in the documentation and/or other materials provided with the 13// distribution. 14// * Neither the name of Google Inc. nor the names of its 15// contributors may be used to endorse or promote products derived from 16// this software without specific prior written permission. 17// 18// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS 19// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT 20// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR 21// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT 22// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, 23// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT 24// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, 25// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY 26// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT 27// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE 28// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 29 30// Functions to canonicalize "standard" URLs, which are ones that have an 31// authority section including a host name. 32 33#include "googleurl/src/url_canon.h" 34#include "googleurl/src/url_canon_internal.h" 35 36namespace url_canon { 37 38namespace { 39 40template<typename CHAR, typename UCHAR> 41bool DoCanonicalizeStandardURL(const URLComponentSource<CHAR>& source, 42 const url_parse::Parsed& parsed, 43 CharsetConverter* query_converter, 44 CanonOutput* output, 45 url_parse::Parsed* new_parsed) { 46 // Scheme: this will append the colon. 47 bool success = CanonicalizeScheme(source.scheme, parsed.scheme, 48 output, &new_parsed->scheme); 49 50 // Authority (username, password, host, port) 51 bool have_authority; 52 if (parsed.username.is_valid() || parsed.password.is_valid() || 53 parsed.host.is_nonempty() || parsed.port.is_valid()) { 54 have_authority = true; 55 56 // Only write the authority separators when we have a scheme. 57 if (parsed.scheme.is_valid()) { 58 output->push_back('/'); 59 output->push_back('/'); 60 } 61 62 // User info: the canonicalizer will handle the : and @. 63 success &= CanonicalizeUserInfo(source.username, parsed.username, 64 source.password, parsed.password, 65 output, 66 &new_parsed->username, 67 &new_parsed->password); 68 69 success &= CanonicalizeHost(source.host, parsed.host, 70 output, &new_parsed->host); 71 72 // Host must not be empty for standard URLs. 73 if (!parsed.host.is_nonempty()) 74 success = false; 75 76 // Port: the port canonicalizer will handle the colon. 77 int default_port = DefaultPortForScheme( 78 &output->data()[new_parsed->scheme.begin], new_parsed->scheme.len); 79 success &= CanonicalizePort(source.port, parsed.port, default_port, 80 output, &new_parsed->port); 81 } else { 82 // No authority, clear the components. 83 have_authority = false; 84 new_parsed->host.reset(); 85 new_parsed->username.reset(); 86 new_parsed->password.reset(); 87 new_parsed->port.reset(); 88 success = false; // Standard URLs must have an authority. 89 } 90 91 // Path 92 if (parsed.path.is_valid()) { 93 success &= CanonicalizePath(source.path, parsed.path, 94 output, &new_parsed->path); 95 } else if (have_authority || 96 parsed.query.is_valid() || parsed.ref.is_valid()) { 97 // When we have an empty path, make up a path when we have an authority 98 // or something following the path. The only time we allow an empty 99 // output path is when there is nothing else. 100 new_parsed->path = url_parse::Component(output->length(), 1); 101 output->push_back('/'); 102 } else { 103 // No path at all 104 new_parsed->path.reset(); 105 } 106 107 // Query 108 CanonicalizeQuery(source.query, parsed.query, query_converter, 109 output, &new_parsed->query); 110 111 // Ref: ignore failure for this, since the page can probably still be loaded. 112 CanonicalizeRef(source.ref, parsed.ref, output, &new_parsed->ref); 113 114 return success; 115} 116 117} // namespace 118 119 120// Returns the default port for the given canonical scheme, or PORT_UNSPECIFIED 121// if the scheme is unknown. 122int DefaultPortForScheme(const char* scheme, int scheme_len) { 123 int default_port = url_parse::PORT_UNSPECIFIED; 124 switch (scheme_len) { 125 case 4: 126 if (!strncmp(scheme, "http", scheme_len)) 127 default_port = 80; 128 break; 129 case 5: 130 if (!strncmp(scheme, "https", scheme_len)) 131 default_port = 443; 132 break; 133 case 3: 134 if (!strncmp(scheme, "ftp", scheme_len)) 135 default_port = 21; 136 else if (!strncmp(scheme, "wss", scheme_len)) 137 default_port = 443; 138 break; 139 case 6: 140 if (!strncmp(scheme, "gopher", scheme_len)) 141 default_port = 70; 142 break; 143 case 2: 144 if (!strncmp(scheme, "ws", scheme_len)) 145 default_port = 80; 146 break; 147 } 148 return default_port; 149} 150 151bool CanonicalizeStandardURL(const char* spec, 152 int spec_len, 153 const url_parse::Parsed& parsed, 154 CharsetConverter* query_converter, 155 CanonOutput* output, 156 url_parse::Parsed* new_parsed) { 157 return DoCanonicalizeStandardURL<char, unsigned char>( 158 URLComponentSource<char>(spec), parsed, query_converter, 159 output, new_parsed); 160} 161 162bool CanonicalizeStandardURL(const char16* spec, 163 int spec_len, 164 const url_parse::Parsed& parsed, 165 CharsetConverter* query_converter, 166 CanonOutput* output, 167 url_parse::Parsed* new_parsed) { 168 return DoCanonicalizeStandardURL<char16, char16>( 169 URLComponentSource<char16>(spec), parsed, query_converter, 170 output, new_parsed); 171} 172 173// It might be nice in the future to optimize this so unchanged components don't 174// need to be recanonicalized. This is especially true since the common case for 175// ReplaceComponents is removing things we don't want, like reference fragments 176// and usernames. These cases can become more efficient if we can assume the 177// rest of the URL is OK with these removed (or only the modified parts 178// recanonicalized). This would be much more complex to implement, however. 179// 180// You would also need to update DoReplaceComponents in url_util.cc which 181// relies on this re-checking everything (see the comment there for why). 182bool ReplaceStandardURL(const char* base, 183 const url_parse::Parsed& base_parsed, 184 const Replacements<char>& replacements, 185 CharsetConverter* query_converter, 186 CanonOutput* output, 187 url_parse::Parsed* new_parsed) { 188 URLComponentSource<char> source(base); 189 url_parse::Parsed parsed(base_parsed); 190 SetupOverrideComponents(base, replacements, &source, &parsed); 191 return DoCanonicalizeStandardURL<char, unsigned char>( 192 source, parsed, query_converter, output, new_parsed); 193} 194 195// For 16-bit replacements, we turn all the replacements into UTF-8 so the 196// regular codepath can be used. 197bool ReplaceStandardURL(const char* base, 198 const url_parse::Parsed& base_parsed, 199 const Replacements<char16>& replacements, 200 CharsetConverter* query_converter, 201 CanonOutput* output, 202 url_parse::Parsed* new_parsed) { 203 RawCanonOutput<1024> utf8; 204 URLComponentSource<char> source(base); 205 url_parse::Parsed parsed(base_parsed); 206 SetupUTF16OverrideComponents(base, replacements, &utf8, &source, &parsed); 207 return DoCanonicalizeStandardURL<char, unsigned char>( 208 source, parsed, query_converter, output, new_parsed); 209} 210 211} // namespace url_canon 212