1 // Copyright (c) 2012 The Chromium Authors. All rights reserved. 2 // Use of this source code is governed by a BSD-style license that can be 3 // found in the LICENSE file. 4 5 // NB: Modelled after Mozilla's code (originally written by Pamela Greene, 6 // later modified by others), but almost entirely rewritten for Chrome. 7 // (netwerk/dns/src/nsEffectiveTLDService.h) 8 /* ***** BEGIN LICENSE BLOCK ***** 9 * Version: MPL 1.1/GPL 2.0/LGPL 2.1 10 * 11 * The contents of this file are subject to the Mozilla Public License Version 12 * 1.1 (the "License"); you may not use this file except in compliance with 13 * the License. You may obtain a copy of the License at 14 * http://www.mozilla.org/MPL/ 15 * 16 * Software distributed under the License is distributed on an "AS IS" basis, 17 * WITHOUT WARRANTY OF ANY KIND, either express or implied. See the License 18 * for the specific language governing rights and limitations under the 19 * License. 20 * 21 * The Original Code is Mozilla TLD Service 22 * 23 * The Initial Developer of the Original Code is 24 * Google Inc. 25 * Portions created by the Initial Developer are Copyright (C) 2006 26 * the Initial Developer. All Rights Reserved. 27 * 28 * Contributor(s): 29 * Pamela Greene <pamg.bugs@gmail.com> (original author) 30 * 31 * Alternatively, the contents of this file may be used under the terms of 32 * either the GNU General Public License Version 2 or later (the "GPL"), or 33 * the GNU Lesser General Public License Version 2.1 or later (the "LGPL"), 34 * in which case the provisions of the GPL or the LGPL are applicable instead 35 * of those above. If you wish to allow use of your version of this file only 36 * under the terms of either the GPL or the LGPL, and not to allow others to 37 * use your version of this file under the terms of the MPL, indicate your 38 * decision by deleting the provisions above and replace them with the notice 39 * and other provisions required by the GPL or the LGPL. If you do not delete 40 * the provisions above, a recipient may use your version of this file under 41 * the terms of any one of the MPL, the GPL or the LGPL. 42 * 43 * ***** END LICENSE BLOCK ***** */ 44 45 /* 46 (Documentation based on the Mozilla documentation currently at 47 http://wiki.mozilla.org/Gecko:Effective_TLD_Service, written by the same 48 author.) 49 50 The RegistryControlledDomainService examines the hostname of a GURL passed to 51 it and determines the longest portion that is controlled by a registrar. 52 Although technically the top-level domain (TLD) for a hostname is the last 53 dot-portion of the name (such as .com or .org), many domains (such as co.uk) 54 function as though they were TLDs, allocating any number of more specific, 55 essentially unrelated names beneath them. For example, .uk is a TLD, but 56 nobody is allowed to register a domain directly under .uk; the "effective" 57 TLDs are ac.uk, co.uk, and so on. We wouldn't want to allow any site in 58 *.co.uk to set a cookie for the entire co.uk domain, so it's important to be 59 able to identify which higher-level domains function as effective TLDs and 60 which can be registered. 61 62 The service obtains its information about effective TLDs from a text resource 63 that must be in the following format: 64 65 * It should use plain ASCII. 66 * It should contain one domain rule per line, terminated with \n, with nothing 67 else on the line. (The last rule in the file may omit the ending \n.) 68 * Rules should have been normalized using the same canonicalization that GURL 69 applies. For ASCII, that means they're not case-sensitive, among other 70 things; other normalizations are applied for other characters. 71 * Each rule should list the entire TLD-like domain name, with any subdomain 72 portions separated by dots (.) as usual. 73 * Rules should neither begin nor end with a dot. 74 * If a hostname matches more than one rule, the most specific rule (that is, 75 the one with more dot-levels) will be used. 76 * Other than in the case of wildcards (see below), rules do not implicitly 77 include their subcomponents. For example, "bar.baz.uk" does not imply 78 "baz.uk", and if "bar.baz.uk" is the only rule in the list, "foo.bar.baz.uk" 79 will match, but "baz.uk" and "qux.baz.uk" won't. 80 * The wildcard character '*' will match any valid sequence of characters. 81 * Wildcards may only appear as the entire most specific level of a rule. That 82 is, a wildcard must come at the beginning of a line and must be followed by 83 a dot. (You may not use a wildcard as the entire rule.) 84 * A wildcard rule implies a rule for the entire non-wildcard portion. For 85 example, the rule "*.foo.bar" implies the rule "foo.bar" (but not the rule 86 "bar"). This is typically important in the case of exceptions (see below). 87 * The exception character '!' before a rule marks an exception to a wildcard 88 rule. If your rules are "*.tokyo.jp" and "!pref.tokyo.jp", then 89 "a.b.tokyo.jp" has an effective TLD of "b.tokyo.jp", but "a.pref.tokyo.jp" 90 has an effective TLD of "tokyo.jp" (the exception prevents the wildcard 91 match, and we thus fall through to matching on the implied "tokyo.jp" rule 92 from the wildcard). 93 * If you use an exception rule without a corresponding wildcard rule, the 94 behavior is undefined. 95 96 Firefox has a very similar service, and it's their data file we use to 97 construct our resource. However, the data expected by this implementation 98 differs from the Mozilla file in several important ways: 99 (1) We require that all single-level TLDs (com, edu, etc.) be explicitly 100 listed. As of this writing, Mozilla's file includes the single-level 101 TLDs too, but that might change. 102 (2) Our data is expected be in pure ASCII: all UTF-8 or otherwise encoded 103 items must already have been normalized. 104 (3) We do not allow comments, rule notes, blank lines, or line endings other 105 than LF. 106 Rules are also expected to be syntactically valid. 107 108 The utility application tld_cleanup.exe converts a Mozilla-style file into a 109 Chrome one, making sure that single-level TLDs are explicitly listed, using 110 GURL to normalize rules, and validating the rules. 111 */ 112 113 #ifndef NET_BASE_REGISTRY_CONTROLLED_DOMAINS_REGISTRY_CONTROLLED_DOMAIN_H_ 114 #define NET_BASE_REGISTRY_CONTROLLED_DOMAINS_REGISTRY_CONTROLLED_DOMAIN_H_ 115 116 #include <string> 117 118 #include "base/basictypes.h" 119 #include "net/base/net_export.h" 120 121 class GURL; 122 123 struct DomainRule; 124 125 namespace net { 126 namespace registry_controlled_domains { 127 128 // This enum is a required parameter to all public methods declared for this 129 // service. The Public Suffix List (http://publicsuffix.org/) this service 130 // uses as a data source splits all effective-TLDs into two groups. The main 131 // group describes registries that are acknowledged by ICANN. The second group 132 // contains a list of private additions for domains that enable external users 133 // to create subdomains, such as appspot.com. 134 // The RegistryFilter enum lets you choose whether you want to include the 135 // private additions in your lookup. 136 // See this for example use cases: 137 // https://wiki.mozilla.org/Public_Suffix_List/Use_Cases 138 enum NET_EXPORT PrivateRegistryFilter { 139 EXCLUDE_PRIVATE_REGISTRIES = 0, 140 INCLUDE_PRIVATE_REGISTRIES 141 }; 142 143 // This enum is a required parameter to the GetRegistryLength functions 144 // declared for this service. Whenever there is no matching rule in the 145 // effective-TLD data (or in the default data, if the resource failed to 146 // load), the result will be dependent on which enum value was passed in. 147 // If EXCLUDE_UNKNOWN_REGISTRIES was passed in, the resulting registry length 148 // will be 0. If INCLUDE_UNKNOWN_REGISTRIES was passed in, the resulting 149 // registry length will be the length of the last subcomponent (eg. 3 for 150 // foobar.baz). 151 enum NET_EXPORT UnknownRegistryFilter { 152 EXCLUDE_UNKNOWN_REGISTRIES = 0, 153 INCLUDE_UNKNOWN_REGISTRIES 154 }; 155 156 // Returns the registered, organization-identifying host and all its registry 157 // information, but no subdomains, from the given GURL. Returns an empty 158 // string if the GURL is invalid, has no host (e.g. a file: URL), has multiple 159 // trailing dots, is an IP address, has only one subcomponent (i.e. no dots 160 // other than leading/trailing ones), or is itself a recognized registry 161 // identifier. If no matching rule is found in the effective-TLD data (or in 162 // the default data, if the resource failed to load), the last subcomponent of 163 // the host is assumed to be the registry. 164 // 165 // Examples: 166 // http://www.google.com/file.html -> "google.com" (com) 167 // http://..google.com/file.html -> "google.com" (com) 168 // http://google.com./file.html -> "google.com." (com) 169 // http://a.b.co.uk/file.html -> "b.co.uk" (co.uk) 170 // file:///C:/bar.html -> "" (no host) 171 // http://foo.com../file.html -> "" (multiple trailing dots) 172 // http://192.168.0.1/file.html -> "" (IP address) 173 // http://bar/file.html -> "" (no subcomponents) 174 // http://co.uk/file.html -> "" (host is a registry) 175 // http://foo.bar/file.html -> "foo.bar" (no rule; assume bar) 176 NET_EXPORT std::string GetDomainAndRegistry(const GURL& gurl, 177 PrivateRegistryFilter filter); 178 179 // Like the GURL version, but takes a host (which is canonicalized internally) 180 // instead of a full GURL. 181 NET_EXPORT std::string GetDomainAndRegistry(const std::string& host, 182 PrivateRegistryFilter filter); 183 184 // This convenience function returns true if the two GURLs both have hosts 185 // and one of the following is true: 186 // * They each have a known domain and registry, and it is the same for both 187 // URLs. Note that this means the trailing dot, if any, must match too. 188 // * They don't have known domains/registries, but the hosts are identical. 189 // Effectively, callers can use this function to check whether the input URLs 190 // represent hosts "on the same site". 191 NET_EXPORT bool SameDomainOrHost(const GURL& gurl1, const GURL& gurl2, 192 PrivateRegistryFilter filter); 193 194 // Finds the length in bytes of the registrar portion of the host in the 195 // given GURL. Returns std::string::npos if the GURL is invalid or has no 196 // host (e.g. a file: URL). Returns 0 if the GURL has multiple trailing dots, 197 // is an IP address, has no subcomponents, or is itself a recognized registry 198 // identifier. The result is also dependent on the UnknownRegistryFilter. 199 // If no matching rule is found in the effective-TLD data (or in 200 // the default data, if the resource failed to load), returns 0 if 201 // |unknown_filter| is EXCLUDE_UNKNOWN_REGISTRIES, or the length of the last 202 // subcomponent if |unknown_filter| is INCLUDE_UNKNOWN_REGISTRIES. 203 // 204 // Examples: 205 // http://www.google.com/file.html -> 3 (com) 206 // http://..google.com/file.html -> 3 (com) 207 // http://google.com./file.html -> 4 (com) 208 // http://a.b.co.uk/file.html -> 5 (co.uk) 209 // file:///C:/bar.html -> std::string::npos (no host) 210 // http://foo.com../file.html -> 0 (multiple trailing 211 // dots) 212 // http://192.168.0.1/file.html -> 0 (IP address) 213 // http://bar/file.html -> 0 (no subcomponents) 214 // http://co.uk/file.html -> 0 (host is a registry) 215 // http://foo.bar/file.html -> 0 or 3, depending (no rule; assume 216 // bar) 217 NET_EXPORT size_t GetRegistryLength(const GURL& gurl, 218 UnknownRegistryFilter unknown_filter, 219 PrivateRegistryFilter private_filter); 220 221 // Like the GURL version, but takes a host (which is canonicalized internally) 222 // instead of a full GURL. 223 NET_EXPORT size_t GetRegistryLength(const std::string& host, 224 UnknownRegistryFilter unknown_filter, 225 PrivateRegistryFilter private_filter); 226 227 typedef const struct DomainRule* (*FindDomainPtr)(const char *, unsigned int); 228 229 // Used for unit tests. Use default domains. 230 NET_EXPORT_PRIVATE void SetFindDomainGraph(); 231 232 // Used for unit tests, so that a frozen list of domains is used. 233 NET_EXPORT_PRIVATE void SetFindDomainGraph(const unsigned char* domains, 234 size_t length); 235 } // namespace registry_controlled_domains 236 } // namespace net 237 238 #endif // NET_BASE_REGISTRY_CONTROLLED_DOMAINS_REGISTRY_CONTROLLED_DOMAIN_H_ 239