1 /* 2 * Copyright (C) 2003, 2004, 2005, 2006, 2007, 2008, 2011, 2012 Apple Inc. 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 6 * are met: 7 * 1. Redistributions of source code must retain the above copyright 8 * notice, this list of conditions and the following disclaimer. 9 * 2. Redistributions in binary form must reproduce the above copyright 10 * notice, this list of conditions and the following disclaimer in the 11 * documentation and/or other materials provided with the distribution. 12 * 13 * THIS SOFTWARE IS PROVIDED BY APPLE COMPUTER, INC. ``AS IS'' AND ANY 14 * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 15 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR 16 * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL APPLE COMPUTER, INC. OR 17 * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, 18 * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, 19 * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR 20 * PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY 21 * OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT 22 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE 23 * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 24 */ 25 26 #ifndef KURL_h 27 #define KURL_h 28 29 #include "platform/PlatformExport.h" 30 #include "wtf/Forward.h" 31 #include "wtf/HashTableDeletedValueType.h" 32 #include "wtf/OwnPtr.h" 33 #include "wtf/text/WTFString.h" 34 #include <url/third_party/mozilla/url_parse.h> 35 #include <url/url_canon.h> 36 37 namespace WTF { 38 class TextEncoding; 39 } 40 41 namespace WebCore { 42 43 struct KURLHash; 44 45 enum ParsedURLStringTag { ParsedURLString }; 46 47 class PLATFORM_EXPORT KURL { 48 public: KURL()49 KURL() 50 : m_isValid(false) 51 , m_protocolIsInHTTPFamily(false) 52 { 53 } 54 55 KURL(const KURL&); 56 KURL& operator=(const KURL&); 57 58 // The argument is an absolute URL string. The string is assumed to be 59 // output of KURL::string() called on a valid KURL object, or indiscernible 60 // from such. It is usually best to avoid repeatedly parsing a string, 61 // unless memory saving outweigh the possible slow-downs. 62 KURL(ParsedURLStringTag, const String&); 63 explicit KURL(WTF::HashTableDeletedValueType); 64 65 // Creates an isolated URL object suitable for sending to another thread. 66 static KURL createIsolated(ParsedURLStringTag, const String&); 67 isHashTableDeletedValue()68 bool isHashTableDeletedValue() const { return string().isHashTableDeletedValue(); } 69 70 // Resolves the relative URL with the given base URL. If provided, the 71 // TextEncoding is used to encode non-ASCII characers. The base URL can be 72 // null or empty, in which case the relative URL will be interpreted as 73 // absolute. 74 // FIXME: If the base URL is invalid, this always creates an invalid 75 // URL. Instead I think it would be better to treat all invalid base URLs 76 // the same way we treate null and empty base URLs. 77 KURL(const KURL& base, const String& relative); 78 KURL(const KURL& base, const String& relative, const WTF::TextEncoding&); 79 80 // For conversions from other structures that have already parsed and 81 // canonicalized the URL. The input must be exactly what KURL would have 82 // done with the same input. 83 KURL(const AtomicString& canonicalString, const url_parse::Parsed&, bool isValid); 84 85 String strippedForUseAsReferrer() const; 86 87 // FIXME: The above functions should be harmonized so that passing a 88 // base of null or the empty string gives the same result as the 89 // standard String constructor. 90 91 // Makes a deep copy. Helpful only if you need to use a KURL on another 92 // thread. Since the underlying StringImpl objects are immutable, there's 93 // no other reason to ever prefer copy() over plain old assignment. 94 KURL copy() const; 95 96 bool isNull() const; 97 bool isEmpty() const; 98 bool isValid() const; 99 100 // Returns true if this URL has a path. Note that "http://foo.com/" has a 101 // path of "/", so this function will return true. Only invalid or 102 // non-hierarchical (like "javascript:") URLs will have no path. 103 bool hasPath() const; 104 105 // Returns true if you can set the host and port for the URL. 106 // Non-hierarchical URLs don't have a host and port. canSetHostOrPort()107 bool canSetHostOrPort() const { return isHierarchical(); } 108 canSetPathname()109 bool canSetPathname() const { return isHierarchical(); } 110 bool isHierarchical() const; 111 string()112 const String& string() const { return m_string; } 113 114 String elidedString() const; 115 116 String protocol() const; 117 String host() const; 118 unsigned short port() const; 119 bool hasPort() const; 120 String user() const; 121 String pass() const; 122 String path() const; 123 String lastPathComponent() const; 124 String query() const; 125 String fragmentIdentifier() const; 126 bool hasFragmentIdentifier() const; 127 128 String baseAsString() const; 129 130 // Returns true if the current URL's protocol is the same as the null- 131 // terminated ASCII argument. The argument must be lower-case. 132 bool protocolIs(const char*) const; protocolIsData()133 bool protocolIsData() const { return protocolIs("data"); } 134 bool protocolIsInHTTPFamily() const; 135 bool isLocalFile() const; 136 bool isBlankURL() const; 137 138 bool setProtocol(const String&); 139 void setHost(const String&); 140 141 void removePort(); 142 void setPort(unsigned short); 143 void setPort(const String&); 144 145 // Input is like "foo.com" or "foo.com:8000". 146 void setHostAndPort(const String&); 147 148 void setUser(const String&); 149 void setPass(const String&); 150 151 // If you pass an empty path for HTTP or HTTPS URLs, the resulting path 152 // will be "/". 153 void setPath(const String&); 154 155 // The query may begin with a question mark, or, if not, one will be added 156 // for you. Setting the query to the empty string will leave a "?" in the 157 // URL (with nothing after it). To clear the query, pass a null string. 158 void setQuery(const String&); 159 160 void setFragmentIdentifier(const String&); 161 void removeFragmentIdentifier(); 162 163 PLATFORM_EXPORT friend bool equalIgnoringFragmentIdentifier(const KURL&, const KURL&); 164 165 unsigned hostStart() const; 166 unsigned hostEnd() const; 167 168 unsigned pathStart() const; 169 unsigned pathEnd() const; 170 unsigned pathAfterLastSlash() const; 171 172 operator const String&() const { return string(); } 173 parsed()174 const url_parse::Parsed& parsed() const { return m_parsed; } 175 innerURL()176 const KURL* innerURL() const { return m_innerURL.get(); } 177 178 #ifndef NDEBUG 179 void print() const; 180 #endif 181 182 bool isSafeToSendToAnotherThread() const; 183 184 private: 185 void init(const KURL& base, const String& relative, const WTF::TextEncoding* queryEncoding); 186 187 String componentString(const url_parse::Component&) const; 188 String stringForInvalidComponent() const; 189 190 template<typename CHAR> 191 void replaceComponents(const url_canon::Replacements<CHAR>&); 192 193 template <typename CHAR> 194 void init(const KURL& base, const CHAR* relative, int relativeLength, const WTF::TextEncoding* queryEncoding); 195 void initInnerURL(); 196 void initProtocolIsInHTTPFamily(); 197 198 bool m_isValid; 199 bool m_protocolIsInHTTPFamily; 200 url_parse::Parsed m_parsed; 201 String m_string; 202 OwnPtr<KURL> m_innerURL; 203 }; 204 205 PLATFORM_EXPORT bool operator==(const KURL&, const KURL&); 206 PLATFORM_EXPORT bool operator==(const KURL&, const String&); 207 PLATFORM_EXPORT bool operator==(const String&, const KURL&); 208 PLATFORM_EXPORT bool operator!=(const KURL&, const KURL&); 209 PLATFORM_EXPORT bool operator!=(const KURL&, const String&); 210 PLATFORM_EXPORT bool operator!=(const String&, const KURL&); 211 212 PLATFORM_EXPORT bool equalIgnoringFragmentIdentifier(const KURL&, const KURL&); 213 214 PLATFORM_EXPORT const KURL& blankURL(); 215 216 // Functions to do URL operations on strings. 217 // These are operations that aren't faster on a parsed URL. 218 // These are also different from the KURL functions in that they don't require the string to be a valid and parsable URL. 219 // This is especially important because valid javascript URLs are not necessarily considered valid by KURL. 220 221 PLATFORM_EXPORT bool protocolIs(const String& url, const char* protocol); 222 PLATFORM_EXPORT bool protocolIsJavaScript(const String& url); 223 224 PLATFORM_EXPORT bool isValidProtocol(const String&); 225 226 // Unescapes the given string using URL escaping rules, given an optional 227 // encoding (defaulting to UTF-8 otherwise). DANGER: If the URL has "%00" 228 // in it, the resulting string will have embedded null characters! 229 PLATFORM_EXPORT String decodeURLEscapeSequences(const String&); 230 PLATFORM_EXPORT String decodeURLEscapeSequences(const String&, const WTF::TextEncoding&); 231 232 PLATFORM_EXPORT String encodeWithURLEscapeSequences(const String&); 233 234 // Inlines. 235 236 inline bool operator==(const KURL& a, const KURL& b) 237 { 238 return a.string() == b.string(); 239 } 240 241 inline bool operator==(const KURL& a, const String& b) 242 { 243 return a.string() == b; 244 } 245 246 inline bool operator==(const String& a, const KURL& b) 247 { 248 return a == b.string(); 249 } 250 251 inline bool operator!=(const KURL& a, const KURL& b) 252 { 253 return a.string() != b.string(); 254 } 255 256 inline bool operator!=(const KURL& a, const String& b) 257 { 258 return a.string() != b; 259 } 260 261 inline bool operator!=(const String& a, const KURL& b) 262 { 263 return a != b.string(); 264 } 265 266 } // namespace WebCore 267 268 namespace WTF { 269 270 // KURLHash is the default hash for String 271 template<typename T> struct DefaultHash; 272 template<> struct DefaultHash<WebCore::KURL> { 273 typedef WebCore::KURLHash Hash; 274 }; 275 276 } // namespace WTF 277 278 #endif // KURL_h 279