1 /*
2 * Copyright (C) 2003, 2004, 2005, 2006, 2007, 2008, 2011, 2012, 2013 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 URL_h
27 #define URL_h
28
29 #include <wtf/Forward.h>
30 #include <wtf/HashMap.h>
31 #include <wtf/RetainPtr.h>
32 #include <wtf/text/WTFString.h>
33
34 #if USE(CF)
35 typedef const struct __CFURL* CFURLRef;
36 #if PLATFORM(JAVA)
37 typedef const struct __CFString* CFString;
38 #endif
39 #endif
40
41 #if USE(SOUP)
42 #include "GUniquePtrSoup.h"
43 #endif
44
45 #if USE(FOUNDATION)
46 OBJC_CLASS NSURL;
47 #endif
48
49 namespace WebCore {
50
51 class TextEncoding;
52 struct URLHash;
53
54 enum ParsedURLStringTag { ParsedURLString };
55
56 class URL {
57 public:
58 // Generates a URL which contains a null string.
59 URL() { invalidate(); }
60
61 // The argument is an absolute URL string. The string is assumed to be output of URL::string() called on a valid
62 // URL object, or indiscernible from such.
63 // It is usually best to avoid repeatedly parsing a string, unless memory saving outweigh the possible slow-downs.
64 URL(ParsedURLStringTag, const String&);
65 explicit URL(WTF::HashTableDeletedValueType) : m_string(WTF::HashTableDeletedValue) { }
66 bool isHashTableDeletedValue() const { return string().isHashTableDeletedValue(); }
67
68 // Resolves the relative URL with the given base URL. If provided, the
69 // TextEncoding is used to encode non-ASCII characers. The base URL can be
70 // null or empty, in which case the relative URL will be interpreted as
71 // absolute.
72 // FIXME: If the base URL is invalid, this always creates an invalid
73 // URL. Instead I think it would be better to treat all invalid base URLs
74 // the same way we treate null and empty base URLs.
75 URL(const URL& base, const String& relative);
76 URL(const URL& base, const String& relative, const TextEncoding&);
77
78 String strippedForUseAsReferrer() const;
79
80 // FIXME: The above functions should be harmonized so that passing a
81 // base of null or the empty string gives the same result as the
82 // standard String constructor.
83
84 // Makes a deep copy. Helpful only if you need to use a URL on another
85 // thread. Since the underlying StringImpl objects are immutable, there's
86 // no other reason to ever prefer copy() over plain old assignment.
87 URL copy() const;
88
89 bool isNull() const;
90 bool isEmpty() const;
91 bool isValid() const;
92
93 // Returns true if you can set the host and port for the URL.
94 // Non-hierarchical URLs don't have a host and port.
95 bool canSetHostOrPort() const { return isHierarchical(); }
96
97 bool canSetPathname() const { return isHierarchical(); }
98 bool isHierarchical() const;
99
100 const String& string() const { return m_string; }
101
102 String stringCenterEllipsizedToLength(unsigned length = 1024) const;
103
104 String protocol() const;
105 String host() const;
106 unsigned short port() const;
107 bool hasPort() const;
108 String user() const;
109 String pass() const;
110 String path() const;
111 String lastPathComponent() const;
112 String query() const;
113 String fragmentIdentifier() const;
114 bool hasFragmentIdentifier() const;
115
116 String baseAsString() const;
117
118 String fileSystemPath() const;
119
120 // Returns true if the current URL's protocol is the same as the null-
121 // terminated ASCII argument. The argument must be lower-case.
122 bool protocolIs(const char*) const;
123 bool protocolIsData() const { return protocolIs("data"); }
124 bool protocolIsInHTTPFamily() const;
125 bool isLocalFile() const;
126 bool isBlankURL() const;
127
128 bool setProtocol(const String&);
129 void setHost(const String&);
130
131 void removePort();
132 void setPort(unsigned short);
133
134 // Input is like "foo.com" or "foo.com:8000".
135 void setHostAndPort(const String&);
136
137 void setUser(const String&);
138 void setPass(const String&);
139
140 // If you pass an empty path for HTTP or HTTPS URLs, the resulting path
141 // will be "/".
142 void setPath(const String&);
143
144 // The query may begin with a question mark, or, if not, one will be added
145 // for you. Setting the query to the empty string will leave a "?" in the
146 // URL (with nothing after it). To clear the query, pass a null string.
147 void setQuery(const String&);
148
149 void setFragmentIdentifier(const String&);
150 void removeFragmentIdentifier();
151
152 friend bool equalIgnoringFragmentIdentifier(const URL&, const URL&);
153
154 friend bool protocolHostAndPortAreEqual(const URL&, const URL&);
155
156 unsigned hostStart() const;
157 unsigned hostEnd() const;
158
159 unsigned pathStart() const;
160 unsigned pathEnd() const;
161 unsigned pathAfterLastSlash() const;
162
163 operator const String&() const { return string(); }
164
165 #if USE(CF)
166 URL(CFURLRef);
167 RetainPtr<CFURLRef> createCFURL() const;
168 #endif
169
170 #if USE(SOUP)
171 URL(SoupURI*);
172 GUniquePtr<SoupURI> createSoupURI() const;
173 #endif
174
175 #if USE(FOUNDATION)
176 URL(NSURL*);
177 operator NSURL*() const;
178 #endif
179 #ifdef __OBJC__
180 operator NSString*() const { return string(); }
181 #endif
182
183 #if PLATFORM(JAVA)
184 String deprecatedString() const;
185 bool isJarFile() const { return m_protocolIsInJar; }
186 #endif
187 const URL* innerURL() const { return 0; }
188
189 #ifndef NDEBUG
190 void print() const;
191 #endif
192
193 bool isSafeToSendToAnotherThread() const;
194
195 private:
196 void invalidate();
197 static bool protocolIs(const String&, const char*);
198 void init(const URL&, const String&, const TextEncoding&);
199 void copyToBuffer(Vector<char, 512>& buffer) const;
200
201 // Parses the given URL. The originalString parameter allows for an
202 // optimization: When the source is the same as the fixed-up string,
203 // it will use the passed-in string instead of allocating a new one.
204 void parse(const String&);
205 void parse(const char* url, const String* originalString = 0);
206
207 bool hasPath() const;
208
209 String m_string;
210 bool m_isValid : 1;
211 bool m_protocolIsInHTTPFamily : 1;
212 #if PLATFORM(JAVA)
213 bool m_protocolIsInJar : 1;
214 #endif
215
216 int m_schemeEnd;
217 int m_userStart;
218 int m_userEnd;
219 int m_passwordEnd;
220 int m_hostEnd;
221 int m_portEnd;
222 int m_pathAfterLastSlash;
223 int m_pathEnd;
224 int m_queryEnd;
225 int m_fragmentEnd;
226 };
227
228 bool operator==(const URL&, const URL&);
229 bool operator==(const URL&, const String&);
230 bool operator==(const String&, const URL&);
231 bool operator!=(const URL&, const URL&);
232 bool operator!=(const URL&, const String&);
233 bool operator!=(const String&, const URL&);
234
235 bool equalIgnoringFragmentIdentifier(const URL&, const URL&);
236 bool protocolHostAndPortAreEqual(const URL&, const URL&);
237
238 const URL& blankURL();
239
240 // Functions to do URL operations on strings.
241 // These are operations that aren't faster on a parsed URL.
242 // These are also different from the URL functions in that they don't require the string to be a valid and parsable URL.
243 // This is especially important because valid javascript URLs are not necessarily considered valid by URL.
244
245 bool protocolIs(const String& url, const char* protocol);
246 bool protocolIsJavaScript(const String& url);
247 bool protocolIsInHTTPFamily(const String& url);
248
249 bool isDefaultPortForProtocol(unsigned short port, const String& protocol);
250 bool portAllowed(const URL&); // Blacklist ports that should never be used for Web resources.
251
252 bool isValidProtocol(const String&);
253
254 String mimeTypeFromDataURL(const String& url);
255 String mimeTypeFromURL(const URL&);
256
257 // Unescapes the given string using URL escaping rules, given an optional
258 // encoding (defaulting to UTF-8 otherwise). DANGER: If the URL has "%00"
259 // in it, the resulting string will have embedded null characters!
260 String decodeURLEscapeSequences(const String&);
261 String decodeURLEscapeSequences(const String&, const TextEncoding&);
262
263 String encodeWithURLEscapeSequences(const String&);
264
265 // Inlines.
266
267 inline bool operator==(const URL& a, const URL& b)
268 {
269 return a.string() == b.string();
270 }
271
272 inline bool operator==(const URL& a, const String& b)
273 {
274 return a.string() == b;
275 }
276
277 inline bool operator==(const String& a, const URL& b)
278 {
279 return a == b.string();
280 }
281
282 inline bool operator!=(const URL& a, const URL& b)
283 {
284 return a.string() != b.string();
285 }
286
287 inline bool operator!=(const URL& a, const String& b)
288 {
289 return a.string() != b;
290 }
291
292 inline bool operator!=(const String& a, const URL& b)
293 {
294 return a != b.string();
295 }
296
297 // Inline versions of some non-GoogleURL functions so we can get inlining
298 // without having to have a lot of ugly ifdefs in the class definition.
299
300 inline bool URL::isNull() const
301 {
302 return m_string.isNull();
303 }
304
305 inline bool URL::isEmpty() const
306 {
307 return m_string.isEmpty();
308 }
309
310 inline bool URL::isValid() const
311 {
312 return m_isValid;
313 }
314
315 inline bool URL::hasPath() const
316 {
317 return m_pathEnd != m_portEnd;
318 }
319
320 inline bool URL::hasPort() const
321 {
322 return m_hostEnd < m_portEnd;
323 }
324
325 inline bool URL::protocolIsInHTTPFamily() const
326 {
327 return m_protocolIsInHTTPFamily;
328 }
329
330 inline unsigned URL::hostStart() const
331 {
332 return (m_passwordEnd == m_userStart) ? m_passwordEnd : m_passwordEnd + 1;
333 }
334
335 inline unsigned URL::hostEnd() const
336 {
337 return m_hostEnd;
338 }
339
340 inline unsigned URL::pathStart() const
341 {
342 return m_portEnd;
343 }
344
345 inline unsigned URL::pathEnd() const
346 {
347 return m_pathEnd;
348 }
349
350 inline unsigned URL::pathAfterLastSlash() const
351 {
352 return m_pathAfterLastSlash;
353 }
354
355 #if PLATFORM(IOS)
356 void enableURLSchemeCanonicalization(bool);
357 #endif
358
359 } // namespace WebCore
360
361 namespace WTF {
362
363 // URLHash is the default hash for String
364 template<typename T> struct DefaultHash;
365 template<> struct DefaultHash<WebCore::URL> {
366 typedef WebCore::URLHash Hash;
367 };
368
369 } // namespace WTF
370
371 #endif // URL_h