]>
Commit | Line | Data |
---|---|---|
dd65d8c8 | 1 | ///////////////////////////////////////////////////////////////////////////// |
80fdcdb9 | 2 | // Name: src/common/uri.cpp |
2186321f VZ |
3 | // Purpose: Implementation of a URI parser |
4 | // Author: Ryan Norton, | |
5 | // Vadim Zeitlin (UTF-8 URI support, many other changes) | |
dd65d8c8 RN |
6 | // Created: 10/26/04 |
7 | // RCS-ID: $Id$ | |
2186321f VZ |
8 | // Copyright: (c) 2004 Ryan Norton, |
9 | // 2008 Vadim Zeitlin | |
10 | // Licence: wxWindows licence | |
dd65d8c8 RN |
11 | ///////////////////////////////////////////////////////////////////////////// |
12 | ||
13 | // =========================================================================== | |
14 | // declarations | |
15 | // =========================================================================== | |
16 | ||
17 | // --------------------------------------------------------------------------- | |
18 | // headers | |
19 | // --------------------------------------------------------------------------- | |
20 | ||
dd65d8c8 RN |
21 | // For compilers that support precompilation, includes "wx.h". |
22 | #include "wx/wxprec.h" | |
23 | ||
24 | #ifdef __BORLANDC__ | |
25 | #pragma hdrstop | |
26 | #endif | |
27 | ||
0bf751e7 VS |
28 | #ifndef WX_PRECOMP |
29 | #include "wx/crt.h" | |
30 | #endif | |
31 | ||
dd65d8c8 RN |
32 | #include "wx/uri.h" |
33 | ||
34 | // --------------------------------------------------------------------------- | |
35 | // definitions | |
36 | // --------------------------------------------------------------------------- | |
37 | ||
4115960d | 38 | IMPLEMENT_CLASS(wxURI, wxObject) |
dd65d8c8 RN |
39 | |
40 | // =========================================================================== | |
2186321f | 41 | // wxURI implementation |
dd65d8c8 RN |
42 | // =========================================================================== |
43 | ||
44 | // --------------------------------------------------------------------------- | |
2186321f | 45 | // Constructors and cleanup |
dd65d8c8 RN |
46 | // --------------------------------------------------------------------------- |
47 | ||
2186321f VZ |
48 | wxURI::wxURI() |
49 | : m_hostType(wxURI_REGNAME), | |
50 | m_fields(0) | |
dd65d8c8 RN |
51 | { |
52 | } | |
846978d7 | 53 | |
2186321f VZ |
54 | wxURI::wxURI(const wxString& uri) |
55 | : m_hostType(wxURI_REGNAME), | |
56 | m_fields(0) | |
dd65d8c8 RN |
57 | { |
58 | Create(uri); | |
59 | } | |
60 | ||
2186321f | 61 | bool wxURI::Create(const wxString& uri) |
dd65d8c8 | 62 | { |
2186321f VZ |
63 | if (m_fields) |
64 | Clear(); | |
dd65d8c8 | 65 | |
2186321f | 66 | return Parse(uri.utf8_str()); |
dd65d8c8 RN |
67 | } |
68 | ||
69 | void wxURI::Clear() | |
70 | { | |
2186321f VZ |
71 | m_scheme = |
72 | m_userinfo = | |
73 | m_server = | |
74 | m_port = | |
75 | m_path = | |
76 | m_query = | |
77 | m_fragment = wxEmptyString; | |
dd65d8c8 RN |
78 | |
79 | m_hostType = wxURI_REGNAME; | |
80 | ||
81 | m_fields = 0; | |
82 | } | |
83 | ||
84 | // --------------------------------------------------------------------------- | |
2186321f | 85 | // Escaped characters handling |
dd65d8c8 RN |
86 | // --------------------------------------------------------------------------- |
87 | ||
2186321f VZ |
88 | // Converts a character into a numeric hexadecimal value, or -1 if the passed |
89 | // in character is not a valid hex character | |
dd65d8c8 | 90 | |
2186321f VZ |
91 | /* static */ |
92 | int wxURI::CharToHex(char c) | |
93 | { | |
94 | if ((c >= 'A') && (c <= 'Z')) | |
95 | return c - 'A' + 10; | |
96 | if ((c >= 'a') && (c <= 'z')) | |
97 | return c - 'a' + 10; | |
98 | if ((c >= '0') && (c <= '9')) | |
99 | return c - '0'; | |
100 | ||
101 | return -1; | |
846978d7 | 102 | } |
dd65d8c8 | 103 | |
2186321f | 104 | int wxURI::DecodeEscape(wxString::const_iterator& i) |
dd65d8c8 | 105 | { |
2186321f VZ |
106 | int hi = CharToHex(*++i); |
107 | if ( hi == -1 ) | |
108 | return -1; | |
8404931e | 109 | |
2186321f VZ |
110 | int lo = CharToHex(*++i); |
111 | if ( lo == -1 ) | |
112 | return -1; | |
c9f78968 | 113 | |
2186321f | 114 | return (hi << 4) | lo; |
dd65d8c8 RN |
115 | } |
116 | ||
2186321f | 117 | /* static */ |
86470d43 RN |
118 | wxString wxURI::Unescape(const wxString& uri) |
119 | { | |
2186321f VZ |
120 | // the unescaped version can't be longer than the original one |
121 | wxCharBuffer buf(uri.length()); | |
122 | char *p = buf.data(); | |
86470d43 | 123 | |
2186321f | 124 | for ( wxString::const_iterator i = uri.begin(); i != uri.end(); ++i, ++p ) |
86470d43 | 125 | { |
2186321f VZ |
126 | char c = *i; |
127 | if ( c == '%' ) | |
86470d43 | 128 | { |
2186321f VZ |
129 | int n = wxURI::DecodeEscape(i); |
130 | if ( n == -1 ) | |
131 | return wxString(); | |
132 | ||
133 | wxASSERT_MSG( n >= 0 && n <= 0xff, "unexpected character value" ); | |
134 | ||
5c33522f | 135 | c = static_cast<char>(n); |
86470d43 | 136 | } |
2186321f VZ |
137 | |
138 | *p = c; | |
86470d43 RN |
139 | } |
140 | ||
2186321f | 141 | *p = '\0'; |
86470d43 | 142 | |
2186321f VZ |
143 | // by default assume that the URI is in UTF-8, this is the most common |
144 | // practice | |
145 | wxString s = wxString::FromUTF8(buf); | |
146 | if ( s.empty() ) | |
147 | { | |
148 | // if it isn't, use latin-1 as a fallback -- at least this always | |
149 | // succeeds | |
150 | s = wxCSConv(wxFONTENCODING_ISO8859_1).cMB2WC(buf); | |
151 | } | |
152 | ||
153 | return s; | |
dd65d8c8 RN |
154 | } |
155 | ||
2186321f | 156 | void wxURI::AppendNextEscaped(wxString& s, const char *& p) |
dd65d8c8 | 157 | { |
2186321f VZ |
158 | // check for an already encoded character: |
159 | // | |
ce321570 | 160 | // pct-encoded = "%" HEXDIG HEXDIG |
2186321f VZ |
161 | if ( p[0] == '%' && IsHex(p[1]) && IsHex(p[2]) ) |
162 | { | |
163 | s += *p++; | |
164 | s += *p++; | |
165 | s += *p++; | |
166 | } | |
167 | else // really needs escaping | |
168 | { | |
169 | static const char* hexDigits = "0123456789abcdef"; | |
170 | ||
171 | const char c = *p++; | |
172 | ||
173 | s += '%'; | |
174 | s += hexDigits[(c >> 4) & 15]; | |
175 | s += hexDigits[c & 15]; | |
176 | } | |
dd65d8c8 RN |
177 | } |
178 | ||
4860d40d RN |
179 | // --------------------------------------------------------------------------- |
180 | // GetUser | |
181 | // GetPassword | |
182 | // | |
183 | // Gets the username and password via the old URL method. | |
184 | // --------------------------------------------------------------------------- | |
185 | wxString wxURI::GetUser() const | |
186 | { | |
62e3e6c2 VZ |
187 | // if there is no colon at all, find() returns npos and this method returns |
188 | // the entire string which is correct as it means that password was omitted | |
189 | return m_userinfo(0, m_userinfo.find(':')); | |
4860d40d RN |
190 | } |
191 | ||
192 | wxString wxURI::GetPassword() const | |
193 | { | |
62e3e6c2 | 194 | size_t posColon = m_userinfo.find(':'); |
4860d40d | 195 | |
62e3e6c2 | 196 | if ( posColon == wxString::npos ) |
2186321f | 197 | return ""; |
62e3e6c2 VZ |
198 | |
199 | return m_userinfo(posColon + 1, wxString::npos); | |
dd65d8c8 RN |
200 | } |
201 | ||
2186321f VZ |
202 | // combine all URI fields in a single string, applying funcDecode to each |
203 | // component which it may make sense to decode (i.e. "unescape") | |
204 | wxString wxURI::DoBuildURI(wxString (*funcDecode)(const wxString&)) const | |
86470d43 RN |
205 | { |
206 | wxString ret; | |
207 | ||
208 | if (HasScheme()) | |
2186321f | 209 | ret += m_scheme + ":"; |
86470d43 RN |
210 | |
211 | if (HasServer()) | |
212 | { | |
2186321f | 213 | ret += "//"; |
86470d43 | 214 | |
4860d40d | 215 | if (HasUserInfo()) |
2186321f | 216 | ret += funcDecode(m_userinfo) + "@"; |
86470d43 RN |
217 | |
218 | if (m_hostType == wxURI_REGNAME) | |
2186321f | 219 | ret += funcDecode(m_server); |
86470d43 RN |
220 | else |
221 | ret += m_server; | |
222 | ||
223 | if (HasPort()) | |
2186321f | 224 | ret += ":" + m_port; |
86470d43 RN |
225 | } |
226 | ||
2186321f | 227 | ret += funcDecode(m_path); |
86470d43 RN |
228 | |
229 | if (HasQuery()) | |
2186321f | 230 | ret += "?" + funcDecode(m_query); |
86470d43 RN |
231 | |
232 | if (HasFragment()) | |
2186321f | 233 | ret += "#" + funcDecode(m_fragment); |
86470d43 RN |
234 | |
235 | return ret; | |
236 | } | |
237 | ||
ce321570 RN |
238 | // --------------------------------------------------------------------------- |
239 | // Comparison | |
240 | // --------------------------------------------------------------------------- | |
241 | ||
2186321f | 242 | bool wxURI::operator==(const wxURI& uri) const |
846978d7 | 243 | { |
dd65d8c8 RN |
244 | if (HasScheme()) |
245 | { | |
246 | if(m_scheme != uri.m_scheme) | |
247 | return false; | |
248 | } | |
249 | else if (uri.HasScheme()) | |
250 | return false; | |
251 | ||
252 | ||
253 | if (HasServer()) | |
254 | { | |
4860d40d | 255 | if (HasUserInfo()) |
dd65d8c8 | 256 | { |
4860d40d | 257 | if (m_userinfo != uri.m_userinfo) |
dd65d8c8 RN |
258 | return false; |
259 | } | |
4860d40d | 260 | else if (uri.HasUserInfo()) |
dd65d8c8 RN |
261 | return false; |
262 | ||
263 | if (m_server != uri.m_server || | |
264 | m_hostType != uri.m_hostType) | |
265 | return false; | |
266 | ||
267 | if (HasPort()) | |
268 | { | |
269 | if(m_port != uri.m_port) | |
270 | return false; | |
271 | } | |
272 | else if (uri.HasPort()) | |
273 | return false; | |
274 | } | |
275 | else if (uri.HasServer()) | |
276 | return false; | |
277 | ||
278 | ||
279 | if (HasPath()) | |
280 | { | |
281 | if(m_path != uri.m_path) | |
282 | return false; | |
283 | } | |
284 | else if (uri.HasPath()) | |
285 | return false; | |
286 | ||
287 | if (HasQuery()) | |
288 | { | |
289 | if (m_query != uri.m_query) | |
290 | return false; | |
291 | } | |
292 | else if (uri.HasQuery()) | |
293 | return false; | |
294 | ||
295 | if (HasFragment()) | |
296 | { | |
297 | if (m_fragment != uri.m_fragment) | |
298 | return false; | |
299 | } | |
300 | else if (uri.HasFragment()) | |
301 | return false; | |
302 | ||
303 | return true; | |
304 | } | |
305 | ||
306 | // --------------------------------------------------------------------------- | |
307 | // IsReference | |
308 | // | |
309 | // if there is no authority or scheme, it is a reference | |
310 | // --------------------------------------------------------------------------- | |
311 | ||
312 | bool wxURI::IsReference() const | |
2186321f VZ |
313 | { |
314 | return !HasScheme() || !HasServer(); | |
315 | } | |
dd65d8c8 RN |
316 | |
317 | // --------------------------------------------------------------------------- | |
318 | // Parse | |
319 | // | |
320 | // Master URI parsing method. Just calls the individual parsing methods | |
321 | // | |
322 | // URI = scheme ":" hier-part [ "?" query ] [ "#" fragment ] | |
4cc52142 | 323 | // URI-reference = URI / relative |
dd65d8c8 RN |
324 | // --------------------------------------------------------------------------- |
325 | ||
2186321f | 326 | bool wxURI::Parse(const char *uri) |
dd65d8c8 RN |
327 | { |
328 | uri = ParseScheme(uri); | |
2186321f VZ |
329 | if ( uri ) |
330 | uri = ParseAuthority(uri); | |
331 | if ( uri ) | |
332 | uri = ParsePath(uri); | |
333 | if ( uri ) | |
334 | uri = ParseQuery(uri); | |
335 | if ( uri ) | |
336 | uri = ParseFragment(uri); | |
337 | ||
338 | // we only succeed if we parsed the entire string | |
339 | return uri && *uri == '\0'; | |
dd65d8c8 RN |
340 | } |
341 | ||
2186321f | 342 | const char* wxURI::ParseScheme(const char *uri) |
dd65d8c8 | 343 | { |
2186321f | 344 | const char * const start = uri; |
dd65d8c8 | 345 | |
2186321f VZ |
346 | // assume that we have a scheme if we have the valid start of it |
347 | if ( IsAlpha(*uri) ) | |
dd65d8c8 RN |
348 | { |
349 | m_scheme += *uri++; | |
350 | ||
351 | //scheme = ALPHA *( ALPHA / DIGIT / "+" / "-" / "." ) | |
846978d7 | 352 | while (IsAlpha(*uri) || IsDigit(*uri) || |
2186321f VZ |
353 | *uri == '+' || |
354 | *uri == '-' || | |
355 | *uri == '.') | |
846978d7 WS |
356 | { |
357 | m_scheme += *uri++; | |
dd65d8c8 RN |
358 | } |
359 | ||
360 | //valid scheme? | |
2186321f | 361 | if (*uri == ':') |
846978d7 | 362 | { |
dd65d8c8 RN |
363 | //mark the scheme as valid |
364 | m_fields |= wxURI_SCHEME; | |
365 | ||
366 | //move reference point up to input buffer | |
2186321f VZ |
367 | ++uri; |
368 | } | |
369 | else // no valid scheme finally | |
370 | { | |
371 | uri = start; // rewind | |
372 | m_scheme.clear(); | |
dd65d8c8 | 373 | } |
dd65d8c8 | 374 | } |
2186321f | 375 | //else: can't have schema, possible a relative URI |
dd65d8c8 | 376 | |
2186321f | 377 | return uri; |
dd65d8c8 RN |
378 | } |
379 | ||
2186321f | 380 | const char* wxURI::ParseAuthority(const char* uri) |
dd65d8c8 RN |
381 | { |
382 | // authority = [ userinfo "@" ] host [ ":" port ] | |
2186321f | 383 | if ( uri[0] == '/' && uri[1] == '/' ) |
dd65d8c8 | 384 | { |
97ad053b | 385 | //skip past the two slashes |
dd65d8c8 RN |
386 | uri += 2; |
387 | ||
97ad053b VZ |
388 | // ############# DEVIATION FROM RFC ######################### |
389 | // Don't parse the server component for file URIs | |
2186321f | 390 | if(m_scheme != "file") |
97ad053b VZ |
391 | { |
392 | //normal way | |
2186321f VZ |
393 | uri = ParseUserInfo(uri); |
394 | uri = ParseServer(uri); | |
395 | return ParsePort(uri); | |
97ad053b | 396 | } |
dd65d8c8 RN |
397 | } |
398 | ||
399 | return uri; | |
400 | } | |
401 | ||
2186321f | 402 | const char* wxURI::ParseUserInfo(const char* uri) |
dd65d8c8 | 403 | { |
2186321f | 404 | const char * const start = uri; |
dd65d8c8 RN |
405 | |
406 | // userinfo = *( unreserved / pct-encoded / sub-delims / ":" ) | |
2186321f | 407 | while ( *uri && *uri != '@' && *uri != '/' && *uri != '#' && *uri != '?' ) |
dd65d8c8 | 408 | { |
2186321f | 409 | if ( IsUnreserved(*uri) || IsSubDelim(*uri) || *uri == ':' ) |
24ca04e7 | 410 | m_userinfo += *uri++; |
dd65d8c8 | 411 | else |
2186321f | 412 | AppendNextEscaped(m_userinfo, uri); |
dd65d8c8 RN |
413 | } |
414 | ||
2186321f | 415 | if ( *uri++ == '@' ) |
dd65d8c8 | 416 | { |
2186321f | 417 | // valid userinfo |
4860d40d | 418 | m_fields |= wxURI_USERINFO; |
dd65d8c8 RN |
419 | } |
420 | else | |
2186321f VZ |
421 | { |
422 | uri = start; // rewind | |
423 | m_userinfo.clear(); | |
424 | } | |
dd65d8c8 | 425 | |
2186321f | 426 | return uri; |
dd65d8c8 RN |
427 | } |
428 | ||
2186321f | 429 | const char* wxURI::ParseServer(const char* uri) |
dd65d8c8 | 430 | { |
2186321f | 431 | const char * const start = uri; |
dd65d8c8 RN |
432 | |
433 | // host = IP-literal / IPv4address / reg-name | |
434 | // IP-literal = "[" ( IPv6address / IPvFuture ) "]" | |
2186321f | 435 | if (*uri == '[') |
dd65d8c8 | 436 | { |
2186321f VZ |
437 | ++uri; |
438 | if (ParseIPv6address(uri) && *uri == ']') | |
dd65d8c8 | 439 | { |
dd65d8c8 | 440 | m_hostType = wxURI_IPV6ADDRESS; |
846978d7 | 441 | |
c4dbb953 | 442 | m_server.assign(start + 1, uri - start - 1); |
2186321f | 443 | ++uri; |
dd65d8c8 RN |
444 | } |
445 | else | |
446 | { | |
2186321f | 447 | uri = start + 1; // skip the leading '[' again |
dd65d8c8 | 448 | |
2186321f | 449 | if (ParseIPvFuture(uri) && *uri == ']') |
dd65d8c8 | 450 | { |
846978d7 WS |
451 | m_hostType = wxURI_IPVFUTURE; |
452 | ||
c4dbb953 | 453 | m_server.assign(start + 1, uri - start - 1); |
2186321f VZ |
454 | ++uri; |
455 | } | |
456 | else // unrecognized IP literal | |
457 | { | |
458 | uri = start; | |
dd65d8c8 | 459 | } |
dd65d8c8 RN |
460 | } |
461 | } | |
2186321f | 462 | else // IPv4 or a reg-name |
dd65d8c8 RN |
463 | { |
464 | if (ParseIPv4address(uri)) | |
465 | { | |
466 | m_hostType = wxURI_IPV4ADDRESS; | |
467 | ||
c4dbb953 | 468 | m_server.assign(start, uri - start); |
dd65d8c8 | 469 | } |
846978d7 | 470 | else |
2186321f VZ |
471 | { |
472 | uri = start; | |
473 | } | |
dd65d8c8 RN |
474 | } |
475 | ||
2186321f | 476 | if ( m_hostType == wxURI_REGNAME ) |
dd65d8c8 | 477 | { |
2186321f | 478 | uri = start; |
dd65d8c8 | 479 | // reg-name = *( unreserved / pct-encoded / sub-delims ) |
2186321f | 480 | while ( *uri && *uri != '/' && *uri != ':' && *uri != '#' && *uri != '?' ) |
dd65d8c8 | 481 | { |
2186321f | 482 | if ( IsUnreserved(*uri) || IsSubDelim(*uri) ) |
24ca04e7 | 483 | m_server += *uri++; |
dd65d8c8 | 484 | else |
2186321f | 485 | AppendNextEscaped(m_server, uri); |
846978d7 | 486 | } |
dd65d8c8 RN |
487 | } |
488 | ||
dd65d8c8 RN |
489 | m_fields |= wxURI_SERVER; |
490 | ||
491 | return uri; | |
492 | } | |
493 | ||
846978d7 | 494 | |
2186321f | 495 | const char* wxURI::ParsePort(const char* uri) |
dd65d8c8 | 496 | { |
dd65d8c8 | 497 | // port = *DIGIT |
2186321f | 498 | if( *uri == ':' ) |
dd65d8c8 RN |
499 | { |
500 | ++uri; | |
2186321f | 501 | while ( IsDigit(*uri) ) |
dd65d8c8 RN |
502 | { |
503 | m_port += *uri++; | |
846978d7 | 504 | } |
dd65d8c8 | 505 | |
dd65d8c8 RN |
506 | m_fields |= wxURI_PORT; |
507 | } | |
508 | ||
509 | return uri; | |
510 | } | |
511 | ||
2186321f | 512 | const char* wxURI::ParsePath(const char* uri) |
dd65d8c8 | 513 | { |
dd65d8c8 RN |
514 | /// hier-part = "//" authority path-abempty |
515 | /// / path-absolute | |
516 | /// / path-rootless | |
517 | /// / path-empty | |
518 | /// | |
519 | /// relative-part = "//" authority path-abempty | |
520 | /// / path-absolute | |
521 | /// / path-noscheme | |
522 | /// / path-empty | |
523 | /// | |
524 | /// path-abempty = *( "/" segment ) | |
525 | /// path-absolute = "/" [ segment-nz *( "/" segment ) ] | |
526 | /// path-noscheme = segment-nz-nc *( "/" segment ) | |
527 | /// path-rootless = segment-nz *( "/" segment ) | |
528 | /// path-empty = 0<pchar> | |
529 | /// | |
530 | /// segment = *pchar | |
531 | /// segment-nz = 1*pchar | |
532 | /// segment-nz-nc = 1*( unreserved / pct-encoded / sub-delims / "@" ) | |
533 | /// ; non-zero-length segment without any colon ":" | |
534 | /// | |
535 | /// pchar = unreserved / pct-encoded / sub-delims / ":" / "@" | |
dd65d8c8 | 536 | |
2186321f VZ |
537 | if ( IsEndPath(*uri) ) |
538 | return uri; | |
dd65d8c8 | 539 | |
2186321f VZ |
540 | const bool isAbs = *uri == '/'; |
541 | if ( isAbs ) | |
542 | m_path += *uri++; | |
543 | ||
544 | wxArrayString segments; | |
545 | wxString segment; | |
546 | for ( ;; ) | |
dd65d8c8 | 547 | { |
2186321f VZ |
548 | const bool endPath = IsEndPath(*uri); |
549 | if ( endPath || *uri == '/' ) | |
dd65d8c8 | 550 | { |
2186321f VZ |
551 | // end of a segment, look at what we got |
552 | if ( segment == ".." ) | |
dd65d8c8 | 553 | { |
2186321f VZ |
554 | if ( !segments.empty() && *segments.rbegin() != ".." ) |
555 | segments.pop_back(); | |
556 | else if ( !isAbs ) | |
557 | segments.push_back(".."); | |
dd65d8c8 | 558 | } |
2186321f | 559 | else if ( segment == "." ) |
dd65d8c8 | 560 | { |
2186321f VZ |
561 | // normally we ignore "." but the last one should be taken into |
562 | // account as "path/." is the same as "path/" and not just "path" | |
563 | if ( endPath ) | |
564 | segments.push_back(""); | |
dd65d8c8 | 565 | } |
2186321f | 566 | else // normal segment |
dd65d8c8 | 567 | { |
2186321f | 568 | segments.push_back(segment); |
dd65d8c8 RN |
569 | } |
570 | ||
2186321f VZ |
571 | if ( endPath ) |
572 | break; | |
573 | ||
574 | segment.clear(); | |
575 | ++uri; | |
576 | continue; | |
dd65d8c8 | 577 | } |
2186321f VZ |
578 | |
579 | if ( IsUnreserved(*uri) || IsSubDelim(*uri) || *uri == ':' || *uri == '@' ) | |
580 | segment += *uri++; | |
581 | else | |
582 | AppendNextEscaped(segment, uri); | |
dd65d8c8 RN |
583 | } |
584 | ||
2186321f VZ |
585 | m_path += wxJoin(segments, '/', '\0'); |
586 | m_fields |= wxURI_PATH; | |
587 | ||
dd65d8c8 RN |
588 | return uri; |
589 | } | |
590 | ||
591 | ||
2186321f | 592 | const char* wxURI::ParseQuery(const char* uri) |
dd65d8c8 | 593 | { |
dd65d8c8 | 594 | // query = *( pchar / "/" / "?" ) |
2186321f | 595 | if ( *uri == '?' ) |
dd65d8c8 RN |
596 | { |
597 | ++uri; | |
2186321f | 598 | while ( *uri && *uri != '#' ) |
dd65d8c8 | 599 | { |
2186321f VZ |
600 | if ( IsUnreserved(*uri) || IsSubDelim(*uri) || |
601 | *uri == ':' || *uri == '@' || *uri == '/' || *uri == '?' ) | |
602 | m_query += *uri++; | |
dd65d8c8 | 603 | else |
2186321f | 604 | AppendNextEscaped(m_query, uri); |
dd65d8c8 RN |
605 | } |
606 | ||
dd65d8c8 RN |
607 | m_fields |= wxURI_QUERY; |
608 | } | |
609 | ||
610 | return uri; | |
611 | } | |
612 | ||
613 | ||
2186321f | 614 | const char* wxURI::ParseFragment(const char* uri) |
dd65d8c8 | 615 | { |
dd65d8c8 | 616 | // fragment = *( pchar / "/" / "?" ) |
2186321f | 617 | if ( *uri == '#' ) |
dd65d8c8 RN |
618 | { |
619 | ++uri; | |
2186321f | 620 | while ( *uri ) |
dd65d8c8 | 621 | { |
2186321f VZ |
622 | if ( IsUnreserved(*uri) || IsSubDelim(*uri) || |
623 | *uri == ':' || *uri == '@' || *uri == '/' || *uri == '?') | |
624 | m_fragment += *uri++; | |
dd65d8c8 | 625 | else |
2186321f | 626 | AppendNextEscaped(m_fragment, uri); |
dd65d8c8 RN |
627 | } |
628 | ||
dd65d8c8 RN |
629 | m_fields |= wxURI_FRAGMENT; |
630 | } | |
631 | ||
632 | return uri; | |
633 | } | |
634 | ||
635 | // --------------------------------------------------------------------------- | |
ce321570 | 636 | // Resolve |
dd65d8c8 | 637 | // |
ce321570 | 638 | // Builds missing components of this uri from a base uri |
dd65d8c8 | 639 | // |
ce321570 RN |
640 | // A version of the algorithm outlined in the RFC is used here |
641 | // (it is shown in comments) | |
642 | // | |
846978d7 | 643 | // Note that an empty URI inherits all components |
dd65d8c8 RN |
644 | // --------------------------------------------------------------------------- |
645 | ||
2186321f VZ |
646 | /* static */ |
647 | wxArrayString wxURI::SplitInSegments(const wxString& path) | |
648 | { | |
649 | return wxSplit(path, '/', '\0' /* no escape character */); | |
650 | } | |
651 | ||
8404931e | 652 | void wxURI::Resolve(const wxURI& base, int flags) |
dd65d8c8 | 653 | { |
846978d7 | 654 | wxASSERT_MSG(!base.IsReference(), |
2186321f | 655 | "wxURI to inherit from must not be a reference!"); |
dd65d8c8 | 656 | |
2186321f VZ |
657 | // If we aren't being strict, enable the older (pre-RFC2396) loophole that |
658 | // allows this uri to inherit other properties from the base uri - even if | |
659 | // the scheme is defined | |
8404931e VZ |
660 | if ( !(flags & wxURI_STRICT) && |
661 | HasScheme() && base.HasScheme() && | |
662 | m_scheme == base.m_scheme ) | |
846978d7 | 663 | { |
dd65d8c8 RN |
664 | m_fields -= wxURI_SCHEME; |
665 | } | |
666 | ||
667 | ||
668 | // Do nothing if this is an absolute wxURI | |
669 | // if defined(R.scheme) then | |
670 | // T.scheme = R.scheme; | |
671 | // T.authority = R.authority; | |
672 | // T.path = remove_dot_segments(R.path); | |
673 | // T.query = R.query; | |
674 | if (HasScheme()) | |
dd65d8c8 | 675 | return; |
dd65d8c8 | 676 | |
ea4daac4 | 677 | //No scheme - inherit |
dd65d8c8 RN |
678 | m_scheme = base.m_scheme; |
679 | m_fields |= wxURI_SCHEME; | |
680 | ||
681 | // All we need to do for relative URIs with an | |
682 | // authority component is just inherit the scheme | |
683 | // if defined(R.authority) then | |
684 | // T.authority = R.authority; | |
685 | // T.path = remove_dot_segments(R.path); | |
686 | // T.query = R.query; | |
687 | if (HasServer()) | |
dd65d8c8 | 688 | return; |
dd65d8c8 RN |
689 | |
690 | //No authority - inherit | |
4860d40d | 691 | if (base.HasUserInfo()) |
dd65d8c8 | 692 | { |
4860d40d RN |
693 | m_userinfo = base.m_userinfo; |
694 | m_fields |= wxURI_USERINFO; | |
dd65d8c8 | 695 | } |
846978d7 | 696 | |
dd65d8c8 RN |
697 | m_server = base.m_server; |
698 | m_hostType = base.m_hostType; | |
699 | m_fields |= wxURI_SERVER; | |
846978d7 | 700 | |
dd65d8c8 RN |
701 | if (base.HasPort()) |
702 | { | |
703 | m_port = base.m_port; | |
704 | m_fields |= wxURI_PORT; | |
705 | } | |
846978d7 | 706 | |
dd65d8c8 RN |
707 | |
708 | // Simple path inheritance from base | |
709 | if (!HasPath()) | |
710 | { | |
711 | // T.path = Base.path; | |
712 | m_path = base.m_path; | |
713 | m_fields |= wxURI_PATH; | |
846978d7 | 714 | |
dd65d8c8 RN |
715 | |
716 | // if defined(R.query) then | |
717 | // T.query = R.query; | |
718 | // else | |
719 | // T.query = Base.query; | |
720 | // endif; | |
721 | if (!HasQuery()) | |
722 | { | |
723 | m_query = base.m_query; | |
724 | m_fields |= wxURI_QUERY; | |
725 | } | |
726 | } | |
2186321f | 727 | else if ( m_path.empty() || m_path[0u] != '/' ) |
dd65d8c8 RN |
728 | { |
729 | // if (R.path starts-with "/") then | |
730 | // T.path = remove_dot_segments(R.path); | |
731 | // else | |
732 | // T.path = merge(Base.path, R.path); | |
733 | // T.path = remove_dot_segments(T.path); | |
734 | // endif; | |
735 | // T.query = R.query; | |
2186321f VZ |
736 | // |
737 | // So we don't do anything for absolute paths and implement merge for | |
738 | // the relative ones | |
c9f78968 | 739 | |
2186321f VZ |
740 | wxArrayString our(SplitInSegments(m_path)), |
741 | result(SplitInSegments(base.m_path)); | |
c9f78968 | 742 | |
2186321f VZ |
743 | if ( !result.empty() ) |
744 | result.pop_back(); | |
846978d7 | 745 | |
2186321f | 746 | if ( our.empty() ) |
dd65d8c8 | 747 | { |
2186321f VZ |
748 | // if we have an empty path it means we were constructed from a "." |
749 | // string or something similar (e.g. "././././"), it should count | |
750 | // as (empty) segment | |
751 | our.push_back(""); | |
dd65d8c8 | 752 | } |
dd65d8c8 | 753 | |
2186321f VZ |
754 | const wxArrayString::const_iterator end = our.end(); |
755 | for ( wxArrayString::const_iterator i = our.begin(); i != end; ++i ) | |
dd65d8c8 | 756 | { |
2186321f | 757 | if ( i->empty() || *i == "." ) |
dd65d8c8 | 758 | { |
2186321f VZ |
759 | // as in ParsePath(), while normally we ignore the empty |
760 | // segments, we need to take account of them at the end | |
761 | if ( i == end - 1 ) | |
762 | result.push_back(""); | |
763 | continue; | |
dd65d8c8 | 764 | } |
dd65d8c8 | 765 | |
2186321f | 766 | if ( *i == ".." ) |
dd65d8c8 | 767 | { |
2186321f VZ |
768 | if ( !result.empty() ) |
769 | { | |
770 | result.pop_back(); | |
dd65d8c8 | 771 | |
2186321f VZ |
772 | if ( i == end - 1 ) |
773 | result.push_back(""); | |
774 | } | |
775 | //else: just ignore, extra ".." don't accumulate | |
dd65d8c8 RN |
776 | } |
777 | else | |
778 | { | |
2186321f VZ |
779 | if ( result.empty() ) |
780 | { | |
781 | // ensure that the resulting path will always be absolute | |
782 | result.push_back(""); | |
783 | } | |
784 | ||
785 | result.push_back(*i); | |
dd65d8c8 RN |
786 | } |
787 | } | |
2186321f VZ |
788 | |
789 | m_path = wxJoin(result, '/', '\0'); | |
dd65d8c8 RN |
790 | } |
791 | ||
2186321f | 792 | //T.fragment = R.fragment; |
dd65d8c8 RN |
793 | } |
794 | ||
795 | // --------------------------------------------------------------------------- | |
ce321570 RN |
796 | // ParseH16 |
797 | // | |
798 | // Parses 1 to 4 hex values. Returns true if the first character of the input | |
2186321f | 799 | // string is a valid hex character. It is the caller's responsibility to move |
ce321570 RN |
800 | // the input string back to its original position on failure. |
801 | // --------------------------------------------------------------------------- | |
802 | ||
2186321f | 803 | bool wxURI::ParseH16(const char*& uri) |
ce321570 RN |
804 | { |
805 | // h16 = 1*4HEXDIG | |
806 | if(!IsHex(*++uri)) | |
807 | return false; | |
808 | ||
809 | if(IsHex(*++uri) && IsHex(*++uri) && IsHex(*++uri)) | |
810 | ++uri; | |
811 | ||
812 | return true; | |
813 | } | |
814 | ||
815 | // --------------------------------------------------------------------------- | |
816 | // ParseIPXXX | |
817 | // | |
846978d7 WS |
818 | // Parses a certain version of an IP address and moves the input string past |
819 | // it. Returns true if the input string contains the proper version of an ip | |
37424888 | 820 | // address. It is the caller's responsibility to move the input string back |
ce321570 | 821 | // to its original position on failure. |
dd65d8c8 RN |
822 | // --------------------------------------------------------------------------- |
823 | ||
2186321f | 824 | bool wxURI::ParseIPv4address(const char*& uri) |
dd65d8c8 RN |
825 | { |
826 | //IPv4address = dec-octet "." dec-octet "." dec-octet "." dec-octet | |
827 | // | |
828 | //dec-octet = DIGIT ; 0-9 | |
829 | // / %x31-39 DIGIT ; 10-99 | |
830 | // / "1" 2DIGIT ; 100-199 | |
831 | // / "2" %x30-34 DIGIT ; 200-249 | |
832 | // / "25" %x30-35 ; 250-255 | |
833 | size_t iIPv4 = 0; | |
834 | if (IsDigit(*uri)) | |
835 | { | |
836 | ++iIPv4; | |
837 | ||
846978d7 | 838 | |
dd65d8c8 RN |
839 | //each ip part must be between 0-255 (dupe of version in for loop) |
840 | if( IsDigit(*++uri) && IsDigit(*++uri) && | |
841 | //100 or less (note !) | |
2186321f | 842 | !( (*(uri-2) < '2') || |
846978d7 | 843 | //240 or less |
2186321f VZ |
844 | (*(uri-2) == '2' && |
845 | (*(uri-1) < '5' || (*(uri-1) == '5' && *uri <= '5')) | |
dd65d8c8 RN |
846 | ) |
847 | ) | |
848 | ) | |
849 | { | |
850 | return false; | |
851 | } | |
852 | ||
853 | if(IsDigit(*uri))++uri; | |
854 | ||
855 | //compilers should unroll this loop | |
856 | for(; iIPv4 < 4; ++iIPv4) | |
857 | { | |
2186321f | 858 | if (*uri != '.' || !IsDigit(*++uri)) |
dd65d8c8 RN |
859 | break; |
860 | ||
861 | //each ip part must be between 0-255 | |
862 | if( IsDigit(*++uri) && IsDigit(*++uri) && | |
863 | //100 or less (note !) | |
2186321f | 864 | !( (*(uri-2) < '2') || |
846978d7 | 865 | //240 or less |
2186321f VZ |
866 | (*(uri-2) == '2' && |
867 | (*(uri-1) < '5' || (*(uri-1) == '5' && *uri <= '5')) | |
dd65d8c8 RN |
868 | ) |
869 | ) | |
870 | ) | |
871 | { | |
872 | return false; | |
873 | } | |
874 | if(IsDigit(*uri))++uri; | |
875 | } | |
876 | } | |
877 | return iIPv4 == 4; | |
878 | } | |
879 | ||
2186321f | 880 | bool wxURI::ParseIPv6address(const char*& uri) |
dd65d8c8 RN |
881 | { |
882 | // IPv6address = 6( h16 ":" ) ls32 | |
883 | // / "::" 5( h16 ":" ) ls32 | |
884 | // / [ h16 ] "::" 4( h16 ":" ) ls32 | |
885 | // / [ *1( h16 ":" ) h16 ] "::" 3( h16 ":" ) ls32 | |
886 | // / [ *2( h16 ":" ) h16 ] "::" 2( h16 ":" ) ls32 | |
887 | // / [ *3( h16 ":" ) h16 ] "::" h16 ":" ls32 | |
888 | // / [ *4( h16 ":" ) h16 ] "::" ls32 | |
889 | // / [ *5( h16 ":" ) h16 ] "::" h16 | |
890 | // / [ *6( h16 ":" ) h16 ] "::" | |
891 | ||
892 | size_t numPrefix = 0, | |
893 | maxPostfix; | |
894 | ||
895 | bool bEndHex = false; | |
896 | ||
897 | for( ; numPrefix < 6; ++numPrefix) | |
898 | { | |
899 | if(!ParseH16(uri)) | |
900 | { | |
901 | --uri; | |
902 | bEndHex = true; | |
903 | break; | |
904 | } | |
846978d7 | 905 | |
2186321f | 906 | if(*uri != ':') |
dd65d8c8 RN |
907 | { |
908 | break; | |
909 | } | |
910 | } | |
911 | ||
912 | if(!bEndHex && !ParseH16(uri)) | |
913 | { | |
914 | --uri; | |
915 | ||
916 | if (numPrefix) | |
917 | return false; | |
918 | ||
2186321f | 919 | if (*uri == ':') |
dd65d8c8 | 920 | { |
2186321f | 921 | if (*++uri != ':') |
dd65d8c8 RN |
922 | return false; |
923 | ||
924 | maxPostfix = 5; | |
925 | } | |
926 | else | |
927 | maxPostfix = 6; | |
928 | } | |
929 | else | |
930 | { | |
2186321f | 931 | if (*uri != ':' || *(uri+1) != ':') |
dd65d8c8 RN |
932 | { |
933 | if (numPrefix != 6) | |
934 | return false; | |
935 | ||
2186321f | 936 | while (*--uri != ':') {} |
dd65d8c8 RN |
937 | ++uri; |
938 | ||
2186321f | 939 | const char * const start = uri; |
dd65d8c8 RN |
940 | //parse ls32 |
941 | // ls32 = ( h16 ":" h16 ) / IPv4address | |
2186321f | 942 | if (ParseH16(uri) && *uri == ':' && ParseH16(uri)) |
dd65d8c8 RN |
943 | return true; |
944 | ||
2186321f | 945 | uri = start; |
dd65d8c8 RN |
946 | |
947 | if (ParseIPv4address(uri)) | |
948 | return true; | |
949 | else | |
950 | return false; | |
951 | } | |
952 | else | |
953 | { | |
954 | uri += 2; | |
846978d7 | 955 | |
dd65d8c8 RN |
956 | if (numPrefix > 3) |
957 | maxPostfix = 0; | |
958 | else | |
959 | maxPostfix = 4 - numPrefix; | |
960 | } | |
961 | } | |
962 | ||
963 | bool bAllowAltEnding = maxPostfix == 0; | |
964 | ||
965 | for(; maxPostfix != 0; --maxPostfix) | |
966 | { | |
2186321f | 967 | if(!ParseH16(uri) || *uri != ':') |
dd65d8c8 RN |
968 | return false; |
969 | } | |
970 | ||
971 | if(numPrefix <= 4) | |
972 | { | |
2186321f | 973 | const char * const start = uri; |
dd65d8c8 RN |
974 | //parse ls32 |
975 | // ls32 = ( h16 ":" h16 ) / IPv4address | |
2186321f | 976 | if (ParseH16(uri) && *uri == ':' && ParseH16(uri)) |
dd65d8c8 RN |
977 | return true; |
978 | ||
2186321f | 979 | uri = start; |
dd65d8c8 RN |
980 | |
981 | if (ParseIPv4address(uri)) | |
982 | return true; | |
983 | ||
2186321f | 984 | uri = start; |
846978d7 | 985 | |
dd65d8c8 RN |
986 | if (!bAllowAltEnding) |
987 | return false; | |
988 | } | |
989 | ||
990 | if(numPrefix <= 5 && ParseH16(uri)) | |
991 | return true; | |
992 | ||
993 | return true; | |
994 | } | |
995 | ||
2186321f | 996 | bool wxURI::ParseIPvFuture(const char*& uri) |
dd65d8c8 RN |
997 | { |
998 | // IPvFuture = "v" 1*HEXDIG "." 1*( unreserved / sub-delims / ":" ) | |
2186321f | 999 | if (*++uri != 'v' || !IsHex(*++uri)) |
dd65d8c8 RN |
1000 | return false; |
1001 | ||
2186321f VZ |
1002 | while (IsHex(*++uri)) |
1003 | ; | |
dd65d8c8 | 1004 | |
2186321f | 1005 | if (*uri != '.' || !(IsUnreserved(*++uri) || IsSubDelim(*uri) || *uri == ':')) |
dd65d8c8 RN |
1006 | return false; |
1007 | ||
2186321f | 1008 | while(IsUnreserved(*++uri) || IsSubDelim(*uri) || *uri == ':') {} |
dd65d8c8 RN |
1009 | |
1010 | return true; | |
1011 | } | |
1012 | ||
1013 | ||
ce321570 RN |
1014 | // --------------------------------------------------------------------------- |
1015 | // IsXXX | |
1016 | // | |
1017 | // Returns true if the passed in character meets the criteria of the method | |
1018 | // --------------------------------------------------------------------------- | |
1019 | ||
2186321f VZ |
1020 | // unreserved = ALPHA / DIGIT / "-" / "." / "_" / "~" |
1021 | bool wxURI::IsUnreserved(char c) | |
1022 | { | |
1023 | return IsAlpha(c) || | |
1024 | IsDigit(c) || | |
1025 | c == '-' || | |
1026 | c == '.' || | |
1027 | c == '_' || | |
1028 | c == '~' | |
846978d7 | 1029 | ; |
dd65d8c8 RN |
1030 | } |
1031 | ||
2186321f | 1032 | bool wxURI::IsReserved(char c) |
846978d7 | 1033 | { |
dd65d8c8 RN |
1034 | return IsGenDelim(c) || IsSubDelim(c); |
1035 | } | |
1036 | ||
2186321f VZ |
1037 | // gen-delims = ":" / "/" / "?" / "#" / "[" / "]" / "@" |
1038 | bool wxURI::IsGenDelim(char c) | |
dd65d8c8 | 1039 | { |
2186321f VZ |
1040 | return c == ':' || |
1041 | c == '/' || | |
1042 | c == '?' || | |
1043 | c == '#' || | |
1044 | c == '[' || | |
1045 | c == ']' || | |
1046 | c == '@'; | |
dd65d8c8 RN |
1047 | } |
1048 | ||
2186321f VZ |
1049 | // sub-delims = "!" / "$" / "&" / "'" / "(" / ")" |
1050 | // / "*" / "+" / "," / ";" / "=" | |
1051 | bool wxURI::IsSubDelim(char c) | |
dd65d8c8 | 1052 | { |
2186321f VZ |
1053 | return c == '!' || |
1054 | c == '$' || | |
1055 | c == '&' || | |
1056 | c == '\'' || | |
1057 | c == '(' || | |
1058 | c == ')' || | |
1059 | c == '*' || | |
1060 | c == '+' || | |
1061 | c == ',' || | |
1062 | c == ';' || | |
1063 | c == '=' | |
dd65d8c8 RN |
1064 | ; |
1065 | } | |
1066 | ||
2186321f VZ |
1067 | bool wxURI::IsHex(char c) |
1068 | { | |
1069 | return IsDigit(c) || | |
1070 | (c >= 'a' && c <= 'f') || | |
1071 | (c >= 'A' && c <= 'F'); | |
1072 | } | |
dd65d8c8 | 1073 | |
2186321f VZ |
1074 | bool wxURI::IsAlpha(char c) |
1075 | { | |
1076 | return (c >= 'a' && c <= 'z') || (c >= 'A' && c <= 'Z'); | |
1077 | } | |
dd65d8c8 | 1078 | |
2186321f VZ |
1079 | bool wxURI::IsDigit(char c) |
1080 | { | |
1081 | return c >= '0' && c <= '9'; | |
1082 | } | |
dd65d8c8 | 1083 | |
2186321f VZ |
1084 | bool wxURI::IsEndPath(char c) |
1085 | { | |
1086 | return c == '\0' || c == '#' || c == '?'; | |
1087 | } | |
dd65d8c8 | 1088 |