1 |
/* ============================================================================ |
2 |
* Douglas Thrift's Search Engine License |
3 |
* |
4 |
* Copyright (C) 2002-2003, Douglas Thrift. All Rights Reserved. |
5 |
* Redistribution and use in source and binary forms, with or without |
6 |
* modification, are permitted provided that the following conditions are met: |
7 |
* |
8 |
* 1. Redistributions of source code must retain the above copyright notice, |
9 |
* this list of conditions and the following disclaimer. |
10 |
* |
11 |
* 2. Redistributions in binary form must reproduce the above copyright notice, |
12 |
* this list of conditions and the following disclaimer in the documentation |
13 |
* and/or other materials provided with the distribution. |
14 |
* |
15 |
* 3. The end-user documentation included with the redistribution, if any, must |
16 |
* include the following acknowledgment: |
17 |
* |
18 |
* "This product includes software developed by Douglas Thrift |
19 |
* (http://computers.douglasthrift.net/searchengine/)." |
20 |
* |
21 |
* Alternately, this acknowledgment may appear in the software itself, if |
22 |
* and wherever such third-party acknowledgments normally appear. |
23 |
* |
24 |
* 4. The names "Douglas Thrift" and "Douglas Thrift's Search Engine" must not |
25 |
* be used to endorse or promote products derived from this software without |
26 |
* specific prior written permission. For written permission, please visit |
27 |
* http://www.douglasthrift.net/contact.cgi for contact information. |
28 |
* |
29 |
* 5. Products derived from this software may not be called "Douglas Thrift's |
30 |
* Search Engine", nor may "Douglas Thrift's Search Engine" appear in their |
31 |
* name, without prior written permission. |
32 |
* |
33 |
* THIS SOFTWARE IS PROVIDED "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, |
34 |
* INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND |
35 |
* FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE |
36 |
* COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, |
37 |
* INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT |
38 |
* LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, |
39 |
* OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF |
40 |
* LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING |
41 |
* NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, |
42 |
* EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. |
43 |
* ============================================================================ |
44 |
*/ |
45 |
// Douglas Thrift's Search Engine HTTP Handler |
46 |
// |
47 |
// Douglas Thrift |
48 |
// |
49 |
// $Id: HttpHandler.cpp,v 1.19 2003/07/15 08:01:00 douglas Exp $ |
50 |
|
51 |
#include "HttpHandler.h" |
52 |
|
53 |
// Lovely C Sockets! |
54 |
#ifndef _WIN32 |
55 |
// BSD Sockets |
56 |
#include <unistd.h> |
57 |
#include <sys/types.h> |
58 |
#include <sys/socket.h> |
59 |
#include <netinet/in.h> |
60 |
#include <netdb.h> |
61 |
|
62 |
#define INVALID_SOCKET -1 |
63 |
#define SOCKET_ERROR -1 |
64 |
|
65 |
inline int closesocket(SOCKET s) { return close(s); } |
66 |
#endif |
67 |
|
68 |
HttpHandler::HttpHandler() |
69 |
{ |
70 |
buffer = new char[BUFSIZ + 1]; |
71 |
|
72 |
#ifdef _WIN32 |
73 |
if (WSAStartup(MAKEWORD(2, 0), &data) != 0) |
74 |
{ |
75 |
error(program + ": WSAStartup"); |
76 |
exit(1); |
77 |
} |
78 |
#endif // _WIN32 |
79 |
|
80 |
length = 0; |
81 |
chunked = false; |
82 |
#ifdef _OpenSSL_ |
83 |
tls = false; |
84 |
#endif |
85 |
} |
86 |
|
87 |
HttpHandler::~HttpHandler() |
88 |
{ |
89 |
delete [] buffer; |
90 |
|
91 |
#ifdef _WIN32 |
92 |
WSACleanup(); |
93 |
#endif // _WIN32 |
94 |
} |
95 |
|
96 |
bool HttpHandler::handle(URL &url, const string referer, bool head) |
97 |
{ |
98 |
bool answer = false; |
99 |
|
100 |
if ((http = socket(PF_INET, SOCK_STREAM, 0)) == INVALID_SOCKET) |
101 |
{ |
102 |
error(program + ": Socket"); |
103 |
exit(1); |
104 |
} |
105 |
|
106 |
sockaddr_in address; |
107 |
hostent* host; |
108 |
|
109 |
address.sin_family = AF_INET; |
110 |
|
111 |
if ((host = gethostbyname(url.getAddress().c_str())) == NULL) |
112 |
{ |
113 |
error(program + ": Host: " + url.getAddress(), true); |
114 |
return answer; |
115 |
} |
116 |
|
117 |
address.sin_addr = *((in_addr*)*host->h_addr_list); |
118 |
address.sin_port = htons(url.getPort()); |
119 |
|
120 |
if (connect(http, (sockaddr*)&address, sizeof(sockaddr_in)) == |
121 |
SOCKET_ERROR) |
122 |
{ |
123 |
error(program + ": Connect"); |
124 |
return answer; |
125 |
} |
126 |
|
127 |
if (head) |
128 |
{ |
129 |
putline("HEAD " + url.getPath() + " HTTP/1.1"); |
130 |
} |
131 |
else |
132 |
{ |
133 |
putline("GET " + url.getPath() + " HTTP/1.1"); |
134 |
} |
135 |
|
136 |
putline("Accept: text/html; text/plain"); |
137 |
#ifndef _OpenSSL_ |
138 |
putline("User-Agent: " + agent(true) + ' ' + platform()); |
139 |
#else |
140 |
putline("User-Agent: " + agent(true) + ' ' + platform() + ' ' |
141 |
+ openssl(true)); |
142 |
#endif |
143 |
|
144 |
if (url.getPort() == 80) |
145 |
{ |
146 |
putline("Host: " + url.getAddress()); |
147 |
} |
148 |
else |
149 |
{ |
150 |
char* port = new char[1024]; |
151 |
sprintf(port, "%u", url.getPort()); |
152 |
|
153 |
putline("Host: " + url.getAddress() + ':' + port); |
154 |
|
155 |
delete [] port; |
156 |
} |
157 |
|
158 |
if (referer != "") |
159 |
{ |
160 |
putline("Referer: " + referer); |
161 |
} |
162 |
|
163 |
putline("Connection: close"); |
164 |
putline(); |
165 |
|
166 |
code response; |
167 |
string line; |
168 |
|
169 |
do |
170 |
{ |
171 |
line = getline(); |
172 |
|
173 |
if (line.find("HTTP/") != 0) |
174 |
{ |
175 |
return answer; |
176 |
} |
177 |
|
178 |
unsigned dot = line.find('.'); |
179 |
unsigned space = line.find(' '); |
180 |
|
181 |
unsigned major = strtoul(line.substr(5, dot - 5).c_str(), 0, 10); |
182 |
unsigned minor = strtoul(line.substr(dot + 1, space - dot - 1).c_str(), |
183 |
0, 10); |
184 |
|
185 |
if (major > 1) |
186 |
{ |
187 |
cerr << program << ": Potentially Incompatible Server: HTTP/" << |
188 |
major << "." << minor << "\n"; |
189 |
|
190 |
return answer; |
191 |
} |
192 |
|
193 |
response = code(strtoul(line.substr(space + 1).c_str(), 0, 10)); |
194 |
|
195 |
if (response < ok) do line = getline(); while (line != ""); |
196 |
} |
197 |
while (response < ok); |
198 |
|
199 |
do |
200 |
{ |
201 |
line = getline(); |
202 |
|
203 |
if (line != "") |
204 |
{ |
205 |
unsigned colon = line.find(':'); |
206 |
|
207 |
string field = line.substr(0, colon); |
208 |
string value = line.substr(colon + 1); |
209 |
|
210 |
while (isspace(value[0])) value.erase(0, 1); |
211 |
|
212 |
if (field == "Content-Type") |
213 |
{ |
214 |
type = value; |
215 |
} |
216 |
else if (field == "Content-Length") |
217 |
{ |
218 |
length = strtoul(value.c_str(), 0, 10); |
219 |
} |
220 |
else if (field == "Location") |
221 |
{ |
222 |
location = value; |
223 |
} |
224 |
else if (field == "Transfer-Encoding") |
225 |
{ |
226 |
chunked = value == "chunked"; |
227 |
} |
228 |
} |
229 |
} |
230 |
while (line != ""); |
231 |
|
232 |
switch (response) |
233 |
{ |
234 |
case ok: |
235 |
if (debug) cerr << "response = " << response << "\n"; |
236 |
answer = true; |
237 |
break; |
238 |
case choices: |
239 |
case moved: |
240 |
case found: |
241 |
if (debug) cerr << "response = " << response << "\n" |
242 |
<< "location = " << location << "\n"; |
243 |
location = getLink(location, url); |
244 |
break; |
245 |
case notfound: |
246 |
case internal: |
247 |
if (debug) cerr << "response = " << response << "\n"; |
248 |
break; |
249 |
default: |
250 |
if (debug) cerr << "response = " << response << "\n"; |
251 |
if (response <= 299) |
252 |
{ |
253 |
answer = true; |
254 |
} |
255 |
else if (response <= 399) |
256 |
{ |
257 |
location = getLink(location, url); |
258 |
} |
259 |
break; |
260 |
} |
261 |
|
262 |
if (!head && answer) populate(); |
263 |
|
264 |
return answer; |
265 |
} |
266 |
|
267 |
HttpHandler& HttpHandler::getline(string& line, char endline) |
268 |
{ |
269 |
unsigned end = page.find(endline); |
270 |
unsigned newline = page.find('\n'); |
271 |
|
272 |
if (newline < end || end == string::npos) |
273 |
{ |
274 |
end = newline; |
275 |
} |
276 |
|
277 |
line = page.substr(0, end); |
278 |
page.erase(0, (end == string::npos ? end : end + 1)); |
279 |
|
280 |
return *this; |
281 |
} |
282 |
|
283 |
void HttpHandler::clear() |
284 |
{ |
285 |
closesocket(http); |
286 |
|
287 |
type = ""; |
288 |
length = 0; |
289 |
location = ""; |
290 |
page = ""; |
291 |
chunked = false; |
292 |
#ifdef _OpenSSL_ |
293 |
tls = false; |
294 |
#endif |
295 |
} |
296 |
|
297 |
void HttpHandler::populate() |
298 |
{ |
299 |
if (!chunked) |
300 |
{ |
301 |
unsigned left = length; |
302 |
|
303 |
while (left > 0) |
304 |
{ |
305 |
memset(buffer, 0, BUFSIZ + 1); |
306 |
|
307 |
unsigned bytes = left > BUFSIZ ? BUFSIZ : left; |
308 |
unsigned received; |
309 |
|
310 |
while (true) |
311 |
{ |
312 |
if ((received = recv(http, buffer, bytes, 0)) == SOCKET_ERROR) |
313 |
{ |
314 |
error(program + ": Recv"); |
315 |
exit(1); |
316 |
} |
317 |
else if (received != bytes) |
318 |
{ |
319 |
left -= received; |
320 |
page += buffer; |
321 |
|
322 |
memset(buffer, 0, BUFSIZ + 1); |
323 |
|
324 |
bytes -= received; |
325 |
} |
326 |
else |
327 |
{ |
328 |
break; |
329 |
} |
330 |
} |
331 |
|
332 |
page += buffer; |
333 |
left -= bytes; |
334 |
} |
335 |
} |
336 |
else |
337 |
{ |
338 |
unsigned chunk; |
339 |
|
340 |
do |
341 |
{ |
342 |
chunk = strtoul(getline().c_str(), 0, 16); |
343 |
|
344 |
unsigned left = chunk; |
345 |
|
346 |
while (left > 0) |
347 |
{ |
348 |
memset(buffer, 0, BUFSIZ + 1); |
349 |
|
350 |
unsigned bytes = left > BUFSIZ ? BUFSIZ : left; |
351 |
unsigned received; |
352 |
|
353 |
while (true) |
354 |
{ |
355 |
if ((received = recv(http, buffer, bytes, 0)) == |
356 |
SOCKET_ERROR) |
357 |
{ |
358 |
error(program + ": Recv"); |
359 |
exit(1); |
360 |
} |
361 |
else if (received != bytes) |
362 |
{ |
363 |
left -= received; |
364 |
page += buffer; |
365 |
|
366 |
memset(buffer, 0, BUFSIZ + 1); |
367 |
|
368 |
bytes -= received; |
369 |
} |
370 |
else |
371 |
{ |
372 |
break; |
373 |
} |
374 |
} |
375 |
|
376 |
page += buffer; |
377 |
left -= bytes; |
378 |
} |
379 |
|
380 |
getline(); |
381 |
length += chunk; |
382 |
} |
383 |
while (chunk > 0); |
384 |
} |
385 |
|
386 |
for (unsigned index = 0; index < page.length(); index++) |
387 |
{ |
388 |
if (page[index] == '\r' && (index + 1 < page.length()) ? page[index + |
389 |
1] == '\n' : false) |
390 |
{ |
391 |
page.erase(index, 1); |
392 |
} |
393 |
else if (page[index] == '\r') |
394 |
{ |
395 |
page[index] = '\n'; |
396 |
} |
397 |
} |
398 |
} |
399 |
|
400 |
void HttpHandler::putline(const string line) |
401 |
{ |
402 |
sprintf(buffer, "%s\r\n", line.c_str()); |
403 |
if (send(http, buffer, strlen(buffer), 0) == SOCKET_ERROR) |
404 |
{ |
405 |
error(program + ": Send"); |
406 |
exit(1); |
407 |
} |
408 |
} |
409 |
|
410 |
string HttpHandler::getline() |
411 |
{ |
412 |
string line; |
413 |
char byte; |
414 |
|
415 |
do |
416 |
{ |
417 |
if (recv(http, &byte, 1, 0) == SOCKET_ERROR) |
418 |
{ |
419 |
error(program + ": Recv"); |
420 |
} |
421 |
|
422 |
if (byte != '\r' && byte != '\n') |
423 |
{ |
424 |
line += byte; |
425 |
} |
426 |
} |
427 |
while (byte != '\n'); |
428 |
|
429 |
return line; |
430 |
} |
431 |
|
432 |
void HttpHandler::error(const string& prefix, bool host) |
433 |
{ |
434 |
#ifdef _WIN32 |
435 |
string error; |
436 |
|
437 |
switch (WSAGetLastError()) |
438 |
{ |
439 |
case WSAEACCES: |
440 |
error = "Permission denied"; |
441 |
break; |
442 |
case WSAEADDRINUSE: |
443 |
error = "Address already in use"; |
444 |
break; |
445 |
case WSAEADDRNOTAVAIL: |
446 |
error = "Cannot assign requested address"; |
447 |
break; |
448 |
case WSAEAFNOSUPPORT: |
449 |
error = "Address family not supported by protocol family"; |
450 |
break; |
451 |
case WSAEALREADY: |
452 |
error = "Operation already in progress"; |
453 |
break; |
454 |
case WSAECONNABORTED: |
455 |
error = "Software caused connection abort"; |
456 |
break; |
457 |
case WSAECONNREFUSED: |
458 |
error = "Connection refused"; |
459 |
break; |
460 |
case WSAECONNRESET: |
461 |
error = "Connection reset by peer"; |
462 |
break; |
463 |
case WSAEDESTADDRREQ: |
464 |
error = "Destination address required"; |
465 |
break; |
466 |
case WSAEFAULT: |
467 |
error = "Bad address"; |
468 |
break; |
469 |
case WSAEHOSTDOWN: |
470 |
error = "Host is down"; |
471 |
break; |
472 |
case WSAEHOSTUNREACH: |
473 |
error = "No route to host"; |
474 |
break; |
475 |
case WSAEINPROGRESS: |
476 |
error = "Operation now in progress"; |
477 |
break; |
478 |
case WSAEINTR: |
479 |
error = "Interrupted function call"; |
480 |
break; |
481 |
case WSAEINVAL: |
482 |
error = "Invalid argument"; |
483 |
break; |
484 |
case WSAEISCONN: |
485 |
error = "Socket is already connected"; |
486 |
break; |
487 |
case WSAEMFILE: |
488 |
error = "Too many open files"; |
489 |
break; |
490 |
case WSAEMSGSIZE: |
491 |
error = "Message too long"; |
492 |
break; |
493 |
case WSAENETDOWN: |
494 |
error = "Network is down"; |
495 |
break; |
496 |
case WSAENETRESET: |
497 |
error = "Network dropped connection on reset"; |
498 |
break; |
499 |
case WSAENETUNREACH: |
500 |
error = "Network is unreachable"; |
501 |
break; |
502 |
case WSAENOBUFS: |
503 |
error = "No buffer space available"; |
504 |
break; |
505 |
case WSAENOPROTOOPT: |
506 |
error = "Bad protocol option"; |
507 |
break; |
508 |
case WSAENOTCONN: |
509 |
error = "Socket is not connected"; |
510 |
break; |
511 |
case WSAENOTSOCK: |
512 |
error = "Socket operation on non-socket"; |
513 |
break; |
514 |
case WSAEOPNOTSUPP: |
515 |
error = "Operation not supported"; |
516 |
break; |
517 |
case WSAEPFNOSUPPORT: |
518 |
error = "Protocol family not supported"; |
519 |
break; |
520 |
case WSAEPROCLIM: |
521 |
error = "Too many processes"; |
522 |
break; |
523 |
case WSAEPROTONOSUPPORT: |
524 |
error = "Protocol not supported"; |
525 |
break; |
526 |
case WSAEPROTOTYPE: |
527 |
error = "Protocol wrong type for socket"; |
528 |
break; |
529 |
case WSAESHUTDOWN: |
530 |
error = "Cannot send after socket shutdown"; |
531 |
break; |
532 |
case WSAESOCKTNOSUPPORT: |
533 |
error = "Socket type not supported"; |
534 |
break; |
535 |
case WSAETIMEDOUT: |
536 |
error = "Connection timed out"; |
537 |
break; |
538 |
case WSATYPE_NOT_FOUND: |
539 |
error = "Class type not found"; |
540 |
break; |
541 |
case WSAEWOULDBLOCK: |
542 |
error = "Resource temporarily unavailable"; |
543 |
break; |
544 |
case WSAHOST_NOT_FOUND: |
545 |
error = "Host not found"; |
546 |
break; |
547 |
case WSA_INVALID_HANDLE: |
548 |
error = "Specified event object handle is invalid"; |
549 |
break; |
550 |
case WSA_INVALID_PARAMETER: |
551 |
error = "One or more parameters are invalid"; |
552 |
break; |
553 |
// case WSAINVALIDPROCTABLE: |
554 |
// error = "Invalid procedure table from service provider"; |
555 |
// break; |
556 |
// case WSAINVALIDPROVIDER: |
557 |
// error = "Invalid service provider version number"; |
558 |
// break; |
559 |
case WSA_IO_INCOMPLETE: |
560 |
error = "Overlapped I/O event object not in signaled state"; |
561 |
break; |
562 |
case WSA_IO_PENDING: |
563 |
error = "Overlapped operations will complete later"; |
564 |
break; |
565 |
case WSA_NOT_ENOUGH_MEMORY: |
566 |
error = "Insufficient memory available"; |
567 |
break; |
568 |
case WSANOTINITIALISED: |
569 |
error = "Successful WSAStartup not yet performed"; |
570 |
break; |
571 |
case WSANO_DATA: |
572 |
error = "Valid name, no data record of requested type"; |
573 |
break; |
574 |
case WSANO_RECOVERY: |
575 |
error = "This is a non-recoverable error"; |
576 |
break; |
577 |
// case WSAPROVIDERFAILEDINIT: |
578 |
// error = "Unable to initialize a service provider"; |
579 |
// break; |
580 |
case WSASYSCALLFAILURE: |
581 |
error = "System call failure"; |
582 |
break; |
583 |
case WSASYSNOTREADY: |
584 |
error = "Network subsystem is unavailable"; |
585 |
break; |
586 |
case WSATRY_AGAIN: |
587 |
error = "Non-authoritative host not found"; |
588 |
break; |
589 |
case WSAVERNOTSUPPORTED: |
590 |
error = "WINSOCK.DLL version out of range"; |
591 |
break; |
592 |
case WSAEDISCON: |
593 |
error = "Graceful shutdown in progress"; |
594 |
break; |
595 |
case WSA_OPERATION_ABORTED: |
596 |
error = "Overlapped operation aborted"; |
597 |
break; |
598 |
default: |
599 |
error = "Unknown error"; |
600 |
break; |
601 |
} |
602 |
|
603 |
cerr << prefix << ": " << error << "\n"; |
604 |
#else |
605 |
if (host) |
606 |
{ |
607 |
string error; |
608 |
|
609 |
switch (h_errno) |
610 |
{ |
611 |
case HOST_NOT_FOUND: |
612 |
error = "Unknown host"; |
613 |
break; |
614 |
case TRY_AGAIN: |
615 |
error = "Host name lookup failure"; |
616 |
break; |
617 |
case NO_RECOVERY: |
618 |
error = "Unknown server error"; |
619 |
break; |
620 |
case NO_DATA: |
621 |
error = "No address associated with name"; |
622 |
break; |
623 |
default: |
624 |
error = "Unknown error"; |
625 |
break; |
626 |
} |
627 |
|
628 |
cerr << prefix << ": " << error << "\n"; |
629 |
} |
630 |
else |
631 |
{ |
632 |
perror(prefix.c_str()); |
633 |
} |
634 |
#endif // _WIN32 |
635 |
} |