| 1 |
|
/* ============================================================================ |
| 2 |
|
* Douglas Thrift's Search Engine License |
| 3 |
|
* |
| 4 |
< |
* Copyright (C) 2002, Douglas Thrift. All Rights Reserved. |
| 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 |
|
* |
| 65 |
|
string title, description, text; |
| 66 |
|
vector<string> headings; |
| 67 |
|
|
| 68 |
< |
if (html(http)) |
| 68 |
> |
if (http.contentType().find("text/html") == 0) |
| 69 |
|
{ |
| 70 |
|
if (!process(http, url, title, description, text, headings)) return |
| 71 |
|
false; |
| 107 |
|
} |
| 108 |
|
else |
| 109 |
|
{ |
| 110 |
– |
bool knowSize = page->getSize() > 0; |
| 111 |
– |
|
| 110 |
|
string line; |
| 111 |
|
while (http.good()) |
| 112 |
|
{ |
| 113 |
|
http.getline(line); |
| 114 |
|
|
| 115 |
|
text += line + "\n"; |
| 118 |
– |
|
| 119 |
– |
if (!knowSize) page->setSize(page->getSize() + line.length() + 1); |
| 116 |
|
} |
| 117 |
|
|
| 118 |
|
normalize(text); |
| 119 |
|
} |
| 120 |
|
|
| 121 |
+ |
page->setSize(http.contentLength()); |
| 122 |
|
page->setURL(url); |
| 123 |
|
page->setTitle(title); |
| 124 |
|
page->setDescription(description); |
| 139 |
|
description, string& text, vector<string>& headings) |
| 140 |
|
{ |
| 141 |
|
bool inHtml = false, inHead = false, inTitle = false, inBody = false, |
| 142 |
< |
inHeading = false, inComment = false, knowSize = page->getSize() > 0, |
| 146 |
< |
follow = true, answer = true; |
| 142 |
> |
inHeading = false, inComment = false, follow = true, answer = true; |
| 143 |
|
unsigned startComment = 0, finishComment = 0; |
| 144 |
|
string line; |
| 145 |
|
while (http.good()) |
| 154 |
|
unsigned close = line.find('>', begin); |
| 155 |
|
|
| 156 |
|
string next; |
| 157 |
< |
while (close == string::npos) |
| 157 |
> |
while (close == string::npos && http.good()) |
| 158 |
|
{ |
| 159 |
|
http.getline(next); |
| 160 |
|
line += '\n' + next; |
| 239 |
|
unsigned start = lowerTag.find("href=\"") + 6; |
| 240 |
|
unsigned finish = lowerTag.find('\"', start); |
| 241 |
|
|
| 242 |
< |
string fred = tag.substr(start, finish - start); |
| 243 |
< |
string link = getLink(fred, url); |
| 242 |
> |
string link = getLink(tag.substr(start, finish - |
| 243 |
> |
start), url); |
| 244 |
|
|
| 245 |
< |
if (link != "bad link") links.insert(link); |
| 245 |
> |
if (link != "") links.insert(link); |
| 246 |
|
} |
| 247 |
|
else if (lowerTag.find("href=") != string::npos) |
| 248 |
|
{ |
| 251 |
|
|
| 252 |
|
if (finish < close) |
| 253 |
|
{ |
| 254 |
< |
string fred = tag.substr(start, finish - start); |
| 255 |
< |
string link = getLink(fred, url); |
| 254 |
> |
string link = getLink(tag.substr(start, finish - |
| 255 |
> |
start), url); |
| 256 |
|
|
| 257 |
< |
if (link != "bad link") links.insert(link); |
| 257 |
> |
if (link != "") links.insert(link); |
| 258 |
|
} |
| 259 |
|
else |
| 260 |
|
{ |
| 261 |
< |
string fred = tag.substr(start, close - start); |
| 262 |
< |
string link = getLink(fred, url); |
| 261 |
> |
string link = getLink(tag.substr(start, close - |
| 262 |
> |
start), url); |
| 263 |
|
|
| 264 |
< |
if (link != "bad link") links.insert(link); |
| 264 |
> |
if (link != "") links.insert(link); |
| 265 |
|
} |
| 266 |
|
} |
| 267 |
|
} |
| 349 |
|
|
| 350 |
|
startComment = 0; |
| 351 |
|
finishComment = 0; |
| 356 |
– |
|
| 357 |
– |
if (!knowSize) page->setSize(page->getSize() + line.length() + 1); |
| 358 |
– |
} |
| 359 |
– |
|
| 360 |
– |
return answer; |
| 361 |
– |
} |
| 362 |
– |
|
| 363 |
– |
bool Processor::html(HttpHandler& http) |
| 364 |
– |
{ |
| 365 |
– |
bool answer = false; |
| 366 |
– |
|
| 367 |
– |
string line; |
| 368 |
– |
http.getline(line); |
| 369 |
– |
|
| 370 |
– |
while (http.good()) |
| 371 |
– |
{ |
| 372 |
– |
string field; |
| 373 |
– |
http.getline(field, ' '); |
| 374 |
– |
if (field == "") break; |
| 375 |
– |
http.getline(line); |
| 376 |
– |
|
| 377 |
– |
if (field == "Content-Type:" || field == "Content-type:") |
| 378 |
– |
{ |
| 379 |
– |
if (line.find("text/html") != string::npos) |
| 380 |
– |
{ |
| 381 |
– |
answer = true; |
| 382 |
– |
} |
| 383 |
– |
} |
| 384 |
– |
|
| 385 |
– |
if (field == "Content-Length:" || field == "Content-length:") |
| 386 |
– |
{ |
| 387 |
– |
page->setSize(strtoul(line.c_str(), 0, 0)); |
| 388 |
– |
} |
| 352 |
|
} |
| 353 |
|
|
| 354 |
|
return answer; |
| 355 |
|
} |
| 356 |
|
|
| 357 |
< |
string Processor::getTag(string& line, unsigned open, unsigned close) |
| 357 |
> |
string Processor::getTag(const string& line, unsigned open, unsigned close) |
| 358 |
|
{ |
| 359 |
|
string tag = line.substr(open + 1, close - open - 1); |
| 360 |
|
|
| 361 |
|
return tag; |
| 362 |
|
} |
| 400 |
– |
|
| 401 |
– |
string Processor::getLink(string& link, URL& url) |
| 402 |
– |
{ |
| 403 |
– |
string hyperlink = "bad link"; |
| 404 |
– |
|
| 405 |
– |
if (link.find('#') != string::npos) |
| 406 |
– |
{ |
| 407 |
– |
unsigned pound = link.find('#'); |
| 408 |
– |
link.erase(pound); |
| 409 |
– |
} |
| 410 |
– |
|
| 411 |
– |
if (link.find("://") != string::npos) |
| 412 |
– |
{ |
| 413 |
– |
if (link.find("http://") == 0) hyperlink = link; |
| 414 |
– |
} |
| 415 |
– |
else if (link.find("mailto:") == 0) |
| 416 |
– |
{ |
| 417 |
– |
// do nothing we are not evil spammers! |
| 418 |
– |
} |
| 419 |
– |
else if (link.find("//") == 0) |
| 420 |
– |
{ |
| 421 |
– |
hyperlink = "http:" + link; |
| 422 |
– |
} |
| 423 |
– |
else if (link.find('/') == 0) |
| 424 |
– |
{ |
| 425 |
– |
hyperlink = url.getURL(); |
| 426 |
– |
|
| 427 |
– |
unsigned path = hyperlink.find('/', 7); |
| 428 |
– |
hyperlink.erase(path); |
| 429 |
– |
|
| 430 |
– |
hyperlink += link; |
| 431 |
– |
} |
| 432 |
– |
else if (link == "") |
| 433 |
– |
{ |
| 434 |
– |
// a blank link is useless |
| 435 |
– |
} |
| 436 |
– |
else |
| 437 |
– |
{ |
| 438 |
– |
hyperlink = url.getURL(); |
| 439 |
– |
string path = url.getPath(); |
| 440 |
– |
|
| 441 |
– |
unsigned cutoff = hyperlink.rfind(path); |
| 442 |
– |
hyperlink.erase(cutoff); |
| 443 |
– |
|
| 444 |
– |
unsigned dir = path.rfind('/') + 1; |
| 445 |
– |
path.erase(dir); |
| 446 |
– |
|
| 447 |
– |
while (link.find("../") == 0) |
| 448 |
– |
{ |
| 449 |
– |
unsigned dot = path.rfind('/') - 1; |
| 450 |
– |
unsigned up = path.rfind('/', dot) + 1; |
| 451 |
– |
|
| 452 |
– |
path.erase(up); |
| 453 |
– |
link.erase(0, 3); |
| 454 |
– |
} |
| 455 |
– |
while (link.find("./") == 0) |
| 456 |
– |
{ |
| 457 |
– |
link.erase(0, 2); |
| 458 |
– |
} |
| 459 |
– |
|
| 460 |
– |
hyperlink += path + link; |
| 461 |
– |
} |
| 462 |
– |
|
| 463 |
– |
return hyperlink; |
| 464 |
– |
} |
