| 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 Ranker |
| 46 |
// |
| 47 |
// Douglas Thrift |
| 48 |
// |
| 49 |
// Ranker.cpp |
| 50 |
|
| 51 |
#include "Ranker.h" |
| 52 |
|
| 53 |
Ranker::Ranker() |
| 54 |
{ |
| 55 |
value = 0; |
| 56 |
requiredValue = 0; |
| 57 |
excludedValue = 0; |
| 58 |
eitherOrValue = 0; |
| 59 |
allIn = all; |
| 60 |
} |
| 61 |
|
| 62 |
Ranker::Ranker(Page& page) : Page(page) |
| 63 |
{ |
| 64 |
value = 0; |
| 65 |
requiredValue = 0; |
| 66 |
excludedValue = 0; |
| 67 |
eitherOrValue = 0; |
| 68 |
allIn = all; |
| 69 |
} |
| 70 |
|
| 71 |
void Ranker::rank(vector<string> query) |
| 72 |
{ |
| 73 |
vector<string> prep; |
| 74 |
|
| 75 |
for (unsigned index = 0; index < query.size(); index++) |
| 76 |
{ |
| 77 |
if (query[index] == "allintitle:" && index == 0) |
| 78 |
{ |
| 79 |
allIn = title; |
| 80 |
} |
| 81 |
else if (query[index] == "allinurl:" && index == 0) |
| 82 |
{ |
| 83 |
allIn = url; |
| 84 |
} |
| 85 |
else if (query[index] == "allintext:" && index == 0) |
| 86 |
{ |
| 87 |
allIn = text; |
| 88 |
} |
| 89 |
else if (query[index].find("site:") == 0 && query[index].size() > 5) |
| 90 |
{ |
| 91 |
site = query[index].substr(5); |
| 92 |
} |
| 93 |
else if (query[index].find("intitle:") == 0 && query[index].size() > 8) |
| 94 |
{ |
| 95 |
prep.push_back("TITLE " + query[index].substr(8)); |
| 96 |
} |
| 97 |
else if (query[index].find("inurl:") == 0 && query[index].size() > 6) |
| 98 |
{ |
| 99 |
prep.push_back("URL " + query[index].substr(6)); |
| 100 |
} |
| 101 |
else if (query[index].find("intext:") == 0 && query[index].size() > 7) |
| 102 |
{ |
| 103 |
prep.push_back("TEXT " + query[index].substr(7)); |
| 104 |
} |
| 105 |
else |
| 106 |
{ |
| 107 |
prep.push_back(query[index]); |
| 108 |
} |
| 109 |
} |
| 110 |
|
| 111 |
if (prep.size() > 0) |
| 112 |
{ |
| 113 |
bool or_ = false; |
| 114 |
for (unsigned index = 0; index < prep.size(); index++) |
| 115 |
{ |
| 116 |
bool exclude = false; |
| 117 |
if (prep[index].find('+') == 0) |
| 118 |
{ |
| 119 |
prep[index].erase(0, 1); |
| 120 |
} |
| 121 |
else if (prep[index].find('-') == 0) |
| 122 |
{ |
| 123 |
exclude = true; |
| 124 |
prep[index].erase(0, 1); |
| 125 |
} |
| 126 |
|
| 127 |
if (or_) |
| 128 |
{ |
| 129 |
if (prep[index].find(" OR") == string::npos) |
| 130 |
{ |
| 131 |
or_ = false; |
| 132 |
} |
| 133 |
|
| 134 |
eitherOr[eitherOr.size() - 1] += ' ' + prep[index]; |
| 135 |
} |
| 136 |
else if (exclude) |
| 137 |
{ |
| 138 |
excluded.push_back(prep[index]); |
| 139 |
} |
| 140 |
else if (prep[index].find(" OR") != string::npos) |
| 141 |
{ |
| 142 |
or_ = true; |
| 143 |
eitherOr.push_back(prep[index]); |
| 144 |
} |
| 145 |
else |
| 146 |
{ |
| 147 |
required.push_back(prep[index]); |
| 148 |
} |
| 149 |
} |
| 150 |
} |
| 151 |
|
| 152 |
rank(); |
| 153 |
} |
| 154 |
|
| 155 |
void Ranker::setSample() |
| 156 |
{ |
| 157 |
map<unsigned, unsigned>::iterator itor; |
| 158 |
|
| 159 |
multimap<unsigned, map<unsigned, unsigned>::iterator> distances; |
| 160 |
|
| 161 |
for (itor = occurrencesText.begin(); itor != occurrencesText.end(); itor++) |
| 162 |
{ |
| 163 |
unsigned distance; |
| 164 |
|
| 165 |
if (++itor != occurrencesText.end()) |
| 166 |
{ |
| 167 |
unsigned next = itor->first; |
| 168 |
itor--; |
| 169 |
|
| 170 |
distance = next - (itor->first + itor->second); |
| 171 |
} |
| 172 |
else |
| 173 |
{ |
| 174 |
distance = UINT_MAX; |
| 175 |
itor--; |
| 176 |
} |
| 177 |
|
| 178 |
distances.insert(pair<unsigned, map<unsigned, |
| 179 |
unsigned>::iterator>(distance, itor)); |
| 180 |
} |
| 181 |
|
| 182 |
if (distances.begin() != distances.end()) |
| 183 |
{ |
| 184 |
itor = distances.begin()->second; |
| 185 |
} |
| 186 |
|
| 187 |
string portion; |
| 188 |
unsigned sampleLength = 0, begin = 0, end = string::npos; |
| 189 |
while (sampleLength < 160 && itor != occurrencesText.end()) |
| 190 |
{ |
| 191 |
unsigned found = itor->first; |
| 192 |
unsigned length = itor->second; |
| 193 |
|
| 194 |
for (unsigned index = found; index > begin; index--) |
| 195 |
{ |
| 196 |
if (index == begin) cerr << "Oh crap, I'm insane!\n"; |
| 197 |
if (found - index >= 160 - sampleLength - length) |
| 198 |
{ |
| 199 |
for (; index < found; index++) |
| 200 |
{ |
| 201 |
if (isspace(getText()[index])) break; |
| 202 |
} |
| 203 |
begin = index + 1; |
| 204 |
break; |
| 205 |
} |
| 206 |
else if ((index > begin ? (isupper(getText()[index]) && |
| 207 |
!isalnum(getText()[index - 1])) : isupper(getText()[index])) && |
| 208 |
index != found) |
| 209 |
{ |
| 210 |
begin = index; |
| 211 |
break; |
| 212 |
} |
| 213 |
} |
| 214 |
|
| 215 |
if (end + 1 != begin) sample += " <strong>...</strong> "; |
| 216 |
|
| 217 |
portion = getText().substr(begin, found - begin); |
| 218 |
sampleLength += portion.length(); |
| 219 |
|
| 220 |
entities(portion, '&', "&"); |
| 221 |
entities(portion, '\"', """); |
| 222 |
entities(portion, '<', "<"); |
| 223 |
entities(portion, '>', ">"); |
| 224 |
|
| 225 |
sample += portion + "<strong>"; |
| 226 |
|
| 227 |
portion = getText().substr(found, length); |
| 228 |
sampleLength += portion.length(); |
| 229 |
|
| 230 |
entities(portion, '&', "&"); |
| 231 |
entities(portion, '\"', """); |
| 232 |
entities(portion, '<', "<"); |
| 233 |
entities(portion, '>', ">"); |
| 234 |
|
| 235 |
sample += portion + "</strong>"; |
| 236 |
|
| 237 |
begin = found + length; |
| 238 |
end = begin - 1; |
| 239 |
|
| 240 |
if (++itor != occurrencesText.end()) |
| 241 |
{ |
| 242 |
if (itor->first + itor->second < begin + 160 - sampleLength) |
| 243 |
{ |
| 244 |
portion = getText().substr(begin, itor->first - begin); |
| 245 |
sampleLength += portion.length(); |
| 246 |
|
| 247 |
entities(portion, '&', "&"); |
| 248 |
entities(portion, '\"', """); |
| 249 |
entities(portion, '<', "<"); |
| 250 |
entities(portion, '>', ">"); |
| 251 |
|
| 252 |
sample += portion; |
| 253 |
|
| 254 |
begin = itor->first; |
| 255 |
end = begin - 1; |
| 256 |
} |
| 257 |
else |
| 258 |
{ |
| 259 |
for (end = begin + 160 - sampleLength; end > begin; end--) |
| 260 |
{ |
| 261 |
if (isspace(getText()[end])) break; |
| 262 |
} |
| 263 |
|
| 264 |
portion = getText().substr(begin, end - begin + 1); |
| 265 |
sampleLength += portion.length(); |
| 266 |
|
| 267 |
entities(portion, '&', "&"); |
| 268 |
entities(portion, '\"', """); |
| 269 |
entities(portion, '<', "<"); |
| 270 |
entities(portion, '>', ">"); |
| 271 |
|
| 272 |
sample += portion + " <strong>...</strong>"; |
| 273 |
|
| 274 |
break; |
| 275 |
} |
| 276 |
} |
| 277 |
else |
| 278 |
{ |
| 279 |
for (end = begin + 160 - sampleLength; end > begin && (end + 1 < |
| 280 |
getText().length()); end--) |
| 281 |
{ |
| 282 |
if (isspace(getText()[end])) break; |
| 283 |
} |
| 284 |
|
| 285 |
if (end >= getText().length()) end = getText().length() - 1; |
| 286 |
|
| 287 |
portion = getText().substr(begin, end - begin + 1); |
| 288 |
sampleLength += portion.length(); |
| 289 |
|
| 290 |
entities(portion, '&', "&"); |
| 291 |
entities(portion, '\"', """); |
| 292 |
entities(portion, '<', "<"); |
| 293 |
entities(portion, '>', ">"); |
| 294 |
|
| 295 |
sample += portion; |
| 296 |
|
| 297 |
if (end + 1 < getText().length()) |
| 298 |
{ |
| 299 |
sample += " <strong>...</strong>"; |
| 300 |
} |
| 301 |
|
| 302 |
break; |
| 303 |
} |
| 304 |
} |
| 305 |
} |
| 306 |
|
| 307 |
string Ranker::getTitle() |
| 308 |
{ |
| 309 |
string title, portion; |
| 310 |
|
| 311 |
unsigned begin = 0; |
| 312 |
for (map<unsigned, unsigned>::iterator itor = occurrencesTitle.begin(); |
| 313 |
itor != occurrencesTitle.end(); itor++) |
| 314 |
{ |
| 315 |
unsigned found = itor->first; |
| 316 |
unsigned length = itor->second; |
| 317 |
|
| 318 |
portion = Page::getTitle().substr(begin, found - begin); |
| 319 |
|
| 320 |
entities(portion, '&', "&"); |
| 321 |
entities(portion, '\"', """); |
| 322 |
entities(portion, '<', "<"); |
| 323 |
entities(portion, '>', ">"); |
| 324 |
|
| 325 |
title += portion + "<strong>"; |
| 326 |
|
| 327 |
portion = Page::getTitle().substr(found, length); |
| 328 |
|
| 329 |
entities(portion, '&', "&"); |
| 330 |
entities(portion, '\"', """); |
| 331 |
entities(portion, '<', "<"); |
| 332 |
entities(portion, '>', ">"); |
| 333 |
|
| 334 |
title += portion + "</strong>"; |
| 335 |
|
| 336 |
begin = found + length; |
| 337 |
} |
| 338 |
|
| 339 |
portion = Page::getTitle().substr(begin); |
| 340 |
|
| 341 |
entities(portion, '&', "&"); |
| 342 |
entities(portion, '\"', """); |
| 343 |
entities(portion, '<', "<"); |
| 344 |
entities(portion, '>', ">"); |
| 345 |
|
| 346 |
title += portion; |
| 347 |
|
| 348 |
return title; |
| 349 |
} |
| 350 |
|
| 351 |
string Ranker::getDescription() |
| 352 |
{ |
| 353 |
string description, portion; |
| 354 |
|
| 355 |
unsigned begin = 0; |
| 356 |
for (map<unsigned, unsigned>::iterator itor = |
| 357 |
occurrencesDescription.begin(); itor != occurrencesDescription.end(); |
| 358 |
itor++) |
| 359 |
{ |
| 360 |
unsigned found = itor->first; |
| 361 |
unsigned length = itor->second; |
| 362 |
|
| 363 |
portion = Page::getDescription().substr(begin, found - begin); |
| 364 |
|
| 365 |
entities(portion, '&', "&"); |
| 366 |
entities(portion, '\"', """); |
| 367 |
entities(portion, '<', "<"); |
| 368 |
entities(portion, '>', ">"); |
| 369 |
|
| 370 |
description += portion + "<strong>"; |
| 371 |
|
| 372 |
portion = Page::getDescription().substr(found, length); |
| 373 |
|
| 374 |
entities(portion, '&', "&"); |
| 375 |
entities(portion, '\"', """); |
| 376 |
entities(portion, '<', "<"); |
| 377 |
entities(portion, '>', ">"); |
| 378 |
|
| 379 |
description += portion + "</strong>"; |
| 380 |
|
| 381 |
begin = found + length; |
| 382 |
} |
| 383 |
|
| 384 |
portion = Page::getDescription().substr(begin); |
| 385 |
|
| 386 |
entities(portion, '&', "&"); |
| 387 |
entities(portion, '\"', """); |
| 388 |
entities(portion, '<', "<"); |
| 389 |
entities(portion, '>', ">"); |
| 390 |
|
| 391 |
description += portion; |
| 392 |
|
| 393 |
return description; |
| 394 |
} |
| 395 |
|
| 396 |
bool Ranker::operator==(const unsigned number) const |
| 397 |
{ |
| 398 |
return value == number; |
| 399 |
} |
| 400 |
|
| 401 |
bool Ranker::operator==(const Ranker& ranker) const |
| 402 |
{ |
| 403 |
return value == ranker.value; |
| 404 |
} |
| 405 |
|
| 406 |
bool Ranker::operator!=(const unsigned number) const |
| 407 |
{ |
| 408 |
return value != number; |
| 409 |
} |
| 410 |
|
| 411 |
bool Ranker::operator!=(const Ranker& ranker) const |
| 412 |
{ |
| 413 |
return value != ranker.value; |
| 414 |
} |
| 415 |
|
| 416 |
bool Ranker::operator<(const unsigned number) const |
| 417 |
{ |
| 418 |
return value < number; |
| 419 |
} |
| 420 |
|
| 421 |
bool Ranker::operator<(const Ranker& ranker) const |
| 422 |
{ |
| 423 |
return value < ranker.value; |
| 424 |
} |
| 425 |
|
| 426 |
bool Ranker::operator>(const unsigned number) const |
| 427 |
{ |
| 428 |
return value > number; |
| 429 |
} |
| 430 |
|
| 431 |
bool Ranker::operator >(const Ranker& ranker) const |
| 432 |
{ |
| 433 |
return value > ranker.value; |
| 434 |
} |
| 435 |
|
| 436 |
void Ranker::rank() |
| 437 |
{ |
| 438 |
lowerAddress = string(getAddress().length(), ' '); |
| 439 |
for (unsigned index = 0; index < lowerAddress.length(); index++) |
| 440 |
{ |
| 441 |
lowerAddress[index] = tolower(getAddress()[index]); |
| 442 |
} |
| 443 |
|
| 444 |
if (site == "" || lowerAddress.rfind(site) == lowerAddress.length() - |
| 445 |
site.length()) |
| 446 |
{ |
| 447 |
bool isRequired = required.size() > 0; |
| 448 |
bool isExcluded = excluded.size() > 0; |
| 449 |
bool isEitherOr = eitherOr.size() > 0; |
| 450 |
|
| 451 |
lowerURL = string(getURL().length(), ' '); |
| 452 |
for (unsigned index = 0; index < lowerURL.length(); index++) |
| 453 |
{ |
| 454 |
lowerURL[index] = tolower(getURL()[index]); |
| 455 |
} |
| 456 |
|
| 457 |
lowerTitle = string(Page::getTitle().length(), ' '); |
| 458 |
for (unsigned index0 = 0; index0 < lowerTitle.length(); index0++) |
| 459 |
{ |
| 460 |
lowerTitle[index0] = tolower(Page::getTitle()[index0]); |
| 461 |
} |
| 462 |
|
| 463 |
lowerText = string(Page::getText().length(), ' '); |
| 464 |
for (unsigned index1 = 0; index1 < lowerText.length(); index1++) |
| 465 |
{ |
| 466 |
lowerText[index1] = tolower(Page::getText()[index1]); |
| 467 |
} |
| 468 |
|
| 469 |
if (isRequired) checkRequired(); |
| 470 |
if (isExcluded && (isRequired || isEitherOr)) checkExcluded(); |
| 471 |
if (isEitherOr) checkEitherOr(); |
| 472 |
|
| 473 |
if (isRequired && isExcluded && isEitherOr) |
| 474 |
{ |
| 475 |
value += requiredValue && !excludedValue && eitherOrValue ? |
| 476 |
requiredValue + eitherOrValue : 0; |
| 477 |
} |
| 478 |
else if (isRequired && isExcluded) |
| 479 |
{ |
| 480 |
value += requiredValue && !excludedValue ? requiredValue : 0; |
| 481 |
} |
| 482 |
else if (isRequired && isEitherOr) |
| 483 |
{ |
| 484 |
value += requiredValue && eitherOrValue ? requiredValue + |
| 485 |
eitherOrValue : 0; |
| 486 |
} |
| 487 |
else if (isExcluded && isEitherOr) |
| 488 |
{ |
| 489 |
value += !excludedValue && eitherOrValue ? eitherOrValue : 0; |
| 490 |
} |
| 491 |
else if (isRequired) |
| 492 |
{ |
| 493 |
value += requiredValue; |
| 494 |
} |
| 495 |
else if (isEitherOr) |
| 496 |
{ |
| 497 |
value += eitherOrValue; |
| 498 |
} |
| 499 |
else |
| 500 |
{ |
| 501 |
// do nothing this is a bad search and warrants no results |
| 502 |
} |
| 503 |
|
| 504 |
if (value > 0) |
| 505 |
{ |
| 506 |
string lowerDescription = string(Page::getDescription().length(), |
| 507 |
' '); |
| 508 |
for (unsigned index = 0; index < lowerDescription.length(); index++) |
| 509 |
{ |
| 510 |
lowerDescription[index] = tolower( |
| 511 |
Page::getDescription()[index]); |
| 512 |
} |
| 513 |
|
| 514 |
for (unsigned index0 = 0; index0 < required.size(); index0++) |
| 515 |
{ |
| 516 |
if (required[index0].find("URL ") == 0) |
| 517 |
{ |
| 518 |
value += find(required[index0].substr(4), lowerDescription, |
| 519 |
occurrencesDescription); |
| 520 |
} |
| 521 |
else if (required[index0].find("TITLE ") == 0) |
| 522 |
{ |
| 523 |
value += find(required[index0].substr(6), lowerDescription, |
| 524 |
occurrencesDescription); |
| 525 |
} |
| 526 |
else if (required[index0].find("TEXT ") == 0) |
| 527 |
{ |
| 528 |
value += find(required[index0].substr(5), lowerDescription, |
| 529 |
occurrencesDescription); |
| 530 |
} |
| 531 |
else |
| 532 |
{ |
| 533 |
value += find(required[index0], lowerDescription, |
| 534 |
occurrencesDescription); |
| 535 |
} |
| 536 |
} |
| 537 |
|
| 538 |
for (unsigned index1 = 0; index1 < eitherOr.size(); index1++) |
| 539 |
{ |
| 540 |
vector<string> words; |
| 541 |
|
| 542 |
unsigned begin = 0, found; |
| 543 |
do |
| 544 |
{ |
| 545 |
found = eitherOr[index1].find(" OR ", begin); |
| 546 |
|
| 547 |
if (found != string::npos) |
| 548 |
{ |
| 549 |
words.push_back(eitherOr[index1].substr(begin, found - |
| 550 |
begin)); |
| 551 |
} |
| 552 |
else |
| 553 |
{ |
| 554 |
words.push_back(eitherOr[index1].substr(begin)); |
| 555 |
} |
| 556 |
|
| 557 |
begin = found + 4; |
| 558 |
} |
| 559 |
while (begin < eitherOr[index1].length() && found != |
| 560 |
string::npos); |
| 561 |
|
| 562 |
for (unsigned number = 0; number < words.size(); number++) |
| 563 |
{ |
| 564 |
if (words[index1].find("URL ") == 0) |
| 565 |
{ |
| 566 |
value += find(words[index1].substr(4), |
| 567 |
lowerDescription, occurrencesDescription); |
| 568 |
} |
| 569 |
else if (words[index1].find("TITLE ") == 0) |
| 570 |
{ |
| 571 |
value += find(words[index1].substr(6), |
| 572 |
lowerDescription, occurrencesDescription); |
| 573 |
} |
| 574 |
else if (words[index1].find("TEXT ") == 0) |
| 575 |
{ |
| 576 |
value += find(words[index1].substr(5), |
| 577 |
lowerDescription, occurrencesDescription); |
| 578 |
} |
| 579 |
else |
| 580 |
{ |
| 581 |
value += find(words[index1], lowerDescription, |
| 582 |
occurrencesDescription); |
| 583 |
} |
| 584 |
} |
| 585 |
} |
| 586 |
|
| 587 |
for (unsigned index2 = 0; index2 < getHeadings().size(); index2++) |
| 588 |
{ |
| 589 |
string lowerHeading = string(getHeadings()[index2].length(), |
| 590 |
' '); |
| 591 |
for (unsigned number = 0; number < |
| 592 |
getHeadings()[index2].length(); number++) |
| 593 |
{ |
| 594 |
lowerHeading[number] = tolower( |
| 595 |
getHeadings()[index2][number]); |
| 596 |
} |
| 597 |
|
| 598 |
for (unsigned number0 = 0; number0 < required.size(); number0++) |
| 599 |
{ |
| 600 |
if (required[number0].find("URL ") == 0) |
| 601 |
{ |
| 602 |
value += find(required[number0].substr(4), |
| 603 |
lowerHeading); |
| 604 |
} |
| 605 |
else if (required[number0].find("TITLE ") == 0) |
| 606 |
{ |
| 607 |
value += find(required[number0].substr(6), |
| 608 |
lowerHeading); |
| 609 |
} |
| 610 |
else if (required[number0].find("TEXT ") == 0) |
| 611 |
{ |
| 612 |
value += find(required[number0].substr(5), |
| 613 |
lowerHeading); |
| 614 |
} |
| 615 |
else |
| 616 |
{ |
| 617 |
value += find(required[number0], lowerHeading); |
| 618 |
} |
| 619 |
} |
| 620 |
|
| 621 |
for (unsigned number1 = 0; number1 < eitherOr.size(); number1++) |
| 622 |
{ |
| 623 |
vector<string> words; |
| 624 |
|
| 625 |
unsigned begin = 0, found; |
| 626 |
do |
| 627 |
{ |
| 628 |
found = eitherOr[number1].find(" OR ", begin); |
| 629 |
|
| 630 |
if (found != string::npos) |
| 631 |
{ |
| 632 |
words.push_back(eitherOr[number1].substr(begin, |
| 633 |
found - begin)); |
| 634 |
} |
| 635 |
else |
| 636 |
{ |
| 637 |
words.push_back(eitherOr[number1].substr(begin)); |
| 638 |
} |
| 639 |
|
| 640 |
begin = found + 4; |
| 641 |
} |
| 642 |
while (begin < eitherOr[number1].length() && found != |
| 643 |
string::npos); |
| 644 |
|
| 645 |
for (unsigned number = 0; number < words.size(); number++) |
| 646 |
{ |
| 647 |
if (words[number].find("URL ") == 0) |
| 648 |
{ |
| 649 |
value += find(words[number].substr(4), |
| 650 |
lowerHeading); |
| 651 |
} |
| 652 |
else if (words[number].find("TITLE ") == 0) |
| 653 |
{ |
| 654 |
value += find(words[number].substr(6), |
| 655 |
lowerHeading); |
| 656 |
} |
| 657 |
else if (words[number].find("TEXT ") == 0) |
| 658 |
{ |
| 659 |
value += find(words[number].substr(5), |
| 660 |
lowerHeading); |
| 661 |
} |
| 662 |
else |
| 663 |
{ |
| 664 |
value += find(words[number], lowerHeading); |
| 665 |
} |
| 666 |
} |
| 667 |
} |
| 668 |
} |
| 669 |
} |
| 670 |
} |
| 671 |
} |
| 672 |
|
| 673 |
void Ranker::checkRequired() |
| 674 |
{ |
| 675 |
vector<unsigned> inURLs, inTitles, inTexts; |
| 676 |
|
| 677 |
for (unsigned index = 0; index < required.size(); index++) |
| 678 |
{ |
| 679 |
unsigned inURL = 0, inTitle = 0, inText = 0; |
| 680 |
|
| 681 |
if (required[index].find("URL ") == 0) |
| 682 |
{ |
| 683 |
inURL = find(required[index].substr(4), lowerURL.substr(7)); |
| 684 |
|
| 685 |
if (inURL) |
| 686 |
{ |
| 687 |
inTitle = find(required[index].substr(4), lowerTitle, |
| 688 |
occurrencesTitle); |
| 689 |
inText = find(required[index].substr(4), lowerText, |
| 690 |
occurrencesText); |
| 691 |
|
| 692 |
if (!inTitle) inTitle++; |
| 693 |
if (!inText) inText++; |
| 694 |
} |
| 695 |
} |
| 696 |
else if (required[index].find("TITLE ") == 0) |
| 697 |
{ |
| 698 |
inTitle = find(required[index].substr(6), lowerTitle, |
| 699 |
occurrencesTitle); |
| 700 |
|
| 701 |
if (inTitle) |
| 702 |
{ |
| 703 |
inURL = find(required[index].substr(6), lowerURL.substr(7)); |
| 704 |
inText = find(required[index].substr(6), lowerText, |
| 705 |
occurrencesText); |
| 706 |
|
| 707 |
if (!inURL) inURL++; |
| 708 |
if (!inText) inText++; |
| 709 |
} |
| 710 |
} |
| 711 |
else if (required[index].find("TEXT ") == 0) |
| 712 |
{ |
| 713 |
inText = find(required[index].substr(5), lowerText, |
| 714 |
occurrencesText); |
| 715 |
|
| 716 |
if (inText) |
| 717 |
{ |
| 718 |
inURL = find(required[index].substr(5), lowerURL.substr(7)); |
| 719 |
inTitle = find(required[index].substr(5), lowerTitle, |
| 720 |
occurrencesTitle); |
| 721 |
|
| 722 |
if (!inURL) inURL++; |
| 723 |
if (!inTitle) inTitle++; |
| 724 |
} |
| 725 |
} |
| 726 |
else |
| 727 |
{ |
| 728 |
inURL = find(required[index], lowerURL.substr(7)); |
| 729 |
inTitle = find(required[index], lowerTitle, occurrencesTitle); |
| 730 |
inText = find(required[index], lowerText, occurrencesText); |
| 731 |
} |
| 732 |
|
| 733 |
inURLs.push_back(inURL); |
| 734 |
inTitles.push_back(inTitle); |
| 735 |
inTexts.push_back(inText); |
| 736 |
} |
| 737 |
|
| 738 |
unsigned inURL = evaluate(inURLs); |
| 739 |
unsigned inTitle = evaluate(inTitles); |
| 740 |
unsigned inText = evaluate(inTexts); |
| 741 |
|
| 742 |
requiredValue += (inURL && (allIn == url)) || (inTitle && (allIn == title)) |
| 743 |
|| (inText && ((allIn == text) || (allIn == all))) ? inURL + inTitle + |
| 744 |
inText : 0; |
| 745 |
} |
| 746 |
|
| 747 |
void Ranker::checkExcluded() |
| 748 |
{ |
| 749 |
vector<unsigned> inURLs, inTitles, inTexts; |
| 750 |
|
| 751 |
for (unsigned index = 0; index < excluded.size(); index++) |
| 752 |
{ |
| 753 |
unsigned inURL = 0, inTitle = 0, inText = 0; |
| 754 |
|
| 755 |
inURL = find(excluded[index], lowerURL.substr(7)); |
| 756 |
inTitle = find(excluded[index], lowerTitle); |
| 757 |
inText = find(excluded[index], lowerText); |
| 758 |
|
| 759 |
inURLs.push_back(inURL); |
| 760 |
inTitles.push_back(inTitle); |
| 761 |
inTexts.push_back(inText); |
| 762 |
} |
| 763 |
|
| 764 |
unsigned inURL = evaluate(inURLs); |
| 765 |
unsigned inTitle = evaluate(inTitles); |
| 766 |
unsigned inText = evaluate(inTexts); |
| 767 |
|
| 768 |
excludedValue += (inURL && (allIn == url)) || (inTitle && (allIn == title)) |
| 769 |
|| (inText && ((allIn == text) || (allIn == all))) ? inURL + inTitle + |
| 770 |
inText : 0; |
| 771 |
} |
| 772 |
|
| 773 |
void Ranker::checkEitherOr() |
| 774 |
{ |
| 775 |
vector<unsigned> inURLs, inTitles, inTexts; |
| 776 |
|
| 777 |
for (unsigned index = 0; index < eitherOr.size(); index++) |
| 778 |
{ |
| 779 |
vector<unsigned> inURLz, inTitlez, inTextz; |
| 780 |
unsigned inURL = 0, inTitle = 0, inText = 0; |
| 781 |
vector<string> words; |
| 782 |
|
| 783 |
unsigned begin = 0, found; |
| 784 |
do |
| 785 |
{ |
| 786 |
found = eitherOr[index].find(" OR ", begin); |
| 787 |
|
| 788 |
if (found != string::npos) |
| 789 |
{ |
| 790 |
words.push_back(eitherOr[index].substr(begin, found - begin)); |
| 791 |
} |
| 792 |
else |
| 793 |
{ |
| 794 |
words.push_back(eitherOr[index].substr(begin)); |
| 795 |
} |
| 796 |
|
| 797 |
begin = found + 4; |
| 798 |
} |
| 799 |
while (begin < eitherOr[index].length() && found != string::npos); |
| 800 |
|
| 801 |
for (unsigned number = 0; number < words.size(); number++) |
| 802 |
{ |
| 803 |
unsigned inURL = 0, inTitle = 0, inText = 0; |
| 804 |
|
| 805 |
if (words[number].find("URL ") == 0) |
| 806 |
{ |
| 807 |
inURL = find(words[number].substr(4), lowerURL.substr(7)); |
| 808 |
|
| 809 |
if (inURL) |
| 810 |
{ |
| 811 |
inTitle = find(words[number].substr(4), lowerTitle, |
| 812 |
occurrencesTitle); |
| 813 |
inText = find(words[number].substr(4), lowerText, |
| 814 |
occurrencesText); |
| 815 |
|
| 816 |
if (!inTitle) inTitle++; |
| 817 |
if (!inText) inText++; |
| 818 |
} |
| 819 |
} |
| 820 |
else if (words[number].find("TITLE ") == 0) |
| 821 |
{ |
| 822 |
inTitle = find(words[number].substr(6), lowerTitle, |
| 823 |
occurrencesTitle); |
| 824 |
|
| 825 |
if (inTitle) |
| 826 |
{ |
| 827 |
inURL = find(words[number].substr(6), lowerURL.substr(7)); |
| 828 |
inText = find(words[number].substr(6), lowerText, |
| 829 |
occurrencesText); |
| 830 |
|
| 831 |
if (!inURL) inURL++; |
| 832 |
if (!inText) inText++; |
| 833 |
} |
| 834 |
} |
| 835 |
else if (words[number].find("TEXT ") == 0) |
| 836 |
{ |
| 837 |
inText = find(words[number].substr(5), lowerText, |
| 838 |
occurrencesText); |
| 839 |
|
| 840 |
if (inText) |
| 841 |
{ |
| 842 |
inURL = find(words[number].substr(5), lowerURL.substr(7)); |
| 843 |
inTitle = find(words[number].substr(5), lowerTitle, |
| 844 |
occurrencesTitle); |
| 845 |
|
| 846 |
if (!inURL) inURL++; |
| 847 |
if (!inTitle) inTitle++; |
| 848 |
} |
| 849 |
} |
| 850 |
else |
| 851 |
{ |
| 852 |
inURL = find(words[number], lowerURL.substr(7)); |
| 853 |
inTitle = find(words[number], lowerTitle, occurrencesTitle); |
| 854 |
inText = find(words[number], lowerText, occurrencesText); |
| 855 |
} |
| 856 |
|
| 857 |
inURLz.push_back(inURL); |
| 858 |
inTitlez.push_back(inTitle); |
| 859 |
inTextz.push_back(inText); |
| 860 |
} |
| 861 |
|
| 862 |
for (unsigned number0 = 0; number0 < inURLz.size(); number0++) |
| 863 |
{ |
| 864 |
inURL += inURLz[number0]; |
| 865 |
} |
| 866 |
|
| 867 |
for (unsigned number1 = 0; number1 < inTitlez.size(); number1++) |
| 868 |
{ |
| 869 |
inTitle += inTitlez[number1]; |
| 870 |
} |
| 871 |
|
| 872 |
for (unsigned number2 = 0; number2 < inTextz.size(); number2++) |
| 873 |
{ |
| 874 |
inText += inTextz[number2]; |
| 875 |
} |
| 876 |
|
| 877 |
inURLs.push_back(inURL); |
| 878 |
inTitles.push_back(inTitle); |
| 879 |
inTexts.push_back(inText); |
| 880 |
|
| 881 |
inURLz.clear(); |
| 882 |
inTitlez.clear(); |
| 883 |
inTextz.clear(); |
| 884 |
words.clear(); |
| 885 |
} |
| 886 |
|
| 887 |
unsigned inURL = evaluate(inURLs); |
| 888 |
unsigned inTitle = evaluate(inTitles); |
| 889 |
unsigned inText = evaluate(inTexts); |
| 890 |
|
| 891 |
eitherOrValue += (inURL && (allIn == url)) || (inTitle && (allIn == title)) |
| 892 |
|| (inText && ((allIn == text) || (allIn == all))) ? inURL + inTitle + |
| 893 |
inText : 0; |
| 894 |
} |
| 895 |
|
| 896 |
unsigned Ranker::find(string word, const string& where) |
| 897 |
{ |
| 898 |
unsigned value = 0; |
| 899 |
|
| 900 |
decrap(word); |
| 901 |
|
| 902 |
if (word == "") |
| 903 |
{ |
| 904 |
// this can happen if a word is all crap characters |
| 905 |
value++; |
| 906 |
} |
| 907 |
else if (word.find_first_of(" \n ") == string::npos) |
| 908 |
{ |
| 909 |
unsigned begin = 0, found; |
| 910 |
do |
| 911 |
{ |
| 912 |
found = where.find(word, begin); |
| 913 |
|
| 914 |
if (found != string::npos) |
| 915 |
{ |
| 916 |
bool isBefore, isAfter, before = false, after = false; |
| 917 |
isBefore = found - 1 > 0; |
| 918 |
isAfter = found + word.length() < where.length(); |
| 919 |
|
| 920 |
if (isBefore) before = isalnum(where[found - 1]) != 0; |
| 921 |
if (isAfter) after = isalnum(where[found + word.length()]) != 0; |
| 922 |
|
| 923 |
if (!before && !after) |
| 924 |
{ |
| 925 |
value++; |
| 926 |
} |
| 927 |
} |
| 928 |
|
| 929 |
begin = found + word.length(); |
| 930 |
} |
| 931 |
while (found != string::npos && begin < where.length()); |
| 932 |
} |
| 933 |
else |
| 934 |
{ |
| 935 |
value = phrase(word, where); |
| 936 |
} |
| 937 |
|
| 938 |
return value; |
| 939 |
} |
| 940 |
|
| 941 |
unsigned Ranker::find(string word, const string& where, map<unsigned, |
| 942 |
unsigned>& occurrences) |
| 943 |
{ |
| 944 |
unsigned value = 0; |
| 945 |
|
| 946 |
decrap(word); |
| 947 |
|
| 948 |
if (word == "") |
| 949 |
{ |
| 950 |
// this can happen if a word is all crap characters |
| 951 |
value++; |
| 952 |
} |
| 953 |
else if (word.find_first_of(" \n ") == string::npos) |
| 954 |
{ |
| 955 |
unsigned begin = 0, found; |
| 956 |
do |
| 957 |
{ |
| 958 |
found = where.find(word, begin); |
| 959 |
|
| 960 |
if (found != string::npos) |
| 961 |
{ |
| 962 |
bool isBefore, isAfter, before = false, after = false; |
| 963 |
isBefore = found - 1 > 0; |
| 964 |
isAfter = found + word.length() < where.length(); |
| 965 |
|
| 966 |
if (isBefore) before = isalnum(where[found - 1]) != 0; |
| 967 |
if (isAfter) after = isalnum(where[found + word.length()]) != 0; |
| 968 |
|
| 969 |
if (!before && !after) |
| 970 |
{ |
| 971 |
value++; |
| 972 |
|
| 973 |
occurrences.insert(pair<unsigned, unsigned>(found, |
| 974 |
word.length())); |
| 975 |
} |
| 976 |
} |
| 977 |
|
| 978 |
begin = found + word.length(); |
| 979 |
} |
| 980 |
while (found != string::npos && begin < where.length()); |
| 981 |
} |
| 982 |
else |
| 983 |
{ |
| 984 |
value = phrase(word, where, occurrences); |
| 985 |
} |
| 986 |
|
| 987 |
return value; |
| 988 |
} |
| 989 |
|
| 990 |
unsigned Ranker::phrase(const string& phrase, const string& where) |
| 991 |
{ |
| 992 |
unsigned value = 0; |
| 993 |
vector<string> words; |
| 994 |
|
| 995 |
unsigned begin = 0, space; |
| 996 |
do |
| 997 |
{ |
| 998 |
space = phrase.find(' ', begin); |
| 999 |
|
| 1000 |
words.push_back(phrase.substr(begin, space - begin)); |
| 1001 |
|
| 1002 |
begin = space + 1; |
| 1003 |
} |
| 1004 |
while (space != string::npos && begin < phrase.length()); |
| 1005 |
|
| 1006 |
begin = 0; |
| 1007 |
unsigned counter = 0; |
| 1008 |
do |
| 1009 |
{ |
| 1010 |
value += this->phrase(words, 0, begin, true, where); |
| 1011 |
} |
| 1012 |
while (begin < where.length()); |
| 1013 |
|
| 1014 |
return value; |
| 1015 |
} |
| 1016 |
|
| 1017 |
unsigned Ranker::phrase(const string& phrase, const string& where, |
| 1018 |
map<unsigned, unsigned>& occurrences) |
| 1019 |
{ |
| 1020 |
unsigned value = 0; |
| 1021 |
vector<string> words; |
| 1022 |
|
| 1023 |
unsigned begin = 0, space; |
| 1024 |
do |
| 1025 |
{ |
| 1026 |
space = phrase.find(' ', begin); |
| 1027 |
|
| 1028 |
words.push_back(phrase.substr(begin, space - begin)); |
| 1029 |
|
| 1030 |
begin = space + 1; |
| 1031 |
} |
| 1032 |
while (space != string::npos && begin < phrase.length()); |
| 1033 |
|
| 1034 |
begin = 0; |
| 1035 |
do |
| 1036 |
{ |
| 1037 |
value += this->phrase(words, 0, begin, true, where, occurrences); |
| 1038 |
} |
| 1039 |
while (begin < where.length()); |
| 1040 |
|
| 1041 |
return value; |
| 1042 |
} |
| 1043 |
|
| 1044 |
unsigned Ranker::phrase(const vector<string>& words, unsigned word, unsigned& |
| 1045 |
begin, bool start, const string& where) |
| 1046 |
{ |
| 1047 |
unsigned value = 0; |
| 1048 |
bool end = !(word + 1 < words.size()); |
| 1049 |
unsigned found = where.find(words[word], begin); |
| 1050 |
unsigned newBegin = found + words[word].length(); |
| 1051 |
|
| 1052 |
if (found != string::npos) |
| 1053 |
{ |
| 1054 |
bool isBefore, isAfter, before = false, after = false; |
| 1055 |
isBefore = found - 1 > 0; |
| 1056 |
isAfter = found + words[word].length() < where.length(); |
| 1057 |
|
| 1058 |
if (isBefore) before = isalnum(where[found - 1]) != 0; |
| 1059 |
if (isAfter) after = isalnum(where[found + words[word].length()]) != 0; |
| 1060 |
|
| 1061 |
if (!before && !after) |
| 1062 |
{ |
| 1063 |
bool between = true; |
| 1064 |
if (!start) |
| 1065 |
{ |
| 1066 |
for (unsigned index = begin + 1; index < found - 1; index++) |
| 1067 |
{ |
| 1068 |
if (isalnum(where[index])) |
| 1069 |
{ |
| 1070 |
between = false; |
| 1071 |
break; |
| 1072 |
} |
| 1073 |
} |
| 1074 |
} |
| 1075 |
|
| 1076 |
if (between) |
| 1077 |
{ |
| 1078 |
if (end) |
| 1079 |
{ |
| 1080 |
begin = newBegin; |
| 1081 |
value = 1; |
| 1082 |
} |
| 1083 |
else |
| 1084 |
{ |
| 1085 |
value = phrase(words, (word + 1), newBegin, false, where); |
| 1086 |
} |
| 1087 |
} |
| 1088 |
} |
| 1089 |
} |
| 1090 |
|
| 1091 |
if (start) |
| 1092 |
{ |
| 1093 |
if (found != string::npos) |
| 1094 |
{ |
| 1095 |
begin = newBegin; |
| 1096 |
} |
| 1097 |
else |
| 1098 |
{ |
| 1099 |
begin = string::npos; |
| 1100 |
} |
| 1101 |
} |
| 1102 |
|
| 1103 |
return value; |
| 1104 |
} |
| 1105 |
|
| 1106 |
unsigned Ranker::phrase(const vector<string>& words, unsigned word, unsigned& |
| 1107 |
begin, bool start, const string& where, map<unsigned, unsigned>& |
| 1108 |
occurrences) |
| 1109 |
{ |
| 1110 |
unsigned value = 0; |
| 1111 |
bool end = !(word + 1 < words.size()); |
| 1112 |
unsigned found = where.find(words[word], begin); |
| 1113 |
unsigned newBegin = found + words[word].length(); |
| 1114 |
|
| 1115 |
if (found != string::npos) |
| 1116 |
{ |
| 1117 |
bool isBefore, isAfter, before = false, after = false; |
| 1118 |
isBefore = found - 1 > 0; |
| 1119 |
isAfter = found + words[word].length() < where.length(); |
| 1120 |
|
| 1121 |
if (isBefore) before = isalnum(where[found - 1]) != 0; |
| 1122 |
if (isAfter) after = isalnum(where[found + words[word].length()]) != 0; |
| 1123 |
|
| 1124 |
if (!before && !after) |
| 1125 |
{ |
| 1126 |
bool between = true; |
| 1127 |
if (!start) |
| 1128 |
{ |
| 1129 |
for (unsigned index = begin + 1; index < found - 1; index++) |
| 1130 |
{ |
| 1131 |
if (isalnum(where[index])) |
| 1132 |
{ |
| 1133 |
between = false; |
| 1134 |
break; |
| 1135 |
} |
| 1136 |
} |
| 1137 |
} |
| 1138 |
|
| 1139 |
if (between) |
| 1140 |
{ |
| 1141 |
occurrences.insert(pair<unsigned, unsigned>(found, |
| 1142 |
words[word].length())); |
| 1143 |
|
| 1144 |
if (end) |
| 1145 |
{ |
| 1146 |
begin = newBegin; |
| 1147 |
value = 1; |
| 1148 |
} |
| 1149 |
else |
| 1150 |
{ |
| 1151 |
value = phrase(words, (word + 1), newBegin, false, where, |
| 1152 |
occurrences); |
| 1153 |
} |
| 1154 |
} |
| 1155 |
} |
| 1156 |
} |
| 1157 |
|
| 1158 |
if (start) |
| 1159 |
{ |
| 1160 |
if (found != string::npos) |
| 1161 |
{ |
| 1162 |
begin = newBegin; |
| 1163 |
} |
| 1164 |
else |
| 1165 |
{ |
| 1166 |
begin = string::npos; |
| 1167 |
} |
| 1168 |
} |
| 1169 |
|
| 1170 |
return value; |
| 1171 |
} |
| 1172 |
|
| 1173 |
unsigned Ranker::evaluate(vector<unsigned>& ins) |
| 1174 |
{ |
| 1175 |
unsigned in = 0; |
| 1176 |
|
| 1177 |
for (unsigned index = 0; index < ins.size(); index++) |
| 1178 |
{ |
| 1179 |
if (ins[index] > 0) |
| 1180 |
{ |
| 1181 |
in += ins[index]; |
| 1182 |
} |
| 1183 |
else |
| 1184 |
{ |
| 1185 |
in = 0; |
| 1186 |
break; |
| 1187 |
} |
| 1188 |
} |
| 1189 |
|
| 1190 |
return in; |
| 1191 |
} |
| 1192 |
|
| 1193 |
void Ranker::decrap(string& crap) |
| 1194 |
{ |
| 1195 |
unsigned begin = 0, found; |
| 1196 |
do |
| 1197 |
{ |
| 1198 |
// &, _, +, and # are not considered crap |
| 1199 |
found = crap.find_first_of("!\"$%\'()*,-./:;<=>?@[\\]^`{|}~", begin); |
| 1200 |
|
| 1201 |
if (found != string::npos) |
| 1202 |
{ |
| 1203 |
crap[found] = ' '; |
| 1204 |
} |
| 1205 |
|
| 1206 |
begin = found + 1; |
| 1207 |
} |
| 1208 |
while (found != string::npos && begin < crap.length()); |
| 1209 |
|
| 1210 |
normalize(crap); |
| 1211 |
} |
