1 |
douglas |
1 |
/* ============================================================================ |
2 |
|
|
* Douglas Thrift's Search Engine License |
3 |
|
|
* |
4 |
|
|
* Copyright (C) 2002, 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 |
|
|
string fred = required[index0].substr(4); |
519 |
|
|
value += find(fred, lowerDescription, |
520 |
|
|
occurrencesDescription); |
521 |
|
|
} |
522 |
|
|
else if (required[index0].find("TITLE ") == 0) |
523 |
|
|
{ |
524 |
|
|
string fred = required[index0].substr(6); |
525 |
|
|
value += find(fred, lowerDescription, |
526 |
|
|
occurrencesDescription); |
527 |
|
|
} |
528 |
|
|
else if (required[index0].find("TEXT ") == 0) |
529 |
|
|
{ |
530 |
|
|
string fred = required[index0].substr(5); |
531 |
|
|
value += find(fred, lowerDescription, |
532 |
|
|
occurrencesDescription); |
533 |
|
|
} |
534 |
|
|
else |
535 |
|
|
{ |
536 |
|
|
value += find(required[index0], lowerDescription, |
537 |
|
|
occurrencesDescription); |
538 |
|
|
} |
539 |
|
|
} |
540 |
|
|
|
541 |
|
|
for (unsigned index1 = 0; index1 < eitherOr.size(); index1++) |
542 |
|
|
{ |
543 |
|
|
vector<string> words; |
544 |
|
|
|
545 |
|
|
unsigned begin = 0, found; |
546 |
|
|
do |
547 |
|
|
{ |
548 |
|
|
found = eitherOr[index1].find(" OR ", begin); |
549 |
|
|
|
550 |
|
|
if (found != string::npos) |
551 |
|
|
{ |
552 |
|
|
words.push_back(eitherOr[index1].substr(begin, found - |
553 |
|
|
begin)); |
554 |
|
|
} |
555 |
|
|
else |
556 |
|
|
{ |
557 |
|
|
words.push_back(eitherOr[index1].substr(begin)); |
558 |
|
|
} |
559 |
|
|
|
560 |
|
|
begin = found + 4; |
561 |
|
|
} |
562 |
|
|
while (begin < eitherOr[index1].length() && found != |
563 |
|
|
string::npos); |
564 |
|
|
|
565 |
|
|
for (unsigned number = 0; number < words.size(); number++) |
566 |
|
|
{ |
567 |
|
|
if (words[index1].find("URL ") == 0) |
568 |
|
|
{ |
569 |
|
|
string fred = words[index1].substr(4); |
570 |
|
|
value += find(fred, lowerDescription, |
571 |
|
|
occurrencesDescription); |
572 |
|
|
} |
573 |
|
|
else if (words[index1].find("TITLE ") == 0) |
574 |
|
|
{ |
575 |
|
|
string fred = words[index1].substr(6); |
576 |
|
|
value += find(fred, lowerDescription, |
577 |
|
|
occurrencesDescription); |
578 |
|
|
} |
579 |
|
|
else if (words[index1].find("TEXT ") == 0) |
580 |
|
|
{ |
581 |
|
|
string fred = words[index1].substr(5); |
582 |
|
|
value += find(fred, lowerDescription, |
583 |
|
|
occurrencesDescription); |
584 |
|
|
} |
585 |
|
|
else |
586 |
|
|
{ |
587 |
|
|
value += find(words[index1], lowerDescription, |
588 |
|
|
occurrencesDescription); |
589 |
|
|
} |
590 |
|
|
} |
591 |
|
|
} |
592 |
|
|
|
593 |
|
|
for (unsigned index2 = 0; index2 < getHeadings().size(); index2++) |
594 |
|
|
{ |
595 |
|
|
string lowerHeading = string(getHeadings()[index2].length(), |
596 |
|
|
' '); |
597 |
|
|
for (unsigned number = 0; number < |
598 |
|
|
getHeadings()[index2].length(); number++) |
599 |
|
|
{ |
600 |
|
|
lowerHeading[number] = tolower( |
601 |
|
|
getHeadings()[index2][number]); |
602 |
|
|
} |
603 |
|
|
|
604 |
|
|
for (unsigned number0 = 0; number0 < required.size(); number0++) |
605 |
|
|
{ |
606 |
|
|
if (required[number0].find("URL ") == 0) |
607 |
|
|
{ |
608 |
|
|
string fred = required[number0].substr(4); |
609 |
|
|
value += find(fred, |
610 |
|
|
lowerHeading); |
611 |
|
|
} |
612 |
|
|
else if (required[number0].find("TITLE ") == 0) |
613 |
|
|
{ |
614 |
|
|
string fred = required[number0].substr(6); |
615 |
|
|
value += find(fred, |
616 |
|
|
lowerHeading); |
617 |
|
|
} |
618 |
|
|
else if (required[number0].find("TEXT ") == 0) |
619 |
|
|
{ |
620 |
|
|
string fred = required[number0].substr(5); |
621 |
|
|
value += find(fred, |
622 |
|
|
lowerHeading); |
623 |
|
|
} |
624 |
|
|
else |
625 |
|
|
{ |
626 |
|
|
value += find(required[number0], lowerHeading); |
627 |
|
|
} |
628 |
|
|
} |
629 |
|
|
|
630 |
|
|
for (unsigned number1 = 0; number1 < eitherOr.size(); number1++) |
631 |
|
|
{ |
632 |
|
|
vector<string> words; |
633 |
|
|
|
634 |
|
|
unsigned begin = 0, found; |
635 |
|
|
do |
636 |
|
|
{ |
637 |
|
|
found = eitherOr[number1].find(" OR ", begin); |
638 |
|
|
|
639 |
|
|
if (found != string::npos) |
640 |
|
|
{ |
641 |
|
|
words.push_back(eitherOr[number1].substr(begin, |
642 |
|
|
found - begin)); |
643 |
|
|
} |
644 |
|
|
else |
645 |
|
|
{ |
646 |
|
|
words.push_back(eitherOr[number1].substr(begin)); |
647 |
|
|
} |
648 |
|
|
|
649 |
|
|
begin = found + 4; |
650 |
|
|
} |
651 |
|
|
while (begin < eitherOr[number1].length() && found != |
652 |
|
|
string::npos); |
653 |
|
|
|
654 |
|
|
for (unsigned number = 0; number < words.size(); number++) |
655 |
|
|
{ |
656 |
|
|
if (words[number].find("URL ") == 0) |
657 |
|
|
{ |
658 |
|
|
string fred = words[number].substr(4); |
659 |
|
|
value += find(fred, |
660 |
|
|
lowerHeading); |
661 |
|
|
} |
662 |
|
|
else if (words[number].find("TITLE ") == 0) |
663 |
|
|
{ |
664 |
|
|
string fred = words[number].substr(6); |
665 |
|
|
value += find(fred, |
666 |
|
|
lowerHeading); |
667 |
|
|
} |
668 |
|
|
else if (words[number].find("TEXT ") == 0) |
669 |
|
|
{ |
670 |
|
|
string fred = words[number].substr(5); |
671 |
|
|
value += find(fred, |
672 |
|
|
lowerHeading); |
673 |
|
|
} |
674 |
|
|
else |
675 |
|
|
{ |
676 |
|
|
value += find(words[number], lowerHeading); |
677 |
|
|
} |
678 |
|
|
} |
679 |
|
|
} |
680 |
|
|
} |
681 |
|
|
} |
682 |
|
|
} |
683 |
|
|
} |
684 |
|
|
|
685 |
|
|
void Ranker::checkRequired() |
686 |
|
|
{ |
687 |
|
|
vector<unsigned> inURLs, inTitles, inTexts; |
688 |
|
|
|
689 |
|
|
for (unsigned index = 0; index < required.size(); index++) |
690 |
|
|
{ |
691 |
|
|
unsigned inURL = 0, inTitle = 0, inText = 0; |
692 |
|
|
|
693 |
|
|
if (required[index].find("URL ") == 0) |
694 |
|
|
{ |
695 |
|
|
string fred = required[index].substr(4); |
696 |
|
|
string martha = lowerURL.substr(7); |
697 |
|
|
inURL = find(fred, martha); |
698 |
|
|
|
699 |
|
|
if (inURL) |
700 |
|
|
{ |
701 |
|
|
string fred = required[index].substr(4); |
702 |
|
|
inTitle = find(fred, lowerTitle, |
703 |
|
|
occurrencesTitle); |
704 |
|
|
string martha = required[index].substr(4); |
705 |
|
|
inText = find(martha, lowerText, |
706 |
|
|
occurrencesText); |
707 |
|
|
|
708 |
|
|
if (!inTitle) inTitle++; |
709 |
|
|
if (!inText) inText++; |
710 |
|
|
} |
711 |
|
|
} |
712 |
|
|
else if (required[index].find("TITLE ") == 0) |
713 |
|
|
{ |
714 |
|
|
string fred = required[index].substr(6); |
715 |
|
|
inTitle = find(fred, lowerTitle, |
716 |
|
|
occurrencesTitle); |
717 |
|
|
|
718 |
|
|
if (inTitle) |
719 |
|
|
{ |
720 |
|
|
string fred = required[index].substr(6); |
721 |
|
|
string martha = lowerURL.substr(7); |
722 |
|
|
inURL = find(fred, martha); |
723 |
|
|
string george = required[index].substr(6); |
724 |
|
|
inText = find(george, lowerText, |
725 |
|
|
occurrencesText); |
726 |
|
|
|
727 |
|
|
if (!inURL) inURL++; |
728 |
|
|
if (!inText) inText++; |
729 |
|
|
} |
730 |
|
|
} |
731 |
|
|
else if (required[index].find("TEXT ") == 0) |
732 |
|
|
{ |
733 |
|
|
string fred = required[index].substr(5); |
734 |
|
|
inText = find(fred, lowerText, |
735 |
|
|
occurrencesText); |
736 |
|
|
|
737 |
|
|
if (inText) |
738 |
|
|
{ |
739 |
|
|
string fred = required[index].substr(5); |
740 |
|
|
string martha = lowerURL.substr(7); |
741 |
|
|
inURL = find(fred, martha); |
742 |
|
|
string george = required[index].substr(5); |
743 |
|
|
inTitle = find(george, lowerTitle, |
744 |
|
|
occurrencesTitle); |
745 |
|
|
|
746 |
|
|
if (!inURL) inURL++; |
747 |
|
|
if (!inTitle) inTitle++; |
748 |
|
|
} |
749 |
|
|
} |
750 |
|
|
else |
751 |
|
|
{ |
752 |
|
|
string fred = lowerURL.substr(7); |
753 |
|
|
inURL = find(required[index], fred); |
754 |
|
|
inTitle = find(required[index], lowerTitle, occurrencesTitle); |
755 |
|
|
inText = find(required[index], lowerText, occurrencesText); |
756 |
|
|
} |
757 |
|
|
|
758 |
|
|
inURLs.push_back(inURL); |
759 |
|
|
inTitles.push_back(inTitle); |
760 |
|
|
inTexts.push_back(inText); |
761 |
|
|
} |
762 |
|
|
|
763 |
|
|
unsigned inURL = evaluate(inURLs); |
764 |
|
|
unsigned inTitle = evaluate(inTitles); |
765 |
|
|
unsigned inText = evaluate(inTexts); |
766 |
|
|
|
767 |
|
|
requiredValue += (inURL && (allIn == url)) || (inTitle && (allIn == title)) |
768 |
|
|
|| (inText && ((allIn == text) || (allIn == all))) ? inURL + inTitle + |
769 |
|
|
inText : 0; |
770 |
|
|
} |
771 |
|
|
|
772 |
|
|
void Ranker::checkExcluded() |
773 |
|
|
{ |
774 |
|
|
vector<unsigned> inURLs, inTitles, inTexts; |
775 |
|
|
|
776 |
|
|
for (unsigned index = 0; index < excluded.size(); index++) |
777 |
|
|
{ |
778 |
|
|
unsigned inURL = 0, inTitle = 0, inText = 0; |
779 |
|
|
|
780 |
|
|
string fred = lowerURL.substr(7); |
781 |
|
|
inURL = find(excluded[index], fred); |
782 |
|
|
inTitle = find(excluded[index], lowerTitle); |
783 |
|
|
inText = find(excluded[index], lowerText); |
784 |
|
|
|
785 |
|
|
inURLs.push_back(inURL); |
786 |
|
|
inTitles.push_back(inTitle); |
787 |
|
|
inTexts.push_back(inText); |
788 |
|
|
} |
789 |
|
|
|
790 |
|
|
unsigned inURL = evaluate(inURLs); |
791 |
|
|
unsigned inTitle = evaluate(inTitles); |
792 |
|
|
unsigned inText = evaluate(inTexts); |
793 |
|
|
|
794 |
|
|
excludedValue += (inURL && (allIn == url)) || (inTitle && (allIn == title)) |
795 |
|
|
|| (inText && ((allIn == text) || (allIn == all))) ? inURL + inTitle + |
796 |
|
|
inText : 0; |
797 |
|
|
} |
798 |
|
|
|
799 |
|
|
void Ranker::checkEitherOr() |
800 |
|
|
{ |
801 |
|
|
vector<unsigned> inURLs, inTitles, inTexts; |
802 |
|
|
|
803 |
|
|
for (unsigned index = 0; index < eitherOr.size(); index++) |
804 |
|
|
{ |
805 |
|
|
vector<unsigned> inURLz, inTitlez, inTextz; |
806 |
|
|
unsigned inURL = 0, inTitle = 0, inText = 0; |
807 |
|
|
vector<string> words; |
808 |
|
|
|
809 |
|
|
unsigned begin = 0, found; |
810 |
|
|
do |
811 |
|
|
{ |
812 |
|
|
found = eitherOr[index].find(" OR ", begin); |
813 |
|
|
|
814 |
|
|
if (found != string::npos) |
815 |
|
|
{ |
816 |
|
|
words.push_back(eitherOr[index].substr(begin, found - begin)); |
817 |
|
|
} |
818 |
|
|
else |
819 |
|
|
{ |
820 |
|
|
words.push_back(eitherOr[index].substr(begin)); |
821 |
|
|
} |
822 |
|
|
|
823 |
|
|
begin = found + 4; |
824 |
|
|
} |
825 |
|
|
while (begin < eitherOr[index].length() && found != string::npos); |
826 |
|
|
|
827 |
|
|
for (unsigned number = 0; number < words.size(); number++) |
828 |
|
|
{ |
829 |
|
|
unsigned inURL = 0, inTitle = 0, inText = 0; |
830 |
|
|
|
831 |
|
|
if (words[number].find("URL ") == 0) |
832 |
|
|
{ |
833 |
|
|
string fred = words[number].substr(4); |
834 |
|
|
string martha = lowerURL.substr(7); |
835 |
|
|
inURL = find(fred, martha); |
836 |
|
|
|
837 |
|
|
if (inURL) |
838 |
|
|
{ |
839 |
|
|
string fred = words[number].substr(4); |
840 |
|
|
inTitle = find(fred, lowerTitle, |
841 |
|
|
occurrencesTitle); |
842 |
|
|
string martha = words[number].substr(4); |
843 |
|
|
inText = find(martha, lowerText, |
844 |
|
|
occurrencesText); |
845 |
|
|
|
846 |
|
|
if (!inTitle) inTitle++; |
847 |
|
|
if (!inText) inText++; |
848 |
|
|
} |
849 |
|
|
} |
850 |
|
|
else if (words[number].find("TITLE ") == 0) |
851 |
|
|
{ |
852 |
|
|
string fred = words[number].substr(6); |
853 |
|
|
inTitle = find(fred, lowerTitle, |
854 |
|
|
occurrencesTitle); |
855 |
|
|
|
856 |
|
|
if (inTitle) |
857 |
|
|
{ |
858 |
|
|
string fred = words[number].substr(6); |
859 |
|
|
string martha = lowerURL.substr(7); |
860 |
|
|
inURL = find(fred, martha); |
861 |
|
|
string george = words[number].substr(6); |
862 |
|
|
inText = find(george, lowerText, |
863 |
|
|
occurrencesText); |
864 |
|
|
|
865 |
|
|
if (!inURL) inURL++; |
866 |
|
|
if (!inText) inText++; |
867 |
|
|
} |
868 |
|
|
} |
869 |
|
|
else if (words[number].find("TEXT ") == 0) |
870 |
|
|
{ |
871 |
|
|
string fred = words[number].substr(5); |
872 |
|
|
inText = find(fred, lowerText, |
873 |
|
|
occurrencesText); |
874 |
|
|
|
875 |
|
|
if (inText) |
876 |
|
|
{ |
877 |
|
|
string fred = words[number].substr(5); |
878 |
|
|
string martha = lowerURL.substr(7); |
879 |
|
|
inURL = find(fred, martha); |
880 |
|
|
string george = words[number].substr(5); |
881 |
|
|
inTitle = find(george, lowerTitle, |
882 |
|
|
occurrencesTitle); |
883 |
|
|
|
884 |
|
|
if (!inURL) inURL++; |
885 |
|
|
if (!inTitle) inTitle++; |
886 |
|
|
} |
887 |
|
|
} |
888 |
|
|
else |
889 |
|
|
{ |
890 |
|
|
string fred = lowerURL.substr(7); |
891 |
|
|
inURL = find(words[number], fred); |
892 |
|
|
inTitle = find(words[number], lowerTitle, occurrencesTitle); |
893 |
|
|
inText = find(words[number], lowerText, occurrencesText); |
894 |
|
|
} |
895 |
|
|
|
896 |
|
|
inURLz.push_back(inURL); |
897 |
|
|
inTitlez.push_back(inTitle); |
898 |
|
|
inTextz.push_back(inText); |
899 |
|
|
} |
900 |
|
|
|
901 |
|
|
for (unsigned number0 = 0; number0 < inURLz.size(); number0++) |
902 |
|
|
{ |
903 |
|
|
inURL += inURLz[number0]; |
904 |
|
|
} |
905 |
|
|
|
906 |
|
|
for (unsigned number1 = 0; number1 < inTitlez.size(); number1++) |
907 |
|
|
{ |
908 |
|
|
inTitle += inTitlez[number1]; |
909 |
|
|
} |
910 |
|
|
|
911 |
|
|
for (unsigned number2 = 0; number2 < inTextz.size(); number2++) |
912 |
|
|
{ |
913 |
|
|
inText += inTextz[number2]; |
914 |
|
|
} |
915 |
|
|
|
916 |
|
|
inURLs.push_back(inURL); |
917 |
|
|
inTitles.push_back(inTitle); |
918 |
|
|
inTexts.push_back(inText); |
919 |
|
|
|
920 |
|
|
inURLz.clear(); |
921 |
|
|
inTitlez.clear(); |
922 |
|
|
inTextz.clear(); |
923 |
|
|
words.clear(); |
924 |
|
|
} |
925 |
|
|
|
926 |
|
|
unsigned inURL = evaluate(inURLs); |
927 |
|
|
unsigned inTitle = evaluate(inTitles); |
928 |
|
|
unsigned inText = evaluate(inTexts); |
929 |
|
|
|
930 |
|
|
eitherOrValue += (inURL && (allIn == url)) || (inTitle && (allIn == title)) |
931 |
|
|
|| (inText && ((allIn == text) || (allIn == all))) ? inURL + inTitle + |
932 |
|
|
inText : 0; |
933 |
|
|
} |
934 |
|
|
|
935 |
|
|
unsigned Ranker::find(string& word, string& where) |
936 |
|
|
{ |
937 |
|
|
unsigned value = 0; |
938 |
|
|
|
939 |
|
|
decrap(word); |
940 |
|
|
|
941 |
|
|
if (word == "") |
942 |
|
|
{ |
943 |
|
|
// this can happen if a word is all crap characters |
944 |
|
|
value++; |
945 |
|
|
} |
946 |
|
|
else if (word.find_first_of(" \n ") == string::npos) |
947 |
|
|
{ |
948 |
|
|
unsigned begin = 0, found; |
949 |
|
|
do |
950 |
|
|
{ |
951 |
|
|
found = where.find(word, begin); |
952 |
|
|
|
953 |
|
|
if (found != string::npos) |
954 |
|
|
{ |
955 |
|
|
bool isBefore, isAfter, before = false, after = false; |
956 |
|
|
isBefore = found - 1 > 0; |
957 |
|
|
isAfter = found + word.length() < where.length(); |
958 |
|
|
|
959 |
|
|
if (isBefore) before = isalnum(where[found - 1]) != 0; |
960 |
|
|
if (isAfter) after = isalnum(where[found + word.length()]) != 0; |
961 |
|
|
|
962 |
|
|
if (!before && !after) |
963 |
|
|
{ |
964 |
|
|
value++; |
965 |
|
|
} |
966 |
|
|
} |
967 |
|
|
|
968 |
|
|
begin = found + word.length(); |
969 |
|
|
} |
970 |
|
|
while (found != string::npos && begin < where.length()); |
971 |
|
|
} |
972 |
|
|
else |
973 |
|
|
{ |
974 |
|
|
value = phrase(word, where); |
975 |
|
|
} |
976 |
|
|
|
977 |
|
|
return value; |
978 |
|
|
} |
979 |
|
|
|
980 |
|
|
unsigned Ranker::find(string& word, string& where, map<unsigned, unsigned>& |
981 |
|
|
occurrences) |
982 |
|
|
{ |
983 |
|
|
unsigned value = 0; |
984 |
|
|
|
985 |
|
|
decrap(word); |
986 |
|
|
|
987 |
|
|
if (word == "") |
988 |
|
|
{ |
989 |
|
|
// this can happen if a word is all crap characters |
990 |
|
|
value++; |
991 |
|
|
} |
992 |
|
|
else if (word.find_first_of(" \n ") == string::npos) |
993 |
|
|
{ |
994 |
|
|
unsigned begin = 0, found; |
995 |
|
|
do |
996 |
|
|
{ |
997 |
|
|
found = where.find(word, begin); |
998 |
|
|
|
999 |
|
|
if (found != string::npos) |
1000 |
|
|
{ |
1001 |
|
|
bool isBefore, isAfter, before = false, after = false; |
1002 |
|
|
isBefore = found - 1 > 0; |
1003 |
|
|
isAfter = found + word.length() < where.length(); |
1004 |
|
|
|
1005 |
|
|
if (isBefore) before = isalnum(where[found - 1]) != 0; |
1006 |
|
|
if (isAfter) after = isalnum(where[found + word.length()]) != 0; |
1007 |
|
|
|
1008 |
|
|
if (!before && !after) |
1009 |
|
|
{ |
1010 |
|
|
value++; |
1011 |
|
|
|
1012 |
|
|
occurrences.insert(pair<unsigned, unsigned>(found, |
1013 |
|
|
word.length())); |
1014 |
|
|
} |
1015 |
|
|
} |
1016 |
|
|
|
1017 |
|
|
begin = found + word.length(); |
1018 |
|
|
} |
1019 |
|
|
while (found != string::npos && begin < where.length()); |
1020 |
|
|
} |
1021 |
|
|
else |
1022 |
|
|
{ |
1023 |
|
|
value = phrase(word, where, occurrences); |
1024 |
|
|
} |
1025 |
|
|
|
1026 |
|
|
return value; |
1027 |
|
|
} |
1028 |
|
|
|
1029 |
|
|
unsigned Ranker::phrase(string& phrase, string& where) |
1030 |
|
|
{ |
1031 |
|
|
unsigned value = 0; |
1032 |
|
|
vector<string> words; |
1033 |
|
|
|
1034 |
|
|
unsigned begin = 0, space; |
1035 |
|
|
do |
1036 |
|
|
{ |
1037 |
|
|
space = phrase.find(' ', begin); |
1038 |
|
|
|
1039 |
|
|
words.push_back(phrase.substr(begin, space - begin)); |
1040 |
|
|
|
1041 |
|
|
begin = space + 1; |
1042 |
|
|
} |
1043 |
|
|
while (space != string::npos && begin < phrase.length()); |
1044 |
|
|
|
1045 |
|
|
begin = 0; |
1046 |
|
|
unsigned counter = 0; |
1047 |
|
|
do |
1048 |
|
|
{ |
1049 |
|
|
value += this->phrase(words, 0, begin, true, where); |
1050 |
|
|
} |
1051 |
|
|
while (begin < where.length()); |
1052 |
|
|
|
1053 |
|
|
return value; |
1054 |
|
|
} |
1055 |
|
|
|
1056 |
|
|
unsigned Ranker::phrase(string& phrase, string& where, map<unsigned, unsigned>& |
1057 |
|
|
occurrences) |
1058 |
|
|
{ |
1059 |
|
|
unsigned value = 0; |
1060 |
|
|
vector<string> words; |
1061 |
|
|
|
1062 |
|
|
unsigned begin = 0, space; |
1063 |
|
|
do |
1064 |
|
|
{ |
1065 |
|
|
space = phrase.find(' ', begin); |
1066 |
|
|
|
1067 |
|
|
words.push_back(phrase.substr(begin, space - begin)); |
1068 |
|
|
|
1069 |
|
|
begin = space + 1; |
1070 |
|
|
} |
1071 |
|
|
while (space != string::npos && begin < phrase.length()); |
1072 |
|
|
|
1073 |
|
|
begin = 0; |
1074 |
|
|
do |
1075 |
|
|
{ |
1076 |
|
|
value += this->phrase(words, 0, begin, true, where, occurrences); |
1077 |
|
|
} |
1078 |
|
|
while (begin < where.length()); |
1079 |
|
|
|
1080 |
|
|
return value; |
1081 |
|
|
} |
1082 |
|
|
|
1083 |
|
|
unsigned Ranker::phrase(vector<string>& words, unsigned word, unsigned& begin, |
1084 |
|
|
bool start, string& where) |
1085 |
|
|
{ |
1086 |
|
|
unsigned value = 0; |
1087 |
|
|
bool end = !(word + 1 < words.size()); |
1088 |
|
|
unsigned found = where.find(words[word], begin); |
1089 |
|
|
unsigned newBegin = found + words[word].length(); |
1090 |
|
|
|
1091 |
|
|
if (found != string::npos) |
1092 |
|
|
{ |
1093 |
|
|
bool isBefore, isAfter, before = false, after = false; |
1094 |
|
|
isBefore = found - 1 > 0; |
1095 |
|
|
isAfter = found + words[word].length() < where.length(); |
1096 |
|
|
|
1097 |
|
|
if (isBefore) before = isalnum(where[found - 1]) != 0; |
1098 |
|
|
if (isAfter) after = isalnum(where[found + words[word].length()]) != 0; |
1099 |
|
|
|
1100 |
|
|
if (!before && !after) |
1101 |
|
|
{ |
1102 |
|
|
bool between = true; |
1103 |
|
|
if (!start) |
1104 |
|
|
{ |
1105 |
|
|
for (unsigned index = begin + 1; index < found - 1; index++) |
1106 |
|
|
{ |
1107 |
|
|
if (isalnum(where[index])) |
1108 |
|
|
{ |
1109 |
|
|
between = false; |
1110 |
|
|
break; |
1111 |
|
|
} |
1112 |
|
|
} |
1113 |
|
|
} |
1114 |
|
|
|
1115 |
|
|
if (between) |
1116 |
|
|
{ |
1117 |
|
|
if (end) |
1118 |
|
|
{ |
1119 |
|
|
begin = newBegin; |
1120 |
|
|
value = 1; |
1121 |
|
|
} |
1122 |
|
|
else |
1123 |
|
|
{ |
1124 |
|
|
value = phrase(words, (word + 1), newBegin, false, where); |
1125 |
|
|
} |
1126 |
|
|
} |
1127 |
|
|
} |
1128 |
|
|
} |
1129 |
|
|
|
1130 |
|
|
if (start) |
1131 |
|
|
{ |
1132 |
|
|
if (found != string::npos) |
1133 |
|
|
{ |
1134 |
|
|
begin = newBegin; |
1135 |
|
|
} |
1136 |
|
|
else |
1137 |
|
|
{ |
1138 |
|
|
begin = string::npos; |
1139 |
|
|
} |
1140 |
|
|
} |
1141 |
|
|
|
1142 |
|
|
return value; |
1143 |
|
|
} |
1144 |
|
|
|
1145 |
|
|
unsigned Ranker::phrase(vector<string>& words, unsigned word, unsigned& begin, |
1146 |
|
|
bool start, string& where, map<unsigned, unsigned>& occurrences) |
1147 |
|
|
{ |
1148 |
|
|
unsigned value = 0; |
1149 |
|
|
bool end = !(word + 1 < words.size()); |
1150 |
|
|
unsigned found = where.find(words[word], begin); |
1151 |
|
|
unsigned newBegin = found + words[word].length(); |
1152 |
|
|
|
1153 |
|
|
if (found != string::npos) |
1154 |
|
|
{ |
1155 |
|
|
bool isBefore, isAfter, before = false, after = false; |
1156 |
|
|
isBefore = found - 1 > 0; |
1157 |
|
|
isAfter = found + words[word].length() < where.length(); |
1158 |
|
|
|
1159 |
|
|
if (isBefore) before = isalnum(where[found - 1]) != 0; |
1160 |
|
|
if (isAfter) after = isalnum(where[found + words[word].length()]) != 0; |
1161 |
|
|
|
1162 |
|
|
if (!before && !after) |
1163 |
|
|
{ |
1164 |
|
|
bool between = true; |
1165 |
|
|
if (!start) |
1166 |
|
|
{ |
1167 |
|
|
for (unsigned index = begin + 1; index < found - 1; index++) |
1168 |
|
|
{ |
1169 |
|
|
if (isalnum(where[index])) |
1170 |
|
|
{ |
1171 |
|
|
between = false; |
1172 |
|
|
break; |
1173 |
|
|
} |
1174 |
|
|
} |
1175 |
|
|
} |
1176 |
|
|
|
1177 |
|
|
if (between) |
1178 |
|
|
{ |
1179 |
|
|
occurrences.insert(pair<unsigned, unsigned>(found, |
1180 |
|
|
words[word].length())); |
1181 |
|
|
|
1182 |
|
|
if (end) |
1183 |
|
|
{ |
1184 |
|
|
begin = newBegin; |
1185 |
|
|
value = 1; |
1186 |
|
|
} |
1187 |
|
|
else |
1188 |
|
|
{ |
1189 |
|
|
value = phrase(words, (word + 1), newBegin, false, where, |
1190 |
|
|
occurrences); |
1191 |
|
|
} |
1192 |
|
|
} |
1193 |
|
|
} |
1194 |
|
|
} |
1195 |
|
|
|
1196 |
|
|
if (start) |
1197 |
|
|
{ |
1198 |
|
|
if (found != string::npos) |
1199 |
|
|
{ |
1200 |
|
|
begin = newBegin; |
1201 |
|
|
} |
1202 |
|
|
else |
1203 |
|
|
{ |
1204 |
|
|
begin = string::npos; |
1205 |
|
|
} |
1206 |
|
|
} |
1207 |
|
|
|
1208 |
|
|
return value; |
1209 |
|
|
} |
1210 |
|
|
|
1211 |
|
|
unsigned Ranker::evaluate(vector<unsigned>& ins) |
1212 |
|
|
{ |
1213 |
|
|
unsigned in = 0; |
1214 |
|
|
|
1215 |
|
|
for (unsigned index = 0; index < ins.size(); index++) |
1216 |
|
|
{ |
1217 |
|
|
if (ins[index] > 0) |
1218 |
|
|
{ |
1219 |
|
|
in += ins[index]; |
1220 |
|
|
} |
1221 |
|
|
else |
1222 |
|
|
{ |
1223 |
|
|
in = 0; |
1224 |
|
|
break; |
1225 |
|
|
} |
1226 |
|
|
} |
1227 |
|
|
|
1228 |
|
|
return in; |
1229 |
|
|
} |
1230 |
|
|
|
1231 |
|
|
void Ranker::decrap(string& crap) |
1232 |
|
|
{ |
1233 |
|
|
unsigned begin = 0, found; |
1234 |
|
|
do |
1235 |
|
|
{ |
1236 |
|
|
// &, +, and # are not considered crap |
1237 |
|
|
found = crap.find_first_of("!\"$%\'()*,-./:;<=>?@[\\]^_`{|}~", begin); |
1238 |
|
|
|
1239 |
|
|
if (found != string::npos) |
1240 |
|
|
{ |
1241 |
|
|
crap[found] = ' '; |
1242 |
|
|
} |
1243 |
|
|
|
1244 |
|
|
begin = found + 1; |
1245 |
|
|
} |
1246 |
|
|
while (found != string::npos && begin < crap.length()); |
1247 |
|
|
|
1248 |
|
|
normalize(crap); |
1249 |
|
|
} |