| 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 Indexer |
| 46 |
// |
| 47 |
// Douglas Thrift |
| 48 |
// |
| 49 |
// $Id: Indexer.cpp,v 1.12 2003/07/15 01:30:46 douglas Exp $ |
| 50 |
|
| 51 |
#include "Indexer.h" |
| 52 |
|
| 53 |
#ifndef _WIN32 |
| 54 |
#include <unistd.h> |
| 55 |
#else // _WIN32 |
| 56 |
inline int unlink(const char* filename) { return DeleteFile(filename); } |
| 57 |
#endif // _WIN32 |
| 58 |
|
| 59 |
Indexer::Indexer(string& indexFile, set<string>& domains, set<string>& |
| 60 |
restrictions) |
| 61 |
{ |
| 62 |
this->indexFile = indexFile; |
| 63 |
this->domains = domains; |
| 64 |
this->restrictions = restrictions; |
| 65 |
} |
| 66 |
|
| 67 |
void Indexer::index(string& begin) |
| 68 |
{ |
| 69 |
unsigned separator = indexFile.rfind(slash); |
| 70 |
string dtd = separator != string::npos ? indexFile.substr(0, separator) + |
| 71 |
slash + "index.dtd" : "index.dtd"; |
| 72 |
|
| 73 |
ifstream fin(dtd.c_str()); |
| 74 |
|
| 75 |
if (!fin.is_open()) |
| 76 |
{ |
| 77 |
ofstream fout(dtd.c_str()); |
| 78 |
|
| 79 |
fout << "<!ELEMENT index (page*)>\n" |
| 80 |
<< "<!ELEMENT page (address, port?, path, title?, description?, ke" |
| 81 |
<< "ywords?, text,\n" |
| 82 |
<< " heading*)\n" |
| 83 |
<< ">\n" |
| 84 |
<< "<!ELEMENT address (#PCDATA)>\n" |
| 85 |
<< "<!ELEMENT port (#PCDATA)>\n" |
| 86 |
<< "<!ELEMENT path (#PCDATA)>\n" |
| 87 |
<< "<!ELEMENT size (#PCDATA)>\n" |
| 88 |
<< "<!ELEMENT title (#PCDATA)>\n" |
| 89 |
<< "<!ELEMENT description (#PCDATA)>\n" |
| 90 |
<< "<!ELEMENT text (#PCDATA)>\n" |
| 91 |
<< "<!ELEMENT heading (#PCDATA)>\n"; |
| 92 |
|
| 93 |
fout.close(); |
| 94 |
} |
| 95 |
|
| 96 |
fin.close(); |
| 97 |
|
| 98 |
string lock = indexFile + ".lock"; |
| 99 |
|
| 100 |
ofstream fout(lock.c_str()); |
| 101 |
fout.close(); |
| 102 |
fout.open(indexFile.c_str()); |
| 103 |
|
| 104 |
fout << "<?xml version=\"1.0\" encoding=\"ISO-8859-1\" standalone=\"no\"?>" |
| 105 |
<< "\n<!DOCTYPE index SYSTEM \"index.dtd\">\n" |
| 106 |
<< "<index>\n"; |
| 107 |
|
| 108 |
URL first(begin); |
| 109 |
|
| 110 |
index(first, fout); |
| 111 |
|
| 112 |
fout << "</index>\n"; |
| 113 |
|
| 114 |
fout.close(); |
| 115 |
|
| 116 |
unlink(lock.c_str()); |
| 117 |
} |
| 118 |
|
| 119 |
void Indexer::index(URL& url, ofstream& fout, const string referer) |
| 120 |
{ |
| 121 |
if (domains.find(url.getAddress() += url.getPort() != 80 ? ":" + |
| 122 |
url.getPort() : "") != domains.end() && pages.find(url.getURL()) == |
| 123 |
pages.end()) |
| 124 |
{ |
| 125 |
if (checked.find(url.getAddress() += url.getPort() != 80 ? ":" + |
| 126 |
url.getPort() : "") == checked.end()) |
| 127 |
{ |
| 128 |
robots(url); |
| 129 |
} |
| 130 |
|
| 131 |
if (!restricted(url)) |
| 132 |
{ |
| 133 |
if (http.handle(url, referer, true)) |
| 134 |
{ |
| 135 |
if (http.contentType().find("text/plain") == 0 || |
| 136 |
http.contentType().find("text/html") == 0) |
| 137 |
{ |
| 138 |
http.clear(); |
| 139 |
if (!http.handle(url, referer)) exit(1); |
| 140 |
|
| 141 |
cout << "Indexing " << url << "..." << flush; |
| 142 |
|
| 143 |
if (processor.process(http, url)) |
| 144 |
{ |
| 145 |
Page page = processor.getPage(); |
| 146 |
fout << page << "\n"; |
| 147 |
|
| 148 |
cout << "done.\n"; |
| 149 |
} |
| 150 |
else |
| 151 |
{ |
| 152 |
cout << "canceled.\n"; |
| 153 |
} |
| 154 |
|
| 155 |
pages.insert(url.getURL()); |
| 156 |
Set pageLinks = processor.getLinks(); |
| 157 |
processor.reset(); |
| 158 |
|
| 159 |
for (SetIterator link = pageLinks.begin(); link != |
| 160 |
pageLinks.end(); link++) |
| 161 |
{ |
| 162 |
if (pages.find(*link) == pages.end()) |
| 163 |
{ |
| 164 |
links.push(URL(*link)); |
| 165 |
referers.push(url.getURL()); |
| 166 |
} |
| 167 |
} |
| 168 |
} |
| 169 |
else |
| 170 |
{ |
| 171 |
// unhandled content |
| 172 |
} |
| 173 |
} |
| 174 |
else if (http.redirect() != "") |
| 175 |
{ |
| 176 |
if (pages.find(http.redirect()) == pages.end()) |
| 177 |
{ |
| 178 |
links.push(URL(http.redirect())); |
| 179 |
referers.push(url.getURL()); |
| 180 |
} |
| 181 |
} |
| 182 |
|
| 183 |
http.clear(); |
| 184 |
} |
| 185 |
} |
| 186 |
|
| 187 |
if (!links.empty()) |
| 188 |
{ |
| 189 |
URL next = links.front(); |
| 190 |
links.pop(); |
| 191 |
|
| 192 |
string referer = referers.front(); |
| 193 |
referers.pop(); |
| 194 |
|
| 195 |
if (debug) cerr << "next = " << next << "\n"; |
| 196 |
|
| 197 |
index(next, fout, referer); |
| 198 |
} |
| 199 |
} |
| 200 |
|
| 201 |
bool Indexer::restricted(URL& url) |
| 202 |
{ |
| 203 |
bool answer = false; |
| 204 |
|
| 205 |
for (SetIterator itor = restrictions.begin(); itor != restrictions.end(); |
| 206 |
itor++) |
| 207 |
{ |
| 208 |
URL checker = *itor; |
| 209 |
|
| 210 |
if (url.getAddress() == checker.getAddress() && url.getPort() == |
| 211 |
checker.getPort()) |
| 212 |
{ |
| 213 |
if (url.getPath().find(checker.getPath()) == 0) |
| 214 |
{ |
| 215 |
answer = true; |
| 216 |
break; |
| 217 |
} |
| 218 |
} |
| 219 |
} |
| 220 |
|
| 221 |
return answer; |
| 222 |
} |
| 223 |
|
| 224 |
void Indexer::robots(URL& url) |
| 225 |
{ |
| 226 |
URL robots = url; |
| 227 |
robots.setPath("/robots.txt"); |
| 228 |
|
| 229 |
if (http.handle(robots)) |
| 230 |
{ |
| 231 |
cout << "Checking " << robots << "..." << flush; |
| 232 |
|
| 233 |
string line; |
| 234 |
|
| 235 |
bool record = false, hasVersion = false, hasName = false, hasAll = |
| 236 |
false; |
| 237 |
robot state = none; |
| 238 |
Set restrictionsVersion, restrictionsName, restrictionsAll; |
| 239 |
|
| 240 |
while (http.good()) |
| 241 |
{ |
| 242 |
http.getline(line); |
| 243 |
|
| 244 |
unsigned comment = line.find('#'); |
| 245 |
if (comment != string::npos) line.erase(comment); |
| 246 |
|
| 247 |
if (line == "" && comment == string::npos) record = false; |
| 248 |
if (line == "") continue; |
| 249 |
|
| 250 |
unsigned colon = line.find(':'); |
| 251 |
|
| 252 |
string field = line.substr(0, colon); |
| 253 |
string value = line.substr(colon + 1); |
| 254 |
|
| 255 |
normalize(value); |
| 256 |
|
| 257 |
if (field == "User-agent" && value == agent(true)) |
| 258 |
{ |
| 259 |
state = version; |
| 260 |
record = true; |
| 261 |
hasVersion = true; |
| 262 |
} |
| 263 |
else if (field == "User-agent" && value == agent(false)) |
| 264 |
{ |
| 265 |
state = name; |
| 266 |
record = true; |
| 267 |
hasName = true; |
| 268 |
} |
| 269 |
else if (field == "User-agent" && value == "*") |
| 270 |
{ |
| 271 |
state = all; |
| 272 |
record = true; |
| 273 |
hasAll = true; |
| 274 |
} |
| 275 |
else if (field == "Disallow" && record && value == "") |
| 276 |
{ |
| 277 |
// no restrictions |
| 278 |
} |
| 279 |
else if (field == "Disallow" && record) |
| 280 |
{ |
| 281 |
URL restriction = robots; |
| 282 |
restriction.setPath(value); |
| 283 |
|
| 284 |
switch (state) |
| 285 |
{ |
| 286 |
case version: |
| 287 |
restrictionsVersion.insert(restriction.getURL()); |
| 288 |
break; |
| 289 |
case name: |
| 290 |
restrictionsName.insert(restriction.getURL()); |
| 291 |
break; |
| 292 |
case all: |
| 293 |
restrictionsAll.insert(restriction.getURL()); |
| 294 |
break; |
| 295 |
default: |
| 296 |
break; |
| 297 |
} |
| 298 |
} |
| 299 |
} |
| 300 |
|
| 301 |
if (hasVersion) |
| 302 |
{ |
| 303 |
state = version; |
| 304 |
} |
| 305 |
else if (hasName) |
| 306 |
{ |
| 307 |
state = name; |
| 308 |
} |
| 309 |
else if (hasAll) |
| 310 |
{ |
| 311 |
state = all; |
| 312 |
} |
| 313 |
else |
| 314 |
{ |
| 315 |
state = none; |
| 316 |
} |
| 317 |
|
| 318 |
SetIterator itor; |
| 319 |
switch (state) |
| 320 |
{ |
| 321 |
case version: |
| 322 |
for (itor = restrictionsVersion.begin(); itor != |
| 323 |
restrictionsVersion.end(); itor++) |
| 324 |
{ |
| 325 |
restrictions.insert(*itor); |
| 326 |
} |
| 327 |
break; |
| 328 |
case name: |
| 329 |
for (itor = restrictionsName.begin(); itor != |
| 330 |
restrictionsName.end(); itor++) |
| 331 |
{ |
| 332 |
restrictions.insert(*itor); |
| 333 |
} |
| 334 |
break; |
| 335 |
case all: |
| 336 |
for (itor = restrictionsAll.begin(); itor != |
| 337 |
restrictionsAll.end(); itor++) |
| 338 |
{ |
| 339 |
restrictions.insert(*itor); |
| 340 |
} |
| 341 |
break; |
| 342 |
default: |
| 343 |
break; |
| 344 |
} |
| 345 |
|
| 346 |
cout << "done.\n"; |
| 347 |
} |
| 348 |
|
| 349 |
http.clear(); |
| 350 |
|
| 351 |
checked.insert(url.getAddress() += url.getPort() != 80 ? ":" + |
| 352 |
url.getPort() : ""); |
| 353 |
} |
