ViewVC Help
View File | Revision Log | Show Annotations | Download File | View Changeset | Root Listing
root/proj/trunk/Search/Indexer.cpp
(Generate patch)

Comparing trunk/Search/Indexer.cpp (file contents):
Revision 197 by douglas, 2003-07-14T18:30:46-07:00 vs.
Revision 312 by douglas, 2004-01-01T15:00:34-08:00

# Line 1 | Line 1
1   /* ============================================================================
2   * Douglas Thrift's Search Engine License
3   *
4 < * Copyright (C) 2002-2003, Douglas Thrift. All Rights Reserved.
4 > * Copyright (C) 2002-2004, 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   *
# Line 46 | Line 46
46   //
47   // Douglas Thrift
48   //
49 < // $Id: Indexer.cpp,v 1.12 2003/07/15 01:30:46 douglas Exp $
49 > // $Id: Indexer.cpp,v 1.16 2004/01/01 23:00:34 douglas Exp $
50  
51   #include "Indexer.h"
52  
# Line 77 | Line 77 | void Indexer::index(string& begin)
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"
80 >                        << "<!ELEMENT page (address, port?, tls?, path, title?, descriptio"
81 >                        << "n?, keywords?, text, heading*)>\n"
82                          << "<!ELEMENT address (#PCDATA)>\n"
83                          << "<!ELEMENT port (#PCDATA)>\n"
84 +                        << "<!ELEMENT tls (#PCDATA)>\n"
85                          << "<!ELEMENT path (#PCDATA)>\n"
86                          << "<!ELEMENT size (#PCDATA)>\n"
87                          << "<!ELEMENT title (#PCDATA)>\n"
# Line 118 | Line 117 | void Indexer::index(string& begin)
117  
118   void Indexer::index(URL& url, ofstream& fout, const string referer)
119   {
120 <        if (domains.find(url.getAddress() += url.getPort() != 80 ? ":" +
121 <                url.getPort() : "") != domains.end() && pages.find(url.getURL()) ==
123 <                pages.end())
120 >        if (domains.find(url.getAddress()) != domains.end() &&
121 >                pages.find(url.getURL()) == pages.end())
122          {
123 <                if (checked.find(url.getAddress() += url.getPort() != 80 ? ":" +
124 <                        url.getPort() : "") == checked.end())
123 >                if (checked.find(url.getAddress() + (url.getPort() != 80 ? ":" +
124 >                        url.getPort() : string(""))) == checked.end())
125                  {
126                          robots(url);
127                  }
# Line 138 | Line 136 | void Indexer::index(URL& url, ofstream&
136                                          http.clear();
137                                          if (!http.handle(url, referer)) exit(1);
138  
139 <                                        cout << "Indexing " << url << "..." << flush;
139 >                                        cout << "Indexing " << url << " ... " << flush;
140  
141                                          if (processor.process(http, url))
142                                          {
# Line 228 | Line 226 | void Indexer::robots(URL& url)
226  
227          if (http.handle(robots))
228          {
229 <                cout << "Checking " << robots << "..." << flush;
229 >                cout << "Checking " << robots << " ... " << flush;
230  
231                  string line;
232  

Diff Legend

– Removed lines
+ Added lines
< Changed lines
> Changed lines