[891] | 1 | /**********************************************************************
|
---|
| 2 | *
|
---|
| 3 | * cstrquerytools.cpp --
|
---|
| 4 | * Copyright (C) 1999 The New Zealand Digital Library Project
|
---|
| 5 | *
|
---|
| 6 | * A component of the Greenstone digital library software
|
---|
| 7 | * from the New Zealand Digital Library Project at the
|
---|
| 8 | * University of Waikato, New Zealand.
|
---|
| 9 | *
|
---|
| 10 | * This program is free software; you can redistribute it and/or modify
|
---|
| 11 | * it under the terms of the GNU General Public License as published by
|
---|
| 12 | * the Free Software Foundation; either version 2 of the License, or
|
---|
| 13 | * (at your option) any later version.
|
---|
| 14 | *
|
---|
| 15 | * This program is distributed in the hope that it will be useful,
|
---|
| 16 | * but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
| 17 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
---|
| 18 | * GNU General Public License for more details.
|
---|
| 19 | *
|
---|
| 20 | * You should have received a copy of the GNU General Public License
|
---|
| 21 | * along with this program; if not, write to the Free Software
|
---|
| 22 | * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
|
---|
| 23 | *
|
---|
| 24 | * $Id: cstrquerytools.cpp 891 2000-02-01 22:32:36Z sjboddie $
|
---|
| 25 | *
|
---|
| 26 | *********************************************************************/
|
---|
| 27 |
|
---|
| 28 | /*
|
---|
| 29 | $Log$
|
---|
| 30 | Revision 1.1 2000/02/01 22:32:37 sjboddie
|
---|
| 31 | Initial revision
|
---|
| 32 |
|
---|
| 33 | Revision 1.1 1999/09/07 21:49:44 sjboddie
|
---|
| 34 | new cstr receptionist
|
---|
| 35 |
|
---|
| 36 |
|
---|
| 37 | */
|
---|
| 38 |
|
---|
| 39 |
|
---|
| 40 | // this is overridden for cstr as we're always going to get
|
---|
| 41 | // maxdocs matches. They can then be sorted in the
|
---|
| 42 | // cstrqueryaction
|
---|
| 43 |
|
---|
| 44 | #include "cstrquerytools.h"
|
---|
| 45 |
|
---|
| 46 |
|
---|
| 47 | // do_query sets the filter options and makes the protocol call to
|
---|
| 48 | // do a query. The results are returned in response.
|
---|
| 49 | // request.filterResultOptions and request.fields (if required)
|
---|
| 50 | // should be set by the calling function.
|
---|
| 51 |
|
---|
| 52 | bool do_query (FilterRequest_t &request, cgiargsclass &args,
|
---|
| 53 | recptproto *collectproto, FilterResponse_t &response,
|
---|
| 54 | ostream &logout) {
|
---|
| 55 |
|
---|
| 56 | request.filterName = "QueryFilter";
|
---|
| 57 |
|
---|
| 58 | comerror_t err;
|
---|
| 59 | OptionValue_t option;
|
---|
| 60 | text_t formattedstring = args["q"];
|
---|
| 61 | format_querystring (formattedstring, args.getintarg("b"));
|
---|
| 62 |
|
---|
| 63 | option.name = "Term";
|
---|
| 64 | option.value = formattedstring;
|
---|
| 65 | request.filterOptions.push_back (option);
|
---|
| 66 |
|
---|
| 67 | option.name = "QueryType";
|
---|
| 68 | option.value = (args.getintarg("t")) ? "ranked" : "boolean";
|
---|
| 69 | request.filterOptions.push_back (option);
|
---|
| 70 |
|
---|
| 71 | option.name = "Casefold";
|
---|
| 72 | option.value = (args.getintarg("k")) ? "true" : "false";
|
---|
| 73 | request.filterOptions.push_back (option);
|
---|
| 74 |
|
---|
| 75 | option.name = "Stem";
|
---|
| 76 | option.value = (args.getintarg("s")) ? "true" : "false";
|
---|
| 77 | request.filterOptions.push_back (option);
|
---|
| 78 |
|
---|
| 79 | if (!args["h"].empty()) {
|
---|
| 80 | option.name = "Index";
|
---|
| 81 | option.value = args["h"];
|
---|
| 82 | request.filterOptions.push_back (option);
|
---|
| 83 | }
|
---|
| 84 |
|
---|
| 85 | if (!args["j"].empty()) {
|
---|
| 86 | option.name = "Subcollection";
|
---|
| 87 | option.value = args["j"];
|
---|
| 88 | request.filterOptions.push_back (option);
|
---|
| 89 | }
|
---|
| 90 |
|
---|
| 91 | if (!args["n"].empty()) {
|
---|
| 92 | option.name = "Language";
|
---|
| 93 | option.value = args["n"];
|
---|
| 94 | request.filterOptions.push_back (option);
|
---|
| 95 | }
|
---|
| 96 |
|
---|
| 97 | // fill in the second query if needed
|
---|
| 98 | if (!args["cq2"].empty()) {
|
---|
| 99 | option.name = "CombineQuery";
|
---|
| 100 | option.value = args["cq2"];
|
---|
| 101 | request.filterOptions.push_back (option);
|
---|
| 102 |
|
---|
| 103 | text_t formattedstring2 = args["q2"];
|
---|
| 104 | format_querystring (formattedstring2, args.getintarg("b"));
|
---|
| 105 |
|
---|
| 106 | option.name = "Term";
|
---|
| 107 | option.value = formattedstring2;
|
---|
| 108 | request.filterOptions.push_back (option);
|
---|
| 109 |
|
---|
| 110 | option.name = "QueryType";
|
---|
| 111 | option.value = (args.getintarg("t")) ? "ranked" : "boolean";
|
---|
| 112 | request.filterOptions.push_back (option);
|
---|
| 113 |
|
---|
| 114 | option.name = "Casefold";
|
---|
| 115 | option.value = (args.getintarg("k")) ? "true" : "false";
|
---|
| 116 | request.filterOptions.push_back (option);
|
---|
| 117 |
|
---|
| 118 | option.name = "Stem";
|
---|
| 119 | option.value = (args.getintarg("s")) ? "true" : "false";
|
---|
| 120 | request.filterOptions.push_back (option);
|
---|
| 121 |
|
---|
| 122 | if (!args["h2"].empty()) {
|
---|
| 123 | option.name = "Index";
|
---|
| 124 | option.value = args["h2"];
|
---|
| 125 | request.filterOptions.push_back (option);
|
---|
| 126 | }
|
---|
| 127 |
|
---|
| 128 | if (!args["j2"].empty()) {
|
---|
| 129 | option.name = "Subcollection";
|
---|
| 130 | option.value = args["j2"];
|
---|
| 131 | request.filterOptions.push_back (option);
|
---|
| 132 | }
|
---|
| 133 |
|
---|
| 134 | if (!args["n2"].empty()) {
|
---|
| 135 | option.name = "Language";
|
---|
| 136 | option.value = args["n2"];
|
---|
| 137 | request.filterOptions.push_back (option);
|
---|
| 138 | }
|
---|
| 139 | }
|
---|
| 140 |
|
---|
| 141 | // we always want to retrieve all documents (i.e. 1 to maxdocs)
|
---|
| 142 | // for this collection
|
---|
| 143 | option.name = "Maxdocs";
|
---|
| 144 | option.value = args["m"];
|
---|
| 145 | request.filterOptions.push_back (option);
|
---|
| 146 |
|
---|
| 147 | option.name = "StartResults";
|
---|
| 148 | option.value = 1;
|
---|
| 149 | request.filterOptions.push_back (option);
|
---|
| 150 |
|
---|
| 151 | option.name = "EndResults";
|
---|
| 152 | option.value = args["m"];
|
---|
| 153 | request.filterOptions.push_back (option);
|
---|
| 154 |
|
---|
| 155 | collectproto->filter (args["c"], request, response, err, logout);
|
---|
| 156 |
|
---|
| 157 | if (err != noError) {
|
---|
| 158 | outconvertclass text_t2ascii;
|
---|
| 159 | logout << text_t2ascii
|
---|
| 160 | << "Error: call to QueryFilter failed in queryaction ("
|
---|
| 161 | << get_comerror_string (err) << ")\n";
|
---|
| 162 | return false;
|
---|
| 163 | }
|
---|
| 164 | return true;
|
---|
| 165 | }
|
---|
| 166 |
|
---|
| 167 | void format_querystring (text_t &querystring, int querymode) {
|
---|
| 168 | text_t formattedstring;
|
---|
| 169 |
|
---|
| 170 | text_t::const_iterator here = querystring.begin();
|
---|
| 171 | text_t::const_iterator end = querystring.end();
|
---|
| 172 |
|
---|
| 173 | // space is used to insert spaces between Chinese
|
---|
| 174 | // characters. No space is needed before the first
|
---|
| 175 | // Chinese character.
|
---|
| 176 | bool space = false;
|
---|
| 177 |
|
---|
| 178 | // want to remove ()|!& from querystring so boolean queries are just
|
---|
| 179 | // "all the words" queries (unless querymode is advanced)
|
---|
| 180 | while (here != end) {
|
---|
| 181 | if ((querymode == 0) && (*here == '(' || *here == ')' || *here == '|' ||
|
---|
| 182 | *here == '!' || *here == '&')) {
|
---|
| 183 | formattedstring.push_back(' ');
|
---|
| 184 | } else {
|
---|
| 185 | if ((*here >= 0x4e00 && *here <= 0x9fa5) ||
|
---|
| 186 | (*here >= 0xf900 && *here <= 0xfa2d)) {
|
---|
| 187 | // Chinese character
|
---|
| 188 | if (space) formattedstring.push_back (0x200b);
|
---|
| 189 | formattedstring.push_back (*here);
|
---|
| 190 | formattedstring.push_back (0x200b);
|
---|
| 191 | space = true;
|
---|
| 192 | } else {
|
---|
| 193 | // non-Chinese character
|
---|
| 194 | formattedstring.push_back (*here);
|
---|
| 195 | space = false;
|
---|
| 196 | }
|
---|
| 197 | }
|
---|
| 198 | here ++;
|
---|
| 199 | }
|
---|
| 200 | querystring = formattedstring;
|
---|
| 201 | }
|
---|
| 202 |
|
---|
| 203 | void get_phrases (const text_t &querystring, text_tarray &phrases) {
|
---|
| 204 |
|
---|
| 205 | phrases.erase (phrases.begin(), phrases.end());
|
---|
| 206 | if (!querystring.empty()) {
|
---|
| 207 |
|
---|
| 208 | text_t::const_iterator end = querystring.end();
|
---|
| 209 | text_t::const_iterator here = findchar (querystring.begin(), end, '"');
|
---|
| 210 | if (here != end) {
|
---|
| 211 | text_t tmptext;
|
---|
| 212 | bool foundquote = false;
|
---|
| 213 | while (here != end) {
|
---|
| 214 | if (*here == '"') {
|
---|
| 215 | if (foundquote) {
|
---|
| 216 | if (!tmptext.empty()) {
|
---|
| 217 | phrases.push_back(tmptext);
|
---|
| 218 | tmptext.clear();
|
---|
| 219 | }
|
---|
| 220 | foundquote = false;
|
---|
| 221 | } else foundquote = true;
|
---|
| 222 | } else {
|
---|
| 223 | if (foundquote) tmptext.push_back (*here);
|
---|
| 224 | }
|
---|
| 225 | here ++;
|
---|
| 226 | }
|
---|
| 227 | }
|
---|
| 228 | }
|
---|
| 229 | }
|
---|
| 230 |
|
---|