1 | /**************************************************************************
|
---|
2 | *
|
---|
3 | * mgpp_invf_dict.cpp -- Program to build the blocked stemmed dictionary
|
---|
4 | * Copyright (C) 1999 Rodger McNab
|
---|
5 | *
|
---|
6 | * This program is free software; you can redistribute it and/or modify
|
---|
7 | * it under the terms of the GNU General Public License as published by
|
---|
8 | * the Free Software Foundation; either version 2 of the License, or
|
---|
9 | * (at your option) any later version.
|
---|
10 | *
|
---|
11 | * This program is distributed in the hope that it will be useful,
|
---|
12 | * but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
13 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
---|
14 | * GNU General Public License for more details.
|
---|
15 | *
|
---|
16 | * You should have received a copy of the GNU General Public License
|
---|
17 | * along with this program; if not, write to the Free Software
|
---|
18 | * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
|
---|
19 | *
|
---|
20 | **************************************************************************/
|
---|
21 |
|
---|
22 | #define _XOPEN_SOURCE 1
|
---|
23 | #define _XOPEN_SOURCE_EXTENDED 1
|
---|
24 |
|
---|
25 | // need this to avoid bizarre compiler problems under VC++ 6.0
|
---|
26 | #if defined (__WIN32__) && !defined (GSDL_USE_IOS_H)
|
---|
27 | # include <iostream>
|
---|
28 | #endif
|
---|
29 |
|
---|
30 | /* getopt is in posix.2, so cygwin should have it in unistd, but doesn't */
|
---|
31 | #if defined (__WIN32__) || defined (__CYGWIN__)
|
---|
32 | # include "getopt_old.h"
|
---|
33 | #else
|
---|
34 | # include <unistd.h>
|
---|
35 | #endif
|
---|
36 |
|
---|
37 | #include "sysfuncs.h"
|
---|
38 | #include "messages.h"
|
---|
39 | #include "mg_files.h"
|
---|
40 | #include "invf.h"
|
---|
41 |
|
---|
42 | static void process_files (char *filename, unsigned long entriesPerBlock) {
|
---|
43 | // open the dictionary
|
---|
44 | FILE *dictFile = open_file (filename, INVF_DICT_SUFFIX, "rb",
|
---|
45 | MAGIC_STEM_BUILD, MG_ABORT);
|
---|
46 | invf_dict_header idh;
|
---|
47 | idh.Read (dictFile);
|
---|
48 |
|
---|
49 | // open the inverted index file
|
---|
50 | FILE *invfIdxFile = open_file (filename, INVF_IDX_SUFFIX, "rb",
|
---|
51 | MAGIC_INVI, MG_ABORT);
|
---|
52 |
|
---|
53 | // create the blocked dictionary
|
---|
54 | FILE *blockDictFile = create_file (filename, INVF_DICT_BLOCKED_SUFFIX, "wb",
|
---|
55 | MAGIC_STEM, MG_ABORT);
|
---|
56 | block_dict_header bdh;
|
---|
57 | bdh.lookback = idh.lookback;
|
---|
58 | bdh.word_dict_start = idh.word_dict_start;
|
---|
59 | bdh.word_dict_size = idh.word_dict_size;
|
---|
60 | bdh.tag_dict_start = idh.tag_dict_start;
|
---|
61 | bdh.tag_dict_size = idh.tag_dict_size;
|
---|
62 | bdh.num_docs = idh.num_docs;
|
---|
63 | bdh.num_frags = idh.num_frags;
|
---|
64 | bdh.num_words = idh.num_words;
|
---|
65 | bdh.total_bytes = idh.total_bytes;
|
---|
66 | bdh.index_string_bytes = idh.index_string_bytes;
|
---|
67 | bdh.num_levels = idh.num_levels;
|
---|
68 | bdh.Write (blockDictFile);
|
---|
69 |
|
---|
70 |
|
---|
71 | // write out the word part of the dictionary
|
---|
72 |
|
---|
73 | bdh.entries_per_wblk = entriesPerBlock;
|
---|
74 | bdh.max_wblk_size = 0;
|
---|
75 | bdh.wblk_start = ftell (blockDictFile);
|
---|
76 |
|
---|
77 | fseek (dictFile, idh.word_dict_start, SEEK_SET);
|
---|
78 |
|
---|
79 | block_idx wordIdx;
|
---|
80 | word_block_dict_el wordBlockEl;
|
---|
81 | wordBlockEl.SetNumLevels (idh.num_levels);
|
---|
82 |
|
---|
83 |
|
---|
84 | unsigned long wordNum;
|
---|
85 | unsigned long wordInvfPtr;
|
---|
86 | UCArray lastEl;
|
---|
87 | word_dict_el wordEl;
|
---|
88 | wordEl.SetNumLevels (idh.num_levels);
|
---|
89 | for (wordNum=0; wordNum<idh.word_dict_size; ++wordNum) {
|
---|
90 | // read in the next word and inverted file pointer
|
---|
91 | wordEl.Read (dictFile, idh.num_levels);
|
---|
92 | ReadUL (invfIdxFile, wordInvfPtr);
|
---|
93 |
|
---|
94 | // remember this word (and position) if this is the start
|
---|
95 | // of a new block
|
---|
96 | if (wordNum % entriesPerBlock == 0) {
|
---|
97 | block_idx_info elIdx;
|
---|
98 | elIdx.el = wordEl.el;
|
---|
99 | elIdx.block_ptr = ftell (blockDictFile);
|
---|
100 |
|
---|
101 | // see if this block is the longest so far
|
---|
102 | if (wordIdx.size() > 0) {
|
---|
103 | unsigned long blockLen = elIdx.block_ptr -
|
---|
104 | (*(wordIdx.end()-1)).block_ptr;
|
---|
105 | if (blockLen > bdh.max_wblk_size) bdh.max_wblk_size = blockLen;
|
---|
106 | }
|
---|
107 |
|
---|
108 | wordIdx.push_back (elIdx);
|
---|
109 | lastEl.erase (lastEl.begin(), lastEl.end()); // output full word
|
---|
110 | }
|
---|
111 |
|
---|
112 | // copy the information for this word
|
---|
113 | wordBlockEl.el = wordEl.el;
|
---|
114 | wordBlockEl.frag_occur = wordEl.frag_occur;
|
---|
115 | wordBlockEl.freq = wordEl.freq;
|
---|
116 | wordBlockEl.invf_ptr = wordInvfPtr;
|
---|
117 | unsigned long tempI;
|
---|
118 | for (tempI=0; tempI<idh.num_levels; ++tempI)
|
---|
119 | wordBlockEl.levelFreqs[tempI] = wordEl.levelFreqs[tempI];
|
---|
120 |
|
---|
121 | // write out the word
|
---|
122 | wordBlockEl.Write (blockDictFile, &lastEl, idh.num_levels);
|
---|
123 |
|
---|
124 | lastEl = wordBlockEl.el;
|
---|
125 | }
|
---|
126 |
|
---|
127 |
|
---|
128 | // write out the tag part of the dictionary
|
---|
129 |
|
---|
130 | bdh.entries_per_tblk = entriesPerBlock;
|
---|
131 | bdh.max_tblk_size = 0;
|
---|
132 | bdh.tblk_start = ftell (blockDictFile);
|
---|
133 |
|
---|
134 | fseek (dictFile, idh.tag_dict_start, SEEK_SET);
|
---|
135 |
|
---|
136 | block_idx tagIdx;
|
---|
137 | block_dict_el tagBlockEl;
|
---|
138 |
|
---|
139 | unsigned long tagNum;
|
---|
140 | unsigned long tagInvfPtr;
|
---|
141 | dict_el tagEl;
|
---|
142 | lastEl.erase (lastEl.begin(), lastEl.end());
|
---|
143 | for (tagNum=0; tagNum<idh.tag_dict_size; ++tagNum) {
|
---|
144 | // read in the next tag and inverted file pointer
|
---|
145 | tagEl.Read (dictFile);
|
---|
146 | ReadUL (invfIdxFile, tagInvfPtr);
|
---|
147 |
|
---|
148 | // remember this tag (and position) if this is the start
|
---|
149 | // of a new block
|
---|
150 | if (tagNum % entriesPerBlock == 0) {
|
---|
151 | block_idx_info elIdx;
|
---|
152 | elIdx.el = tagEl.el;
|
---|
153 | elIdx.block_ptr = ftell (blockDictFile);
|
---|
154 |
|
---|
155 | // see if this block is the longest so far
|
---|
156 | if (tagIdx.size() > 0) {
|
---|
157 | unsigned long blockLen = elIdx.block_ptr -
|
---|
158 | (*(tagIdx.end()-1)).block_ptr;
|
---|
159 | if (blockLen > bdh.max_tblk_size) bdh.max_tblk_size = blockLen;
|
---|
160 | }
|
---|
161 |
|
---|
162 | tagIdx.push_back (elIdx);
|
---|
163 | lastEl.erase (lastEl.begin(), lastEl.end()); // output full word
|
---|
164 | }
|
---|
165 |
|
---|
166 | // copy the information for this tag
|
---|
167 | tagBlockEl.el = tagEl.el;
|
---|
168 | tagBlockEl.frag_occur = tagEl.frag_occur;
|
---|
169 | tagBlockEl.freq = tagEl.freq;
|
---|
170 | tagBlockEl.invf_ptr = tagInvfPtr;
|
---|
171 |
|
---|
172 | // write out the tag
|
---|
173 | tagBlockEl.Write (blockDictFile, &lastEl);
|
---|
174 |
|
---|
175 | lastEl = tagBlockEl.el;
|
---|
176 | }
|
---|
177 |
|
---|
178 |
|
---|
179 | // write out the element indexes
|
---|
180 | bdh.num_wblks = wordIdx.size();
|
---|
181 | bdh.wblk_idx_start = ftell (blockDictFile);
|
---|
182 | WriteBlockIdx (blockDictFile, wordIdx);
|
---|
183 |
|
---|
184 | bdh.num_tblks = tagIdx.size();
|
---|
185 | bdh.tblk_idx_start = ftell (blockDictFile);
|
---|
186 | WriteBlockIdx (blockDictFile, tagIdx);
|
---|
187 |
|
---|
188 | // write out the blocked dictionary header
|
---|
189 | fseek (blockDictFile, sizeof(unsigned long), SEEK_SET);
|
---|
190 | bdh.Write (blockDictFile);
|
---|
191 |
|
---|
192 |
|
---|
193 | // close open files
|
---|
194 | fclose (blockDictFile);
|
---|
195 | fclose (invfIdxFile);
|
---|
196 | fclose (dictFile);
|
---|
197 |
|
---|
198 | // print out information
|
---|
199 | #ifndef SILENT
|
---|
200 | Message ("Max word block size = %d\n", bdh.max_wblk_size);
|
---|
201 | Message ("Max tag block size = %d\n", bdh.max_tblk_size);
|
---|
202 | Message ("Number of word blocks written = %d\n", bdh.num_wblks);
|
---|
203 | Message ("Number of tag blocks written = %d\n", bdh.num_tblks);
|
---|
204 | #endif
|
---|
205 | }
|
---|
206 |
|
---|
207 |
|
---|
208 | int main (int argc, char **argv) {
|
---|
209 | unsigned long entriesPerBlock = 16;
|
---|
210 | char *filename = "";
|
---|
211 | int ch;
|
---|
212 | msg_prefix = argv[0];
|
---|
213 | opterr = 0;
|
---|
214 |
|
---|
215 | while ((ch = getopt (argc, argv, "f:d:b:h")) != -1) {
|
---|
216 | switch (ch) {
|
---|
217 | case 'f': // input file
|
---|
218 | filename = optarg;
|
---|
219 | break;
|
---|
220 | case 'd':
|
---|
221 | set_basepath (optarg);
|
---|
222 | break;
|
---|
223 | case 'b':
|
---|
224 | entriesPerBlock = atoi (optarg);
|
---|
225 | break;
|
---|
226 | case 'h':
|
---|
227 | case '?':
|
---|
228 | fprintf (stderr, "usage: %s [-f input_file] "
|
---|
229 | "[-d data directory] [-b entries-per-block] "
|
---|
230 | "[-h]\n", argv[0]);
|
---|
231 | exit (1);
|
---|
232 | }
|
---|
233 | }
|
---|
234 |
|
---|
235 | process_files (filename, entriesPerBlock);
|
---|
236 | return 0;
|
---|
237 | }
|
---|