1 | /**********************************************************************
|
---|
2 | *
|
---|
3 | * txt2tdb.cpp -- A utility to convert a stream of text, ala buildproc
|
---|
4 | * encoded output, into a TDB file.
|
---|
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 | * Copyright (C) 2011 The New Zealand Digital Library Project
|
---|
11 | *
|
---|
12 | * This program is free software; you can redistribute it and/or modify
|
---|
13 | * it under the terms of the GNU General Public License as published by
|
---|
14 | * the Free Software Foundation; either version 2 of the License, or
|
---|
15 | * (at your option) any later version.
|
---|
16 | *
|
---|
17 | * This program is distributed in the hope that it will be useful,
|
---|
18 | * but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
19 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
---|
20 | * GNU General Public License for more details.
|
---|
21 | *
|
---|
22 | * You should have received a copy of the GNU General Public License
|
---|
23 | * along with this program; if not, write to the Free Software
|
---|
24 | * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
|
---|
25 | *
|
---|
26 | **********************************************************************/
|
---|
27 |
|
---|
28 | #if defined(GSDL_USE_OBJECTSPACE)
|
---|
29 | #include <ospace\std\iostream>
|
---|
30 | #elif defined(GSDL_USE_IOS_H)
|
---|
31 | #include <iostream.h>
|
---|
32 | #else
|
---|
33 | #include <iostream>
|
---|
34 | #endif
|
---|
35 |
|
---|
36 | #include "tdb.h"
|
---|
37 | #include "text_t.h"
|
---|
38 |
|
---|
39 | // use the standard namespace
|
---|
40 | #if !defined (GSDL_NAMESPACE_BROKEN)
|
---|
41 | #if defined(GSDL_USE_OBJECTSPACE)
|
---|
42 | using namespace ospace::std;
|
---|
43 | #else
|
---|
44 | using namespace std;
|
---|
45 | #endif
|
---|
46 | #endif
|
---|
47 |
|
---|
48 | void
|
---|
49 | printUsage (char *program_name)
|
---|
50 | {
|
---|
51 | cerr << "usage: " << program_name << " [options] database-name" << endl << endl;
|
---|
52 | cerr << "options:" << endl;
|
---|
53 | cerr << " -append append to existing database" << endl << endl;
|
---|
54 | }
|
---|
55 | /** printUsage() **/
|
---|
56 |
|
---|
57 | int
|
---|
58 | main (int argc, char *argv[])
|
---|
59 | {
|
---|
60 | // sanity check
|
---|
61 | if (argc != 2 && argc != 3)
|
---|
62 | {
|
---|
63 | printUsage (argv[0]);
|
---|
64 | exit (0);
|
---|
65 | }
|
---|
66 |
|
---|
67 | char *dbname;
|
---|
68 | int append = 0;
|
---|
69 | int delkey = 0;
|
---|
70 | if (argc == 3)
|
---|
71 | {
|
---|
72 | if (strcmp (argv[1], "-append") == 0)
|
---|
73 | {
|
---|
74 | append = 1;
|
---|
75 | dbname = argv[2];
|
---|
76 | }
|
---|
77 | else
|
---|
78 | {
|
---|
79 | cerr << argv[1] << " is not a valid option." << endl << endl;
|
---|
80 | printUsage(argv[0]);
|
---|
81 | exit (0);
|
---|
82 | }
|
---|
83 | }
|
---|
84 | else
|
---|
85 | {
|
---|
86 | dbname = argv[1];
|
---|
87 | }
|
---|
88 |
|
---|
89 | // open the database
|
---|
90 | int hash_size = 0;
|
---|
91 | int tdb_flags = TDB_DEFAULT; // Default = 0
|
---|
92 | if (append == 0)
|
---|
93 | {
|
---|
94 | tdb_flags = TDB_CLEAR_IF_FIRST;
|
---|
95 | }
|
---|
96 | int tdb_store_flags = TDB_DEFAULT; // used later when storing
|
---|
97 | int open_flags = O_RDWR | O_CREAT;
|
---|
98 | TDB_CONTEXT *tdb = tdb_open(dbname, hash_size, tdb_flags, open_flags, 0664);
|
---|
99 | if (!tdb)
|
---|
100 | {
|
---|
101 | cerr << "couldn't create " << dbname << endl;
|
---|
102 | exit (0);
|
---|
103 | }
|
---|
104 |
|
---|
105 | char c;
|
---|
106 | cin.get(c);
|
---|
107 | while (!cin.eof())
|
---|
108 | {
|
---|
109 | int num_dashes = 0;
|
---|
110 | text_t key = "";
|
---|
111 | text_t value = "";
|
---|
112 |
|
---|
113 | // Parse out 'key' from [key]\n
|
---|
114 | // - scan for first occurrence of [
|
---|
115 | while (!cin.eof() && c != '[')
|
---|
116 | {
|
---|
117 | cin.get(c);
|
---|
118 | }
|
---|
119 | // - skip [
|
---|
120 | if (!cin.eof())
|
---|
121 | {
|
---|
122 | cin.get(c);
|
---|
123 | }
|
---|
124 | // - now look for closing ], building up 'key' as we go
|
---|
125 | while (!cin.eof() && c != ']')
|
---|
126 | {
|
---|
127 | key.push_back ((unsigned char)c);
|
---|
128 | cin.get(c);
|
---|
129 | }
|
---|
130 | if (!cin.eof())
|
---|
131 | {
|
---|
132 | // most likely an eol char, but if '-', then signifies record
|
---|
133 | // is to be deleted, not added
|
---|
134 | cin.get(c);
|
---|
135 | if (c == '-') {
|
---|
136 | delkey = 1;
|
---|
137 | }
|
---|
138 | else {
|
---|
139 | delkey = 0;
|
---|
140 | }
|
---|
141 | }
|
---|
142 | while (!cin.eof() && (c == '\n' || c == '\r'))
|
---|
143 | {
|
---|
144 | cin.get(c);
|
---|
145 | }
|
---|
146 | // - read in the value, watching for 70 dashes (the end)
|
---|
147 | text_t tmp = "";
|
---|
148 | while (!cin.eof() && (num_dashes < 70))
|
---|
149 | {
|
---|
150 | if (c == '\n')
|
---|
151 | {
|
---|
152 | tmp.push_back ((unsigned char)c);
|
---|
153 | num_dashes = 0;
|
---|
154 | }
|
---|
155 | else if (c == '\r')
|
---|
156 | {
|
---|
157 | // Here we are able to process both Windows-specific text files
|
---|
158 | // (containing carriage-return, newline) and Linux text files
|
---|
159 | // (containing only newline characters) by ignoring the Windows'
|
---|
160 | // carriage-return altogether so that we produce a uniform database
|
---|
161 | // file from either system's type of text file.
|
---|
162 | // If we don't ignore the carriage return here, txt.gz files
|
---|
163 | // produced on Windows cause a GS library running on Linux to break.
|
---|
164 | num_dashes = 0;
|
---|
165 | }
|
---|
166 | else if (c == '-')
|
---|
167 | {
|
---|
168 | tmp.push_back ((unsigned char)c);
|
---|
169 | ++num_dashes;
|
---|
170 | }
|
---|
171 | else
|
---|
172 | {
|
---|
173 | value += tmp;
|
---|
174 | value.push_back ((unsigned char)c);
|
---|
175 | tmp = "";
|
---|
176 | num_dashes = 0;
|
---|
177 | }
|
---|
178 | cin.get(c);
|
---|
179 | }
|
---|
180 |
|
---|
181 | // We should now have a key/value pair. If the key is not an empty string
|
---|
182 | // store this key-value pair
|
---|
183 | if (!key.empty())
|
---|
184 | {
|
---|
185 | // convert key to a datum datatype
|
---|
186 | TDB_DATA key_data;
|
---|
187 | // [why are cstrings from text_t not unsigned? from what I can see we
|
---|
188 | // explicitly cast to char * when we return]
|
---|
189 | key_data.dptr = (unsigned char*)key.getcstr();
|
---|
190 | if (key_data.dptr == NULL)
|
---|
191 | {
|
---|
192 | cerr << "NULL key_data.dptr" << endl;
|
---|
193 | exit (0);
|
---|
194 | }
|
---|
195 | key_data.dsize = key.size();
|
---|
196 | // - delete the given key if we've been asked to
|
---|
197 | if (delkey)
|
---|
198 | {
|
---|
199 | if (tdb_delete(tdb, key_data) < 0)
|
---|
200 | {
|
---|
201 | cerr << "tdb_delete returned an error" << endl;
|
---|
202 | }
|
---|
203 | }
|
---|
204 | // - otherwise add
|
---|
205 | else {
|
---|
206 | // - convert value to a datum datatype
|
---|
207 | TDB_DATA value_data;
|
---|
208 | value_data.dptr = (unsigned char*)value.getcstr();
|
---|
209 | if (value_data.dptr == NULL)
|
---|
210 | {
|
---|
211 | cerr << "NULL value_data.dptr" << endl;
|
---|
212 | exit (0);
|
---|
213 | }
|
---|
214 | value_data.dsize = value.size();
|
---|
215 | // - store the value
|
---|
216 | if (tdb_store(tdb, key_data, value_data, tdb_store_flags) < 0)
|
---|
217 | {
|
---|
218 | cerr << "tdb_store returned an error" << endl;
|
---|
219 | exit (0);
|
---|
220 | }
|
---|
221 | free(value_data.dptr);
|
---|
222 | }
|
---|
223 | free(key_data.dptr);
|
---|
224 | }
|
---|
225 | }
|
---|
226 |
|
---|
227 | // Close the database connection
|
---|
228 | if (tdb_close(tdb) < 0)
|
---|
229 | {
|
---|
230 | cerr << "tdb_close returned an error" << endl;
|
---|
231 | exit (0);
|
---|
232 | }
|
---|
233 |
|
---|
234 | return 0;
|
---|
235 | }
|
---|