1 /* xapian-dump: Create a textual dump of a Xapian database.
3 * Copyright © 2009 Carl Worth
5 * This program is free software: you can redistribute it and/or modify
6 * it under the terms of the GNU General Public License as published by
7 * the Free Software Foundation, either version 3 of the License, or
8 * (at your option) any later version.
10 * This program is distributed in the hope that it will be useful,
11 * but WITHOUT ANY WARRANTY; without even the implied warranty of
12 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
13 * GNU General Public License for more details.
15 * You should have received a copy of the GNU General Public License
16 * along with this program. If not, see http://www.gnu.org/licenses/ .
18 * Author: Carl Worth <cworth@cworth.org>
21 /* Currently the dumped data includes:
25 * And for each document ID:
40 vector<int> UNSERIALIZE;
42 unsigned int MAX_TERMS = 0;
45 print_escaped_string (const char *s)
60 print_document_terms (Xapian::Document doc)
62 Xapian::TermIterator it;
67 for (it = doc.termlist_begin (), i = 0;
68 it != doc.termlist_end ();
72 print_escaped_string ((*it).c_str());
76 for ( ; i < MAX_TERMS; i++)
83 vector_int_contains (vector<int> v, int i)
85 vector<int>::iterator result;
87 result = find (v.begin(), v.end(), i);
89 return result != v.end();
93 print_document_values (Xapian::Document doc)
95 Xapian::ValueIterator i;
96 int value_no, value_int;
99 for (i = doc.values_begin (); i != doc.values_end (); i++) {
100 value_no = i.get_valueno();
104 if (vector_int_contains (UNSERIALIZE, value_no)) {
105 value_float = Xapian::sortable_unserialise (*i);
106 value_int = value_float;
107 if (value_int == value_float)
108 printf ("%d", value_int);
110 printf ("\"%f\"", value_float);
112 print_escaped_string ((*i).c_str ());
121 print_document (Xapian::Database db, Xapian::docid id)
123 Xapian::Document doc;
127 doc = db.get_document (id);
129 printf (" \"%s\",\n", doc.get_data ().c_str());
131 print_document_terms (doc);
133 print_document_values (doc);
139 main (int argc, char *argv[])
141 const char *database_path;
145 fprintf (stderr, "Usage: %s <path-to-xapian-database> [value_nos...]\n",
147 fprintf (stderr, "Dumps data from the given database.\n");
148 fprintf (stderr, "The values corresponding to any value numbers given on the command line\n");
149 fprintf (stderr, "will be unserialized to an before being printed.\n");
153 database_path = argv[1];
155 UNSERIALIZE = vector<int> ();
157 for (i = 2; i < argc; i++)
158 UNSERIALIZE.push_back (atoi (argv[i]));
162 Xapian::PostingIterator i;
163 Xapian::docid doc_id;
165 db = Xapian::Database (database_path);
167 for (i = db.postlist_begin (""); i != db.postlist_end (""); i++) {
168 Xapian::Document doc;
172 doc = db.get_document (doc_id);
174 if (doc.termlist_count () > MAX_TERMS)
175 MAX_TERMS = doc.termlist_count ();
178 printf ("#define MAX_TERMS %d\n\n", MAX_TERMS);
180 printf ("typedef struct {\n"
182 " char terms[MAX_TERMS][255];\n"
183 " char message_id[255];\n"
184 " char thread_id[4096];\n"
186 "} document_dump_t;\n\n");
188 printf ("document_dump_t dump[] = {\n");
190 for (i = db.postlist_begin (""); i != db.postlist_end (""); i++) {
193 print_document (db, doc_id);
198 } catch (const Xapian::Error &error) {
199 cerr << "A Xapian exception occurred: " << error.get_msg () << endl;