-/*
- * Copyright (C) 1994-1995, Index Data I/S
- * All rights reserved.
- * Sebastian Hammer, Adam Dickmeiss
- *
- * $Log: main.c,v $
- * Revision 1.24 1995-11-30 17:01:38 adam
- * New setting commitCache: points to commit directories/files.
- * New command commit: commits at the end of a zebraidx run.
- *
- * Revision 1.23 1995/11/30 08:34:31 adam
- * Started work on commit facility.
- * Changed a few malloc/free to xmalloc/xfree.
- *
- * Revision 1.22 1995/11/28 09:09:42 adam
- * Zebra config renamed.
- * Use setting 'recordId' to identify record now.
- * Bug fix in recindex.c: rec_release_blocks was invokeded even
- * though the blocks were already released.
- * File traversal properly deletes records when needed.
- *
- * Revision 1.21 1995/11/27 14:27:39 adam
- * Renamed 'update' command to 'dir'.
- *
- * Revision 1.20 1995/11/27 13:58:53 adam
- * New option -t. storeStore data implemented in server.
- *
- * Revision 1.19 1995/11/25 10:24:06 adam
- * More record fields - they are enumerated now.
- * New options: flagStoreData flagStoreKey.
- *
- * Revision 1.18 1995/11/22 17:19:17 adam
- * Record management uses the bfile system.
- *
- * Revision 1.17 1995/11/21 15:01:16 adam
- * New general match criteria implemented.
- * New feature: document groups.
- *
- * Revision 1.16 1995/11/20 11:56:27 adam
- * Work on new traversal.
- *
- * Revision 1.15 1995/11/01 16:25:51 quinn
- * *** empty log message ***
- *
- * Revision 1.14 1995/10/17 18:02:09 adam
- * New feature: databases. Implemented as prefix to words in dictionary.
- *
- * Revision 1.13 1995/10/10 12:24:39 adam
- * Temporary sort files are compressed.
- *
- * Revision 1.12 1995/10/04 16:57:20 adam
- * Key input and merge sort in one pass.
- *
- * Revision 1.11 1995/09/29 14:01:45 adam
- * Bug fixes.
- *
- * Revision 1.10 1995/09/28 14:22:57 adam
- * Sort uses smaller temporary files.
- *
- * Revision 1.9 1995/09/14 07:48:24 adam
- * Record control management.
- *
- * Revision 1.8 1995/09/06 16:11:18 adam
- * Option: only one word key per file.
- *
- * Revision 1.7 1995/09/05 15:28:39 adam
- * More work on search engine.
- *
- * Revision 1.6 1995/09/04 12:33:43 adam
- * Various cleanup. YAZ util used instead.
- *
- * Revision 1.5 1995/09/04 09:10:39 adam
- * More work on index add/del/update.
- * Merge sort implemented.
- * Initial work on z39 server.
- *
- * Revision 1.4 1995/09/01 14:06:36 adam
- * Split of work into more files.
- *
- * Revision 1.3 1995/09/01 10:57:07 adam
- * Minor changes.
- *
- * Revision 1.2 1995/09/01 10:30:24 adam
- * More work on indexing. Not working yet.
- *
- * Revision 1.1 1995/08/31 14:50:24 adam
- * New simple file index tool.
- *
- */
+/* $Id: main.c,v 1.96 2002-09-03 11:44:54 adam Exp $
+ Copyright (C) 1995,1996,1997,1998,1999,2000,2001,2002
+ Index Data Aps
+
+This file is part of the Zebra server.
+
+Zebra is free software; you can redistribute it and/or modify it under
+the terms of the GNU General Public License as published by the Free
+Software Foundation; either version 2, or (at your option) any later
+version.
+
+Zebra is distributed in the hope that it will be useful, but WITHOUT ANY
+WARRANTY; without even the implied warranty of MERCHANTABILITY or
+FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
+for more details.
+
+You should have received a copy of the GNU General Public License
+along with Zebra; see the file LICENSE.zebra. If not, write to the
+Free Software Foundation, 59 Temple Place - Suite 330, Boston, MA
+02111-1307, USA.
+*/
+
+
#include <stdio.h>
+#include <string.h>
#include <assert.h>
+#ifdef WIN32
+#include <io.h>
+#else
#include <unistd.h>
+#include <sys/time.h>
+#endif
+#include <time.h>
+#if HAVE_SYS_TIMES_H
+#include <sys/times.h>
+#endif
-#include <alexutil.h>
-#include <data1.h>
-#include "index.h"
+#include <yaz/data1.h>
+#include "zebraapi.h"
char *prog;
-size_t mem_max = 4*1024*1024;
-extern char *data1_tabpath;
int main (int argc, char **argv)
{
- int commit_at_end = 0;
int ret;
int cmd = 0;
char *arg;
- char *configName = NULL;
- int nsections;
- int key_open_flag = 0;
-
+ char *configName = 0;
+ int nsections = 0;
+ int disableCommit = 0;
+ size_t mem_max = 0;
+#if HAVE_SYS_TIMES_H
+ struct tms tms1, tms2;
+ struct timeval start_time, end_time;
+ long usec;
+#endif
+#ifndef WIN32
+ char nbuf[100];
+#endif
struct recordGroup rGroupDef;
-
+ ZebraService zs = 0;
+ ZebraHandle zh = 0;
+
+ nmem_init ();
+
+#ifdef WIN32
+#else
+ sprintf(nbuf, "%.40s(%d)", *argv, getpid());
+ yaz_log_init_prefix (nbuf);
+#endif
+#if HAVE_SYS_TIMES_H
+ times(&tms1);
+ gettimeofday(&start_time, 0);
+#endif
+
rGroupDef.groupName = NULL;
rGroupDef.databaseName = NULL;
rGroupDef.path = NULL;
rGroupDef.recordType = NULL;
rGroupDef.flagStoreData = -1;
rGroupDef.flagStoreKeys = -1;
+ rGroupDef.flagRw = 1;
+ rGroupDef.databaseNamePath = 0;
+ rGroupDef.explainDatabase = 0;
+ rGroupDef.fileVerboseLimit = 100000;
+ rGroupDef.followLinks = -1;
prog = *argv;
if (argc < 2)
{
- fprintf (stderr, "zebraidx [options] command <dir> ...\n"
+ fprintf (stderr, "%s [options] command <dir> ...\n"
"Commands:\n"
" update <dir> Update index with files below <dir>.\n"
" If <dir> is empty filenames are read from stdin.\n"
" delete <dir> Delete index with files below <dir>.\n"
+ " commit Commit changes\n"
+ " clean Clean shadow files\n"
"Options:\n"
" -t <type> Index files as <type> (grs or text).\n"
" -c <config> Read configuration file <config>.\n"
" -g <group> Index files according to group settings.\n"
" -d <database> Records belong to Z39.50 database <database>.\n"
" -m <mbytes> Use <mbytes> before flushing keys to disk.\n"
- " -v <level> Set logging to <level>.\n");
+ " -n Don't use shadow system.\n"
+ " -s Show analysis on stdout, but do no work.\n"
+ " -v <level> Set logging to <level>.\n"
+ " -l <file> Write log to <file>.\n"
+ " -L Don't follow symbolic links.\n"
+ " -f <n> Display information for the first <n> records.\n"
+ " -V Show version.\n", *argv
+ );
exit (1);
}
- while ((ret = options ("t:c:g:d:m:v:", argv, argc, &arg)) != -2)
+ while ((ret = options ("sVt:c:g:d:m:v:nf:l:L"
+ , argv, argc, &arg)) != -2)
{
if (ret == 0)
{
if(cmd == 0) /* command */
{
- if (!common_resource)
+ if (!zs)
{
- common_resource = res_open (configName ?
- configName : FNAME_CONFIG);
- if (!common_resource)
+ const char *config = configName ? configName : "zebra.cfg";
+ logf (LOG_LOG, "Zebra version %s %s",
+ ZEBRAVER, ZEBRADATE);
+ zs = zebra_start (config);
+ if (!zs)
{
- logf (LOG_FATAL, "Cannot open resource `%s'",
- configName);
+ yaz_log (LOG_FATAL, "Cannot read config %s", config);
+ exit (1);
+ }
+ zh = zebra_open (zs);
+ if (disableCommit)
+ zebra_shadow_enable (zh, 0);
+ }
+ if (rGroupDef.databaseName)
+ {
+ if (zebra_select_database (zh, rGroupDef.databaseName))
+ exit (1);
+ }
+ else
+ {
+ if (zebra_select_database (zh, "Default"))
exit (1);
- }
- data1_tabpath = res_get (common_resource, "profilePath");
- bf_cache (res_get (common_resource, "commitCache"));
}
+
if (!strcmp (arg, "update"))
cmd = 'u';
+ else if (!strcmp (arg, "update1"))
+ cmd = 'U';
+ else if (!strcmp (arg, "update2"))
+ cmd = 'm';
+ else if (!strcmp (arg, "dump"))
+ cmd = 's';
else if (!strcmp (arg, "del") || !strcmp(arg, "delete"))
cmd = 'd';
+ else if (!strcmp (arg, "init"))
+ {
+ zebra_init (zh);
+ }
else if (!strcmp (arg, "commit"))
- commit_at_end = 1;
+ {
+ zebra_commit (zh);
+ }
+ else if (!strcmp (arg, "clean"))
+ {
+ assert (!"todo");
+ }
+ else if (!strcmp (arg, "stat") || !strcmp (arg, "status"))
+ {
+ zebra_register_statistics (zh,0);
+ }
+ else if (!strcmp (arg, "dump") || !strcmp (arg, "dumpdict"))
+ {
+ zebra_register_statistics (zh,1);
+ }
+ else if (!strcmp (arg, "compact"))
+ {
+ zebra_compact (zh);
+ }
else
{
- logf (LOG_FATAL, "Unknown command: %s", arg);
+ logf (LOG_FATAL, "unknown command: %s", arg);
exit (1);
}
}
- else
+ else
{
- struct recordGroup rGroup;
+ rGroupDef.path = arg;
+ zebra_set_group (zh, &rGroupDef);
+ zebra_begin_trans (zh);
- memcpy (&rGroup, &rGroupDef, sizeof(rGroup));
- if (!key_open_flag)
+ switch (cmd)
{
- key_open (mem_max);
- key_open_flag = 1;
+ case 'u':
+ zebra_repository_update (zh);
+ break;
+ case 'd':
+ zebra_repository_delete (zh);
+ break;
+ case 's':
+ logf (LOG_LOG, "dumping %s", rGroupDef.path);
+ zebra_repository_show (zh);
+ nsections = 0;
+ break;
+ default:
+ nsections = 0;
}
- rGroup.path = arg;
- if (cmd == 'u')
- repositoryUpdate (&rGroup);
- else if (cmd == 'd')
- repositoryDelete (&rGroup);
cmd = 0;
+ zebra_end_trans (zh);
+ log_event_end (NULL, NULL);
}
}
- else if (ret == 'v')
+ else if (ret == 'V')
{
- log_init (log_mask_str(arg), prog, NULL);
+ fprintf (stderr, "Zebra %s %s\n", ZEBRAVER, ZEBRADATE);
+ fprintf (stderr, " (C) 1994-2002, Index Data ApS\n");
+#ifdef WIN32
+#ifdef _DEBUG
+ fprintf (stderr, " WIN32 Debug\n");
+#else
+ fprintf (stderr, " WIN32 Release\n");
+#endif
+#endif
+#if HAVE_BZLIB_H
+ fprintf (stderr, "libbzip2\n"
+ " (C) 1996-1999 Julian R Seward. All rights reserved.\n");
+#endif
}
+ else if (ret == 'v')
+ yaz_log_init_level (yaz_log_mask_str(arg));
+ else if (ret == 'l')
+ yaz_log_init_file (arg);
else if (ret == 'm')
- {
mem_max = 1024*1024*atoi(arg);
- }
else if (ret == 'd')
- {
rGroupDef.databaseName = arg;
- }
+ else if (ret == 's')
+ rGroupDef.flagRw = 0;
else if (ret == 'g')
- {
rGroupDef.groupName = arg;
- }
+ else if (ret == 'f')
+ rGroupDef.fileVerboseLimit = atoi(arg);
else if (ret == 'c')
configName = arg;
else if (ret == 't')
rGroupDef.recordType = arg;
+ else if (ret == 'n')
+ disableCommit = 1;
+ else if (ret == 'L')
+ rGroupDef.followLinks = 0;
else
- {
- logf (LOG_FATAL, "Unknown option '-%s'", arg);
- exit (1);
- }
- }
- if (key_open_flag)
- {
- nsections = key_close ();
- if (nsections)
- {
- logf (LOG_LOG, "Merging with index");
- key_input (FNAME_WORD_DICT, FNAME_WORD_ISAM, nsections, 60);
- }
- }
- if (commit_at_end)
- {
- logf (LOG_LOG, "commiting");
- bf_commit (res_get (common_resource, "commitCache"));
+ logf (LOG_WARN, "unknown option '-%s'", arg);
}
+ zebra_close (zh);
+ zebra_stop (zs);
+#if HAVE_SYS_TIMES_H
+ gettimeofday(&end_time, 0);
+ usec = (end_time.tv_sec - start_time.tv_sec) * 1000000L +
+ end_time.tv_usec - start_time.tv_usec;
+ times(&tms2);
+ yaz_log (LOG_LOG, "zebraidx times: %5.2f %5.2f %5.2f",
+ (double) usec / 1000000.0,
+ (double) (tms2.tms_utime - tms1.tms_utime)/100,
+ (double) (tms2.tms_stime - tms1.tms_stime)/100);
+#endif
exit (0);
+ return 0;
}