+ * Revision 1.62 1999-05-12 13:08:06 adam
+ * First version of ISAMS.
+ *
+ * Revision 1.61 1999/03/09 16:27:49 adam
+ * More work on SDRKit integration.
+ *
+ * Revision 1.60 1998/10/16 08:14:31 adam
+ * Updated record control system.
+ *
+ * Revision 1.59 1998/06/08 14:43:11 adam
+ * Added suport for EXPLAIN Proxy servers - added settings databasePath
+ * and explainDatabase to facilitate this. Increased maximum number
+ * of databases and attributes in one register.
+ *
+ * Revision 1.58 1998/05/20 10:12:16 adam
+ * Implemented automatic EXPLAIN database maintenance.
+ * Modified Zebra to work with ASN.1 compiled version of YAZ.
+ *
+ * Revision 1.57 1998/03/05 08:45:12 adam
+ * New result set model and modular ranking system. Moved towards
+ * descent server API. System information stored as "SGML" records.
+ *
+ * Revision 1.56 1998/01/12 15:04:08 adam
+ * The test option (-s) only uses read-lock (and not write lock).
+ *
+ * Revision 1.55 1997/10/27 14:33:04 adam
+ * Moved towards generic character mapping depending on "structure"
+ * field in abstract syntax file. Fixed a few memory leaks. Fixed
+ * bug with negative integers when doing searches with relational
+ * operators.
+ *
+ * Revision 1.54 1997/09/29 09:08:36 adam
+ * Revised locking system to be thread safe for the server.
+ *
+ * Revision 1.53 1997/09/25 14:54:43 adam
+ * WIN32 files lock support.
+ *
+ * Revision 1.52 1997/09/22 12:39:06 adam
+ * Added get_pos method for the ranked result sets.
+ *
+ * Revision 1.51 1997/09/18 08:59:19 adam
+ * Extra generic handle for the character mapping routines.
+ *
+ * Revision 1.50 1997/09/17 12:19:13 adam
+ * Zebra version corresponds to YAZ version 1.4.
+ * Changed Zebra server so that it doesn't depend on global common_resource.
+ *
+ * Revision 1.49 1997/09/05 15:30:08 adam
+ * Changed prototype for chr_map_input - added const.
+ * Added support for C++, headers uses extern "C" for public definitions.
+ *
+ * Revision 1.48 1997/02/12 20:39:45 adam
+ * Implemented options -f <n> that limits the log to the first <n>
+ * records.
+ * Changed some log messages also.
+ *
+ * Revision 1.47 1996/12/23 15:30:44 adam
+ * Work on truncation.
+ * Bug fix: result sets weren't deleted after server shut down.
+ *
+ * Revision 1.46 1996/11/08 11:10:19 adam
+ * Buffers used during file match got bigger.
+ * Compressed ISAM support everywhere.
+ * Bug fixes regarding masking characters in queries.
+ * Redesigned Regexp-2 queries.
+ *
+ * Revision 1.45 1996/10/29 14:09:42 adam
+ * Use of cisam system - enabled if setting isamc is 1.
+ *
+ * Revision 1.44 1996/06/06 12:08:40 quinn
+ * Added showRecord function
+ *
+ * Revision 1.43 1996/06/04 10:18:12 adam
+ * Search/scan uses character mapping module.
+ *
+ * Revision 1.42 1996/06/04 08:20:16 quinn
+ * Smallish
+ *
+ * Revision 1.41 1996/06/04 07:54:55 quinn
+ * Added output-map.
+ *
+ * Revision 1.40 1996/05/31 09:06:58 quinn
+ * Work on character-set handling
+ *
+ * Revision 1.39 1996/05/14 14:04:33 adam
+ * In zebraidx, the 'stat' command is improved. Statistics about ISAM/DICT
+ * is collected.
+ *
+ * Revision 1.38 1996/04/12 07:02:23 adam
+ * File update of single files.
+ *
+ * Revision 1.37 1996/03/26 16:01:13 adam
+ * New setting lockPath: directory of various lock files.
+ *
+ * Revision 1.36 1996/03/21 14:50:09 adam
+ * File update uses modify-time instead of change-time.
+ *
+ * Revision 1.35 1996/02/12 18:45:36 adam
+ * New fileVerboseFlag in record group control.
+ *
+ * Revision 1.34 1995/12/11 11:43:29 adam
+ * Locking based on fcntl instead of flock.
+ * Setting commitEnable removed. Command line option -n can be used to
+ * prevent commit if commit setting is defined in the configuration file.
+ *
+ * Revision 1.33 1995/12/08 16:22:53 adam
+ * Work on update while servers are running. Three lock files introduced.
+ * The servers reload their registers when necessary, but they don't
+ * reestablish result sets yet.
+ *
+ * Revision 1.32 1995/12/07 17:38:46 adam
+ * Work locking mechanisms for concurrent updates/commit.
+ *
+ * Revision 1.31 1995/12/06 12:41:22 adam
+ * New command 'stat' for the index program.
+ * Filenames can be read from stdin by specifying '-'.
+ * Bug fix/enhancement of the transformation from terms to regular
+ * expressons in the search engine.
+ *
+ * Revision 1.30 1995/12/05 11:25:02 adam
+ * Include of zebraver.h.
+ *
+ * Revision 1.29 1995/11/28 09:09:40 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 1.28 1995/11/27 13:58:53 adam
+ * New option -t. storeStore data implemented in server.
+ *
+ * Revision 1.27 1995/11/25 10:24:06 adam
+ * More record fields - they are enumerated now.
+ * New options: flagStoreData flagStoreKey.
+ *
+ * Revision 1.26 1995/11/22 17:19:17 adam
+ * Record management uses the bfile system.
+ *
+ * Revision 1.25 1995/11/21 15:29:12 adam
+ * Config file 'base' read by default by both indexer and server.
+ *
+ * Revision 1.24 1995/11/21 15:01:15 adam
+ * New general match criteria implemented.
+ * New feature: document groups.
+ *
+ * Revision 1.23 1995/11/20 16:59:45 adam