X-Git-Url: http://git.indexdata.com/?p=idzebra-moved-to-github.git;a=blobdiff_plain;f=index%2Frank1.c;h=f968e810c991133c717442f15262a2ccf5b778a3;hp=17f8616cd93a86437d900b784e881eacf93d2ce1;hb=896c0427df9d8eff5de6a1735dcd992e067df844;hpb=3c5f6226f97612c0d6ac40591f600587c5ffa858 diff --git a/index/rank1.c b/index/rank1.c index 17f8616..f968e81 100644 --- a/index/rank1.c +++ b/index/rank1.c @@ -1,23 +1,26 @@ -/* - * Copyright (C) 1998-1999, Index Data - * All rights reserved. - * Sebastian Hammer, Adam Dickmeiss - * - * $Log: rank1.c,v $ - * Revision 1.4 1999-02-02 14:51:01 adam - * Updated WIN32 code specific sections. Changed header. - * - * Revision 1.3 1998/06/12 12:21:53 adam - * Fixed memory-leak. - * - * Revision 1.2 1998/03/05 13:03:29 adam - * Improved ranking. - * - * Revision 1.1 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. - * - */ +/* $Id: rank1.c,v 1.10 2002-08-02 19:26:55 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 #include @@ -27,7 +30,7 @@ #include #endif -#include "zserver.h" +#include "index.h" struct rank_class_info { int dummy; @@ -38,6 +41,7 @@ struct rank_term_info { int global_occur; int global_inv; int rank_flag; + int rank_weight; }; struct rank_set_info { @@ -59,9 +63,10 @@ static int log2_int (unsigned g) * create: Creates/Initialises this rank handler. This routine is * called exactly once. The routine returns the class_handle. */ -static void *create (ZebraHandle zh) +static void *create (struct zebra_register *reg) { - struct rank_class_info *ci = xmalloc (sizeof(*ci)); + struct rank_class_info *ci = (struct rank_class_info *) + xmalloc (sizeof(*ci)); logf (LOG_DEBUG, "rank-1 create"); return ci; @@ -72,9 +77,9 @@ static void *create (ZebraHandle zh) * when the handler is no longer needed - i.e. when the server * dies. The class_handle was previously returned by create. */ -static void destroy (ZebraHandle zh, void *class_handle) +static void destroy (struct zebra_register *reg, void *class_handle) { - struct rank_class_info *ci = class_handle; + struct rank_class_info *ci = (struct rank_class_info *) class_handle; logf (LOG_DEBUG, "rank-1 destroy"); xfree (ci); @@ -86,21 +91,25 @@ static void destroy (ZebraHandle zh, void *class_handle) * each result set. The returned handle is a "set handle" and * will be used in each of the handlers below. */ -static void *begin (ZebraHandle zh, void *class_handle, RSET rset) +static void *begin (struct zebra_register *reg, void *class_handle, RSET rset) { - struct rank_set_info *si = xmalloc (sizeof(*si)); + struct rank_set_info *si = (struct rank_set_info *) xmalloc (sizeof(*si)); int i; logf (LOG_DEBUG, "rank-1 begin"); si->no_entries = rset->no_rset_terms; si->no_rank_entries = 0; - si->entries = xmalloc (sizeof(*si->entries)*si->no_entries); + si->entries = (struct rank_term_info *) + xmalloc (sizeof(*si->entries)*si->no_entries); for (i = 0; i < si->no_entries; i++) { int g = rset->rset_terms[i]->nn; - if (!strcmp (rset->rset_terms[i]->flags, "rank")) + if (!strncmp (rset->rset_terms[i]->flags, "rank,", 5)) { + yaz_log (LOG_LOG, "%s", rset->rset_terms[i]->flags); si->entries[i].rank_flag = 1; + si->entries[i].rank_weight = atoi (rset->rset_terms[i]->flags+5); + yaz_log (LOG_LOG, "i=%d weight=%d", i, si->entries[i].rank_weight); (si->no_rank_entries)++; } else @@ -117,9 +126,9 @@ static void *begin (ZebraHandle zh, void *class_handle, RSET rset) * end: Terminates ranking process. Called after a result set * has been ranked. */ -static void end (ZebraHandle zh, void *set_handle) +static void end (struct zebra_register *reg, void *set_handle) { - struct rank_set_info *si = set_handle; + struct rank_set_info *si = (struct rank_set_info *) set_handle; logf (LOG_DEBUG, "rank-1 end"); xfree (si->entries); xfree (si); @@ -132,8 +141,8 @@ static void end (ZebraHandle zh, void *set_handle) */ static void add (void *set_handle, int seqno, int term_index) { - struct rank_set_info *si = set_handle; - logf (LOG_DEBUG, "rank-1 add seqno=%d term_index=%d", seqno, term_index); + struct rank_set_info *si = (struct rank_set_info *) set_handle; + /*logf (LOG_DEBUG, "rank-1 add seqno=%d term_index=%d", seqno, term_index);*/ si->last_pos = seqno; si->entries[term_index].local_occur++; } @@ -147,18 +156,18 @@ static void add (void *set_handle, int seqno, int term_index) static int calc (void *set_handle, int sysno) { int i, lo, divisor, score = 0; - struct rank_set_info *si = set_handle; - - logf (LOG_DEBUG, "rank-1 calc sysno=%d", sysno); + struct rank_set_info *si = (struct rank_set_info *) set_handle; if (!si->no_rank_entries) return -1; + for (i = 0; i < si->no_entries; i++) if (si->entries[i].rank_flag && (lo = si->entries[i].local_occur)) - score += (8+log2_int (lo)) * si->entries[i].global_inv; - score *= 34; + score += (8+log2_int (lo)) * si->entries[i].global_inv * + si->entries[i].rank_weight; divisor = si->no_rank_entries * (8+log2_int (si->last_pos/si->no_entries)); score = score / divisor; + yaz_log (LOG_LOG, "sysno=%d score=%d", sysno, score); if (score > 1000) score = 1000; for (i = 0; i < si->no_entries; i++)