X-Git-Url: http://git.indexdata.com/?a=blobdiff_plain;f=index%2Frank1.c;h=c79edeaf241fe74a2e05025e81977204ec616676;hb=0e56fa84bb4a5985c19a21926e86862c08d84689;hp=1d12085f9bc035748ced16e969522c83c9f8fb6a;hpb=f091b701a8326e533f4877e755ec561e109b47b9;p=idzebra-moved-to-github.git diff --git a/index/rank1.c b/index/rank1.c index 1d12085..c79edea 100644 --- a/index/rank1.c +++ b/index/rank1.c @@ -1,5 +1,5 @@ -/* $Id: rank1.c,v 1.11 2003-01-13 22:37:12 adam Exp $ - Copyright (C) 1995,1996,1997,1998,1999,2000,2001,2002 +/* $Id: rank1.c,v 1.21 2004-11-29 21:55:25 adam Exp $ + Copyright (C) 1995,1996,1997,1998,1999,2000,2001,2002,2003,2004 Index Data Aps This file is part of the Zebra server. @@ -20,8 +20,6 @@ Free Software Foundation, 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. */ - - #include #include #ifdef WIN32 @@ -30,18 +28,24 @@ Free Software Foundation, 59 Temple Place - Suite 330, Boston, MA #include #endif + #include "index.h" +static int log_level=0; +static int log_initialized=0; + struct rank_class_info { int dummy; }; struct rank_term_info { int local_occur; - int global_occur; + zint global_occur; int global_inv; int rank_flag; int rank_weight; + TERMID term; + int term_index; }; struct rank_set_info { @@ -49,6 +53,7 @@ struct rank_set_info { int no_entries; int no_rank_entries; struct rank_term_info *entries; + NMEM nmem; }; static int log2_int (unsigned g) @@ -63,12 +68,17 @@ 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 (struct zebra_register *reg) +static void *create (ZebraHandle zh) { - struct rank_class_info *ci = (struct rank_class_info *) - xmalloc (sizeof(*ci)); + struct rank_class_info *ci = + (struct rank_class_info *) xmalloc (sizeof(*ci)); - logf (LOG_DEBUG, "rank-1 create"); + if (!log_initialized) + { + log_level=yaz_log_module_level("rank1"); + log_initialized=1; + } + yaz_log (log_level, "rank-1 create"); return ci; } @@ -81,43 +91,57 @@ static void destroy (struct zebra_register *reg, void *class_handle) { struct rank_class_info *ci = (struct rank_class_info *) class_handle; - logf (LOG_DEBUG, "rank-1 destroy"); + yaz_log (log_level, "rank-1 destroy"); xfree (ci); } -/* +/** * begin: Prepares beginning of "real" ranking. Called once for * each result set. The returned handle is a "set handle" and * will be used in each of the handlers below. */ -static void *begin (struct zebra_register *reg, void *class_handle, RSET rset) +static void *begin (struct zebra_register *reg, + void *class_handle, RSET rset, NMEM nmem, + TERMID *terms, int numterms) { - struct rank_set_info *si = (struct rank_set_info *) xmalloc (sizeof(*si)); + struct rank_set_info *si = + (struct rank_set_info *) nmem_malloc (nmem,sizeof(*si)); int i; - logf (LOG_LOG, "rank-1 begin"); - si->no_entries = rset->no_rset_terms; + yaz_log (log_level, "rank-1 begin"); + si->no_entries = numterms; si->no_rank_entries = 0; + si->nmem=nmem; si->entries = (struct rank_term_info *) - xmalloc (sizeof(*si->entries)*si->no_entries); - for (i = 0; i < si->no_entries; i++) + nmem_malloc (si->nmem, sizeof(*si->entries)*numterms); + for (i = 0; i < numterms; i++) { - int g = rset->rset_terms[i]->nn; - yaz_log(LOG_LOG, "i=%d flags=%s", i, rset->rset_terms[i]->flags); - if (!strncmp (rset->rset_terms[i]->flags, "rank,", 5)) + zint g = rset_count(terms[i]->rset); + yaz_log(log_level, "i=%d flags=%s '%s'", i, + terms[i]->flags, terms[i]->name ); + if (!strncmp (terms[i]->flags, "rank,", 5)) { + const char *cp = strstr(terms[i]->flags+4, ",w="); si->entries[i].rank_flag = 1; - si->entries[i].rank_weight = atoi (rset->rset_terms[i]->flags+5); - yaz_log (LOG_LOG, " weight=%d", i, si->entries[i].rank_weight); + if (cp) + si->entries[i].rank_weight = atoi (cp+3); + else + si->entries[i].rank_weight = 34; + yaz_log (log_level, " i=%d weight=%d g="ZINT_FORMAT, i, + si->entries[i].rank_weight, g); (si->no_rank_entries)++; } else si->entries[i].rank_flag = 0; - si->entries[i].local_occur = 0; + si->entries[i].local_occur = 0; /* FIXME */ si->entries[i].global_occur = g; si->entries[i].global_inv = 32 - log2_int (g); - logf (LOG_DEBUG, "-------- %d ------", 32 - log2_int (g)); + yaz_log (log_level, " global_inv = %d g = " ZINT_FORMAT, + (int) (32-log2_int (g)), g); + si->entries[i].term=terms[i]; + si->entries[i].term_index=i; + terms[i]->rankpriv=&(si->entries[i]); } return si; } @@ -128,23 +152,32 @@ static void *begin (struct zebra_register *reg, void *class_handle, RSET rset) */ static void end (struct zebra_register *reg, void *set_handle) { - struct rank_set_info *si = (struct rank_set_info *) set_handle; - logf (LOG_DEBUG, "rank-1 end"); - xfree (si->entries); - xfree (si); + yaz_log (log_level, "rank-1 end"); + /* no need to free anything, they are in nmems */ } -/* + +/** * add: Called for each word occurence in a result set. This routine * should be as fast as possible. This routine should "incrementally" * update the score. */ -static void add (void *set_handle, int seqno, int term_index) +static void add (void *set_handle, int seqno, TERMID term) { struct rank_set_info *si = (struct rank_set_info *) set_handle; - yaz_log (LOG_LOG, "rank-1 add seqno=%d term_index=%d", seqno, term_index); + struct rank_term_info *ti; + assert(si); + if (!term) + { + yaz_log (log_level, "rank-1 add NULL term"); + return; + } + ti= (struct rank_term_info *) term->rankpriv; + assert(ti); si->last_pos = seqno; - si->entries[term_index].local_occur++; + ti->local_occur++; + yaz_log (log_level, "rank-1 add seqno=%d term=%s count=%d", + seqno, term->name,ti->local_occur); } /* @@ -153,7 +186,7 @@ static void add (void *set_handle, int seqno, int term_index) * score should be between 0 and 1000. If score cannot be obtained * -1 should be returned. */ -static int calc (void *set_handle, int sysno) +static int calc (void *set_handle, zint sysno) { int i, lo, divisor, score = 0; struct rank_set_info *si = (struct rank_set_info *) set_handle; @@ -161,10 +194,9 @@ static int calc (void *set_handle, int sysno) if (!si->no_rank_entries) return -1; - yaz_log(LOG_LOG, "calc"); for (i = 0; i < si->no_entries; i++) { - yaz_log(LOG_LOG, "i=%d rank_flag=%d lo=%d", + yaz_log(log_level, "calc: i=%d rank_flag=%d lo=%d", i, si->entries[i].rank_flag, si->entries[i].local_occur); if (si->entries[i].rank_flag && (lo = si->entries[i].local_occur)) score += (8+log2_int (lo)) * si->entries[i].global_inv * @@ -172,9 +204,10 @@ static int calc (void *set_handle, int sysno) } 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); + yaz_log (log_level, "calc sysno=" ZINT_FORMAT " score=%d", sysno, score); if (score > 1000) score = 1000; + /* reset the counts for the next term */ for (i = 0; i < si->no_entries; i++) si->entries[i].local_occur = 0; return score;