Update of /cvsroot/monetdb/MonetDB5/src/modules/kernel
In directory sc8-pr-cvs16.sourceforge.net:/tmp/cvs-serv29764/src/modules/kernel
Modified Files:
group.mx
Log Message:
added an optimized group for ordered data.
Index: group.mx
===================================================================
RCS file: /cvsroot/monetdb/MonetDB5/src/modules/kernel/group.mx,v
retrieving revision 1.106
retrieving revision 1.107
diff -u -d -r1.106 -r1.107
--- group.mx 8 Feb 2008 22:36:33 -0000 1.106
+++ group.mx 13 Feb 2008 09:10:17 -0000 1.107
@@ -244,7 +244,6 @@
grp_new(BAT *b, BAT *h)
{
if (h) {
- if (!(h->batDirty&2)) h = BATsetaccess(h, BAT_READ);
BATkey(h, TRUE);
h->tsorted = 0;
if ((h->hsorted = BAThordered(b)) & 1) {
@@ -255,6 +254,7 @@
BBPreclaim(h);
return GDK_FAIL;
}
+ if (!(h->batDirty&2)) h = BATsetaccess(h, BAT_READ);
BBPkeepref(h->batCacheid);
} else {
assert(h);
@@ -492,6 +492,7 @@
GDKfree(hash);
return NULL;
}
+
@c
static int
tailtype(BAT *b, int str_trick)
@@ -1244,6 +1245,76 @@
return MAL_SUCCEED;
}
[EMAIL PROTECTED] group_ordered
+BAT *
[EMAIL PROTECTED](BAT *b, BAT **res)
+{
+ BAT *bn = BATnew(TYPE_void, TYPE_oid, BATcount(b));
+ BAT *histo = BATnew(TYPE_oid, TYPE_int, 1024);
+ @1 *t = (@1*)Tloc(b, BUNfirst(b)), *e = t + BATcount(b), cur;
+ int cnt = 0;
+ oid *rt = (oid*)Tloc(bn, BUNfirst(bn)), curid;
+
+ if (bn == NULL || histo == NULL)
+ return NULL;
+
+ bn->hseqbase = b->hseqbase;
+ cur = *t;
+ if (b->htype == TYPE_void) {
+ oid h = b->hseqbase;
+
+ curid = h;
+ for(; t < e; t++, h++) {
+ if (*t != cur) {
+ BUNins(histo, &curid, &cnt, FALSE);
+ cur = *t;
+ curid = h;
+ cnt = 0;
+ }
+ cnt++;
+ *rt++ = curid;
+ }
+ } else { /* TYPE_oid */
+ oid *h = (oid*)Hloc(b, BUNfirst(b));
+
+ curid = *h;
+ for(; t < e; t++, h++) {
+ if (*t != cur) {
+ BUNins(histo, &curid, &cnt, FALSE);
+ cur = *t;
+ curid = *h;
+ cnt = 0;
+ }
+ cnt++;
+ *rt++ = curid;
+ }
+ }
+ BATsetcount(bn, BATcount(b));
+ if (BATcount(b))
+ BUNins(histo, &curid, &cnt, FALSE);
+ if (b->htype != bn->htype) {
+ BAT *r = VIEWcreate(b,bn);
+
+ BBPreleaseref(bn->batCacheid);
+ bn = r;
+ }
+ if (BAThordered(b)) {
+ bn->tsorted = b->tsorted;
+ histo->hsorted = b->hsorted;
+ histo->tsorted = b->tsorted;
+ BATkey(histo, TRUE);
+ }
+ if (b->hkey)
+ BATkey(bn, TRUE);
+ *res = bn;
+ return histo;
+}
[EMAIL PROTECTED]
+@:group_ordered(bte)@
+@:group_ordered(sht)@
+@:group_ordered(int)@
+@:group_ordered(lng)@
+
group_export str GRPgroup_custom(int *rethisto, int *retbid, int *bid, int *N,
int *rng);
group_export str GRPgroup(int *rethisto, int *retbid, int *bid);
str
@@ -1254,6 +1325,33 @@
if ((b = BATdescriptor(*bid)) == NULL) {
throw(MAL, "group.group", "Cannot access descriptor");
}
+ if (BATcount(b) > 1 &&
+ BATtordered(b)&1 &&
+ ATOMtype(b->htype) == TYPE_oid &&
+ b->ttype <= TYPE_str) {
+ switch(tailtype(b,TRUE)) {
+ case TYPE_bte:
+ histo = CTgroup_bte_ordered(b, &bn);
+ break;
+ case TYPE_sht:
+ histo = CTgroup_sht_ordered(b, &bn);
+ break;
+ case TYPE_int:
+ histo = CTgroup_int_ordered(b, &bn);
+ break;
+ case TYPE_lng:
+ histo = CTgroup_lng_ordered(b, &bn);
+ break;
+ default:
+ throw(MAL, "group.group", "Impossible type");
+ }
+ if (!histo)
+ throw(MAL, "group.group", "ordered group failed");
+ BBPkeepref(*rethisto = histo->batCacheid);
+ BBPkeepref(*retbid = bn->batCacheid);
+ BBPreleaseref(b->batCacheid);
+ return MAL_SUCCEED;
+ }
if (BATcount(b) > 1024*1024 &&
(ATOMstorage(b->ttype) == TYPE_int ||
ATOMstorage(b->ttype) == TYPE_lng)) {
-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Monetdb-checkins mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/monetdb-checkins
