May 2008 - developers-list

Re: [Monetdb-developers] [Monetdb-pf-checkins] pathfinder/modules/pftijah serialize_pftijah.mx, , 1.60, 1.61
by Ying Zhang 15 May '08

15 May '08

Hi Jan, Your checkin breaks the compilation due to the following error: $SRCBASE_DIR/pathfinder/modules/pftijah/serialize_pftijah.mx: In function 'dbat_extend': $SRCBASE_DIR/pathfinder/modules/pftijah/serialize_pftijah.mx:175: error: too few arguments to function 'BATmmap' BATmmap has recently got an extra argument (please see Niels' checkin yesterday), so you might want to update your MonetDB modules firstly :) Regards, Jennie On May 15, 2008, at 11:59 , Jan Flokstra wrote: > Update of /cvsroot/monetdb/pathfinder/modules/pftijah > In directory sc8-pr-cvs16.sourceforge.net:/tmp/cvs-serv12778 > > Modified Files: > serialize_pftijah.mx > Log Message: > + repair pftijah direct memory acces problem which caused Henning's > vitalis > collection problems. > > > > U serialize_pftijah.mx > Index: serialize_pftijah.mx > =================================================================== > RCS file: /cvsroot/monetdb/pathfinder/modules/pftijah/ > serialize_pftijah.mx,v > retrieving revision 1.60 > retrieving revision 1.61 > diff -u -d -r1.60 -r1.61 > --- serialize_pftijah.mx 17 Apr 2008 10:29:10 -0000 1.60 > +++ serialize_pftijah.mx 15 May 2008 09:59:43 -0000 1.61 > @@ -135,9 +135,22 @@ > #define MINCHUNK 8192 > #define MAXCHUNK 67108864 > > -int dbat_extend(dbat* dbat, oid min_i /*raw-index*/, size_t > forced_size) { > +static void > +dbat_setcount(dbat *dbat, oid n) > +{ > + BAT *b = dbat->bat; > + BATsetcount(b, n); > + if (b->ttype) { > + b->tsorted = 0; > + b->tdense = 0; > + } > +} > + > +static int > +dbat_extend(dbat* dbat, oid min_i /*raw-index*/, size_t > forced_size) { > size_t newsize; > > + dbat_setcount(dbat, dbat->raw_max); > if ( forced_size ) { > newsize = forced_size; > } else { > @@ -156,10 +169,19 @@ > return -1; > } > } > +#if 1 > + /* copied this form shredder.mx, keep it here ??? */ > + if (newsize+newsize > (size_t) REMAP_PAGE_MAXSIZE) { /* try to > use mmap() */ > + BATmmap(dbat->bat, STORE_MMAP, STORE_MMAP, STORE_MMAP); > + } > + > +#endif > + > #if 0 > stream_printf(GDKout,"dbat_extend[%s](%d -> %d)\n",dbat- > >name,dbat->raw_max,newsize); > #endif > - dbat->raw_max= newsize; > + > + dbat->raw_max = newsize; > dbat->seq_max = dbat->raw_max + dbat->seqbase; > if ( !(dbat->bat = BATextend(dbat->bat,newsize)) ) { > GDKerror("dbat_extend: BATextend[\"%s\"](to %d) fails > \n","incomplete",newsize); > @@ -169,17 +191,30 @@ > return 1; > } > > -int dbat_sizeHint(dbat* dbat, int sizeHint_mark) { > +#if 0 > +static int > +dbat_sizeHint(dbat* dbat, int sizeHint_mark) { > int sizeHint = sizeHint_mark - dbat->seqbase; > int estimate = dbat->raw_max + sizeHint; > > return dbat_extend(dbat, estimate, 0); > } > +#endif > > #define dbat_set_oid(DBAT,I,V) (DBAT)->cast.oidCAST[I-(DBAT)- > >seqbase] = V > > #define dbat_set_int(DBAT,I,V) (DBAT)->cast.intCAST[I-(DBAT)- > >seqbase] = V > > +#if 0 > +static void > +dbat_int_debug(char* t, dbat* dbat) { > + stream_printf(GDKout,"+ DEBUGGING INT DBAT[sz=%d]@=[%s]:\n",dbat- > >seq_max,t); > + for(int i=0; i<10; i++) { > + stream_printf(GDKout,"+ dbat[%d] = %d\n",i,dbat- > >cast.intCAST[i]); > + } > +} > +#endif > + > /************************************************ > * > * The Tijah-Pathfinder index creation context > @@ -391,7 +426,7 @@ > if ( (tjPre = tj_newPre(tjctx) ) == oid_nil ) > return 0; > dbat_set_oid(&tjctx->dbat_collPre, tjPre, termOid); > - dbat_set_int(&tjctx->dbat_collSize,(int)tjPre,0); > + dbat_set_int(&tjctx->dbat_collSize,tjPre,0); > #ifdef TJ_TRACE > if ( TJ_TRACE ) stream_printf(GDKout,"C[%s]:TERM: \"%s > \", termoid=%d, tjPre=%d\n",tjctx->name,term,termOid,tjPre); > #endif > @@ -1230,7 +1265,7 @@ > return GDK_FAIL; > if ( dbat_finalize(&tjctx->dbat_collSize, tjctx->tijahPre) < 0 ) > return GDK_FAIL; > - > + > tjctx->b_collPre->batDirty = TRUE; > tjctx->b_collSize->batDirty = TRUE; > tjctx->b_collPfPre->batDirty = TRUE; > > > ------------------------------------------------------------------------- > 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-pf-checkins mailing list > Monetdb-pf-checkins(a)lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/monetdb-pf-checkins

1 0

Re: [Monetdb-developers] php5 interface
by Martin Kersten 09 May '08

09 May '08

Michiel Kamermans wrote: > Martin, > > A test using windows XP with PHP 5.2.6 and MonetDB seems to work just > fine. I suspect that this means that in Vista monetDB won't be able to > interface with PHP either, as the core is essentially the same as the > one used in windows server 2008... I am not a windows expert myself > when it comes to figuring out API differences, so I wouldn't know > where to look... but I do have one improvement to the php file in the > /share/php dir: > > Right now it auto-assumes running on *nix, automatically calling > dl('monetdb.so'). A change for this line to the following would make > the script usable on both *nix and windows: > > $ext="monetdb"; > if (!extension_loaded($ext)) { > $prefix = (PHP_SHLIB_SUFFIX === 'dll') ? 'php_' : ''; > dl($prefix . $ext . '.' . PHP_SHLIB_SUFFIX); } > > Regards, Thank you very much for the information provided, we will take this suggestion into consideration > > Mike Kamermans > (who will now no doubt spend a nice bit of time in XP doing > benchmarking =)

1 0

[Monetdb-developers] [Fwd: Re: php5 interface]
by Martin Kersten 09 May '08

09 May '08

-------- Original Message -------- Subject: Re: php5 interface Date: Fri, 09 May 2008 19:13:40 +0200 From: Michiel Kamermans <mkamerma(a)science.uva.nl> To: Martin Kersten <Martin.Kersten(a)cwi.nl> References: <48248084.5060306(a)science.uva.nl> <4824835A.10101(a)cwi.nl> Hi Martin, I've tested this on windows server 2008 (which is basically vista, without all the vista bits stripped off), with the MonetDB installed using "MonetDB5-SQL-Installer-i386-20080214.msi". While technically the setup also involves Apache 2.2.8, the fact that PHP as standalone engine will crash means it doesn't really enter into the picture. I will see if I have the same problems in windows XP with PHP 5.2.6 and report back on the issue. Regards, -- Michiel Kamermans ,-------------------------------------------------------------- | Masters student AI at the University of Amsterdam [0024538] | http://www.nihongoresources.com `--------------------------------------------------------------

1 0

[Monetdb-developers] Reporting a bug
by Nadeem Iftikhar 07 May '08

07 May '08

Hi I tried the following link to report the bug, but it does not work http://monetdb.wiki.sourceforge.net/ Bug..... I need to insert some null values into a table via a procedure. I created the table and did insert two rows. One with (insert into) command and second with (insert into ..select) command. Both commands work well and I could see two rows added in result. When I used the same (insert into.. select) command in a procedure. It behaves in a strange fashion. It shows that a row has been affected, but does not show the newly added row. If the rows are counted, it gives the correct count which is (3) in this case but still I could not see the last row which is added via a procedure. ''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''' create sequence "ff" as integer start with 1; CREATE TABLE t1(timeid INT DEFAULT NEXT VALUE FOR "ff" PRIMARY KEY, a INT, b INT, c INT, d INT, e INT, f VARCHAR(20), g VARCHAR(20), h INT, i VARCHAR(10), j VARCHAR(10), k VARCHAR(10), l INT, m INT, n VARCHAR(10), o VARCHAR(10), p VARCHAR(10), q VARCHAR(15), r INT ); ''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''' And then insert a row ''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''' insert into t1(a, b, c, d, e, f, g, h, i, j, k, l, m, n, o, p, q, r) values(1, 1, 1, 1, 1, 'first row', 'first row',1, 'first row', 'first row', 'first row',1,1, 'first row', 'first row', 'first row', 'first row', 1); select * from t1;sql>sql>sql>sql>more>more>more>more>more>more>more>more>more>more>more>more>more>more>more>more>more>more>more>sql> sql>Rows affected 1 sql> +------+--+--+--+--+--+-----+-----+--+-----+------+------+--+--+------+------+------+------+--+ | time |a |b |c |d |e |f |g |h |i |j |k |l |m |n |o |p |q |r | : id | | | | | | | | | | | | | | | | | | | +======+==+==+==+==+==+=====+=====+==+=====+======+======+==+==+======+======+======+======+==+ | 1 |1 |1 |1 |1 |1 |firs |firs |1 |firs |first |first |1 |1 |first |first |first |first |1 | : | | | | | |t |t | |t |row |row | | |row |row |row |row | | : | | | | | |row |row | |row | | | | | | | | | | +------+--+--+--+--+--+-----+-----+--+-----+------+------+--+--+------+------+------+------+--+ '''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''' Every thing is fine so far '''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''' sql>insert into t1(a, b, c, d, e, f, g, h) select a, b, c, d, e, f, g, h from t1 where timeid = 1; Rows affected 1 sql>select * from t1; +------+--+--+--+--+--+-----+-----+--+-----+------+------+--+--+------+------+------+------+--+ | time |a |b |c |d |e |f |g |h |i |j |k |l |m |n |o |p |q |r | : id | | | | | | | | | | | | | | | | | | | +======+==+==+==+==+==+=====+=====+==+=====+======+======+==+==+======+======+======+======+==+ | 1 |1 |1 |1 |1 |1 |firs |firs |1 |firs |first |first |1 |1 |first |first |first |first |1 | : | | | | | |t |t | |t |row |row | | |row |row |row |row | | : | | | | | |row |row | |row | | | | | | | | | | | 2 |1 |1 |1 |1 |1 |firs |firs |1 |null |null |null |n |n |null |null |null |null |n | : | | | | | |t |t | | | | |u |u | | | | |u | : | | | | | |row |row | | | | |l |l | | | | |l | : | | | | | | | | | | | |l |l | | | | |l | +------+--+--+--+--+--+-----+-----+--+-----+------+------+--+--+------+------+------+------+--+ ''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''' inserted another row and it works fine '''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''' sql>create procedure tt() begin insert into t1(a, b, c, d, e, f, g, h) select a, b, c, d, e, f, g, h from t1 where timeid = 1; end; call tt(); select * from t1; more>more>more> sql>Rows affected 1 sql>+------+--+--+--+--+--+-----+-----+--+-----+------+------+--+--+------+------+------+------+--+ | time |a |b |c |d |e |f |g |h |i |j |k |l |m |n |o |p |q |r | : id | | | | | | | | | | | | | | | | | | | +======+==+==+==+==+==+=====+=====+==+=====+======+======+==+==+======+======+======+======+==+ | 1 |1 |1 |1 |1 |1 |firs |firs |1 |firs |first |first |1 |1 |first |first |first |first |1 | : | | | | | |t |t | |t |row |row | | |row |row |row |row | | : | | | | | |row |row | |row | | | | | | | | | | | 2 |1 |1 |1 |1 |1 |firs |firs |1 |null |null |null |n |n |null |null |null |null |n | : | | | | | |t |t | | | | |u |u | | | | |u | : | | | | | |row |row | | | | |l |l | | | | |l | : | | | | | | | | | | | |l |l | | | | |l | +------+--+--+--+--+--+-----+-----+--+-----+------+------+--+--+------+------+------+------+--+ ''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''' The problem starts now: when I try to insert the row via a procedure it shows that row effected 1 but does not show the row. ''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''' sql>select count(*) from t1; +--------+ | count_ | +========+ | 3 | +--------+ sql>select timeid from t1; +--------+ | timeid | +========+ | 1 | | 2 | | 3 | +--------+ sql>select * from t1; +------+--+--+--+--+--+-----+-----+--+-----+------+------+--+--+------+------+------+------+--+ | time |a |b |c |d |e |f |g |h |i |j |k |l |m |n |o |p |q |r | : id | | | | | | | | | | | | | | | | | | | +======+==+==+==+==+==+=====+=====+==+=====+======+======+==+==+======+======+======+======+==+ | 1 |1 |1 |1 |1 |1 |firs |firs |1 |firs |first |first |1 |1 |first |first |first |first |1 | : | | | | | |t |t | |t |row |row | | |row |row |row |row | | : | | | | | |row |row | |row | | | | | | | | | | | 2 |1 |1 |1 |1 |1 |firs |firs |1 |null |null |null |n |n |null |null |null |null |n | : | | | | | |t |t | | | | |u |u | | | | |u | : | | | | | |row |row | | | | |l |l | | | | |l | : | | | | | | | | | | | |l |l | | | | |l | +------+--+--+--+--+--+-----+-----+--+-----+------+------+--+--+------+------+------+------+--+ sql> ''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''' I could not see the 3rd row *********************************** Regards, Nadeem Iftikhar, Ph.D. Fellow Department of Computer Science, Aalborg University Selma Lagerlöfs Vej 300, DK-9220 Aalborg Øst, Denmark Phone: (+45) 9940 9830, fax: (+45) 9940 9798 E-mail: nadeem(a)cs.aau.dk; nadeem(a)ieee.org Home Page: http://www.cs.aau.dk/~nadeem ---------------------------------------------------------------- This message was sent using IMP, the Internet Messaging Program.

2 1

Re: [Monetdb-developers] [Monetdb-pf-checkins] pathfinder/runtime pathfinder.mx, , 1.411, 1.412
by Lefteris 07 May '08

07 May '08

That was drastic:) On Tue, May 6, 2008 at 7:20 PM, Jan Rittinger <tsheyar(a)users.sourceforge.net> wrote: > Update of /cvsroot/monetdb/pathfinder/runtime > In directory sc8-pr-cvs16.sourceforge.net:/tmp/cvs-serv14794 > > Modified Files: > pathfinder.mx > Log Message: > -- Check whether Lefteris changes to ws_create() are complete :) > > > U pathfinder.mx > Index: pathfinder.mx > =================================================================== > RCS file: /cvsroot/monetdb/pathfinder/runtime/pathfinder.mx,v > retrieving revision 1.411 > retrieving revision 1.412 > diff -u -d -r1.411 -r1.412 > --- pathfinder.mx 6 May 2008 11:36:10 -0000 1.411 > +++ pathfinder.mx 6 May 2008 16:20:00 -0000 1.412 > @@ -1348,7 +1348,6 @@ > return bat(void,tpe).seqbase(PRE_BASE); # a constant bat > } > > -PROC ws_create() : BAT[void,bat] { return ws_create(0); } # still exists for algebra > PROC ws_create(int update) : BAT[void,bat] > { > # wsid = 1bit(0) | 31bits(unique-id) | 2bits(update) | 30bits(batid) > > > ------------------------------------------------------------------------- > This SF.net email is sponsored by the 2008 JavaOne(SM) Conference > Don't miss this year's exciting event. There's still time to save $100. > Use priority code J8TL2D2. > http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/java… > _______________________________________________ > Monetdb-pf-checkins mailing list > Monetdb-pf-checkins(a)lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/monetdb-pf-checkins >

2 2

[Monetdb-developers] check-in "killed" pf/tijah
by Henning Rode 07 May '08

07 May '08

hej, one of the check-ins yesterday more or less "killed" pf/tijah. the test-set is now completely red for our module. since i also checked in one line of code yesterday, i tested already whether it was the cause of the trouble, but undoing my check-in does not solve anything. so, i suppose there must be some other reason. -henning

3 5

[Monetdb-developers] How to debug MonetDB project on Windows OS?
by 廉小伟 06 May '08

06 May '08

Hi.. I want to know about the method of debugging MonetDB on Windows OS. After compiering MonetDB use "namke" , how can I debug it. Is special debug tool needed?

1 0