April 2010 - developers-list

Re: [Monetdb-developers] [Monetdb-sql-checkins] sql configure.ag, , 1.166, 1.167
by Milena Ivanova 14 Apr '10

14 Apr '10

Fabian, where should MERO_LOG be defined? Is it an env. variable? A after your last check in, I got /ufs/milena/src/sql/src/backends/monet5/merovingian/merovingian.c: In function ‘main’: /ufs/milena/src/sql/src/backends/monet5/merovingian/merovingian.c:452: error: ‘MERO_LOG’ undeclared (first use in this function) Milena On 14/04/10 11:54, Fabian wrote: > Update of /cvsroot/monetdb/sql > In directory sfp-cvsdas-4.v30.ch3.sourceforge.com:/tmp/cvs-serv6441 > > Modified Files: > configure.ag > Log Message: > Hardwire the default location for the logs in the sources, document them appropriately > > Index: configure.ag > =================================================================== > RCS file: /cvsroot/monetdb/sql/configure.ag,v > retrieving revision 1.166 > retrieving revision 1.167 > diff -u -d -r1.166 -r1.167 > --- configure.ag 13 Apr 2010 16:39:28 -0000 1.166 > +++ configure.ag 14 Apr 2010 09:54:16 -0000 1.167 > @@ -124,6 +124,8 @@ > AC_CHECK_HEADERS([uuid/uuid.h]) > AC_SEARCH_LIBS([uuid_generate], [uuid]) > AC_SUBST([MONETDB5_CONFFILE]) > +AC_SUBST([MERO_LOG], [$MONETDB5_LOGDIR/merovingian.log]) > +AC_DEFINE_UNQUOTED([MERO_LOG], ["$MONETDB5_LOGDIR/merovingian.log"], [where Merovingian will store its logs by default]) > > dnl CFLAGS for our code are stricter than what autoconf can cope with. > CFLAGS="$CFLAGS \$(X_CFLAGS)" > > > ------------------------------------------------------------------------------ > Download Intel® Parallel Studio Eval > Try the new software tools for yourself. Speed compiling, find bugs > proactively, and fine-tune applications for parallel performance. > See why Intel Parallel Studio got high marks during beta. > http://p.sf.net/sfu/intel-sw-dev > _______________________________________________ > Monetdb-sql-checkins mailing list > Monetdb-sql-checkins(a)lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/monetdb-sql-checkins >

1 1

Re: [Monetdb-developers] [Monetdb-checkins] clients/src/mapiclient MapiClient.mx, , 1.201, 1.202
by Stefan Manegold 14 Apr '10

14 Apr '10

On Tue, Apr 13, 2010 at 12:43:06PM +0000, Sjoerd Mullender wrote: > Update of /cvsroot/monetdb/clients/src/mapiclient > In directory sfp-cvsdas-4.v30.ch3.sourceforge.com:/tmp/cvs-serv15694 > > Modified Files: > MapiClient.mx > Log Message: > Once again print timer information to stderr. > If you want to combine timer information with the regular output, you > can do > mclient -t ... 2>&1 > This works since the standard output is flushed before the timer > information is printed. > And if you only want timer information, you can do > mclient -t ... >/dev/null > I briefly tried to do the same for break-down timings given by pathfinder (MonetDB/XQuery), but the changes (see diff below) don't help, mainly since the break-down timings are printed in MIL, and even fprintf(stderr(), ...) in MIL goes to stdout due to MonetDB/src/gdk/gdk.mx:2737:#define GDKerr GDKout which is probably required because MAPI only used one stream. I'm not sure whether it's worth to persue this fruther, e.g., by making mclient aware of the pathfinder/XQuery break-down timings and have re-direct them to (its) stderr ... Stefan ======== Index: pathfinder/compiler/include/mil.h =================================================================== RCS file: /cvsroot/monetdb/pathfinder/compiler/include/mil.h,v retrieving revision 1.89 diff -u -r1.89 mil.h --- pathfinder/compiler/include/mil.h 7 Jan 2010 15:24:27 -0000 1.89 +++ pathfinder/compiler/include/mil.h 14 Apr 2010 10:33:26 -0000 @@ -385,6 +385,7 @@ , m_print /**< MIL print() function */ , m_printf /**< MIL printf() function */ + , m_printerr /**< MIL fprintf(stderr(),...) function */ , m_col_name /**< assign BAT column name (for debugging only) */ @@ -1014,6 +1015,7 @@ PFmil_t * PFmil_print (const PFmil_t *); PFmil_t * PFmil_printf (const PFmil_t *); +PFmil_t * PFmil_printerr (const PFmil_t *); PFmil_t * PFmil_col_name (const PFmil_t *, const PFmil_t *); PFmil_t * PFmil_comment (const char *, ...) __attribute__ ((format (printf, 1, 2))); Index: pathfinder/compiler/compile.c =================================================================== RCS file: /cvsroot/monetdb/pathfinder/compiler/compile.c,v retrieving revision 1.186 diff -u -r1.186 compile.c --- pathfinder/compiler/compile.c 7 Jan 2010 15:24:27 -0000 1.186 +++ pathfinder/compiler/compile.c 14 Apr 2010 10:33:26 -0000 @@ -801,7 +801,7 @@ /* make runtime timing available */ if (status->timing) { mroot = PFmil_seq (mroot, - PFmil_printf ( + PFmil_printerr ( PFmil_arg (PFmil_lit_str ("\nShred % 10.3f msec" "\nQuery % 10.3f msec" "\nPrint % 10.3f msec\n"), @@ -1082,7 +1082,7 @@ if (!strncmp ("timing", genType, 6)) /* add timing information */ mroot = PFmil_seq (mroot, - PFmil_printf ( + PFmil_printerr ( PFmil_arg (PFmil_lit_str ("\nTrans % 10.3f msec" "\nShred % 10.3f msec" "\nQuery % 10.3f msec" Index: pathfinder/compiler/mil/mil.c =================================================================== RCS file: /cvsroot/monetdb/pathfinder/compiler/mil/mil.c,v retrieving revision 1.86 diff -u -r1.86 mil.c --- pathfinder/compiler/mil/mil.c 7 Jan 2010 15:24:28 -0000 1.86 +++ pathfinder/compiler/mil/mil.c 14 Apr 2010 10:33:26 -0000 @@ -1896,6 +1896,12 @@ } PFmil_t * +PFmil_printerr (const PFmil_t *args) +{ + return wire1 (m_printerr, args); +} + +PFmil_t * PFmil_col_name (const PFmil_t *bat, const PFmil_t *name) { return wire2 (m_col_name, bat, name); Index: pathfinder/compiler/mil/mil_dce.c =================================================================== RCS file: /cvsroot/monetdb/pathfinder/compiler/mil/mil_dce.c,v retrieving revision 1.34 diff -u -r1.34 mil_dce.c --- pathfinder/compiler/mil/mil_dce.c 7 Jan 2010 15:24:28 -0000 1.34 +++ pathfinder/compiler/mil/mil_dce.c 14 Apr 2010 10:33:27 -0000 @@ -474,6 +474,7 @@ case m_error: case m_print: case m_printf: + case m_printerr: case m_break: case m_use: case m_update_tape: Index: pathfinder/compiler/mil/milprint.c =================================================================== RCS file: /cvsroot/monetdb/pathfinder/compiler/mil/milprint.c,v retrieving revision 1.101 diff -u -r1.101 milprint.c --- pathfinder/compiler/mil/milprint.c 7 Jan 2010 15:24:28 -0000 1.101 +++ pathfinder/compiler/mil/milprint.c 14 Apr 2010 10:33:27 -0000 @@ -28,6 +28,7 @@ | 'ERROR (' expression ')' <m_error> | 'print (' args ')' <m_print> | 'printf (' args ')' <m_printf> + | 'printerr (' args ')' <m_printerr> | 'col_name (' expr ',' expr ')' <m_col_name> | 'destroy_ws (' expression ')' <m_destroy_ws> | 'end_ws (' expression ',' expression ')' <m_end_ws> @@ -673,6 +674,13 @@ milprintf (")"); break; + /* statement: 'fprintf ( stderr (), ' expression ')' */ + case m_printerr: + milprintf ("fprintf ( stderr (),"); + print_args (n->child[0]); + milprintf (")"); + break; + case m_col_name: print_expression (n->child[0]); milprintf (".%s (", ID[n->kind]); Index: pathfinder/compiler/mil/milprint_summer.c =================================================================== RCS file: /cvsroot/monetdb/pathfinder/compiler/mil/milprint_summer.c,v retrieving revision 1.438 diff -u -r1.438 milprint_summer.c --- pathfinder/compiler/mil/milprint_summer.c 12 Jan 2010 09:22:14 -0000 1.438 +++ pathfinder/compiler/mil/milprint_summer.c 14 Apr 2010 10:33:27 -0000 @@ -11616,7 +11616,7 @@ " ERROR(err); -(try);\n"\ "} else if (genType.startsWith(\"timing\")) {\n"\ " time_print := usec() - time_print;\n"\ - " printf(\"\\nTrans %% 10.3f msec\\nShred %% 10.3f msec\\nQuery %% 10.3f msec\\n" LASTPHASE " %% 10.3f msec\\n\","\ + " fprintf(stderr(),\"\\nTrans %% 10.3f msec\\nShred %% 10.3f msec\\nQuery %% 10.3f msec\\n" LASTPHASE " %% 10.3f msec\\n\","\ " dbl(time_compile)/1000.0, dbl(time_shred)/1000.0, dbl(time_exec - time_shred)/1000.0, time_print/1000.0);\n}" const char* PFstopMIL(int statement_type) { Index: pathfinder/runtime/pathfinder.mx =================================================================== RCS file: /cvsroot/monetdb/pathfinder/runtime/pathfinder.mx,v retrieving revision 1.480 diff -u -r1.480 pathfinder.mx --- pathfinder/runtime/pathfinder.mx 12 Apr 2010 09:20:51 -0000 1.480 +++ pathfinder/runtime/pathfinder.mx 14 Apr 2010 10:33:27 -0000 @@ -6336,7 +6336,7 @@ /* execute query */ xquery = p+1; if (p[1] == '4' && p[2] == '2' && p[3] == '\n' && p[4] == 0 && genType == buf && output[2] == 0) { /* use '42' as latency query test */ - stream_printf(GDKout, "42\n\nTrans 0.000 msec\nShred 0.000 msec\nQuery 0.000 msec\nPrint 0.000 msec\n"); + stream_printf(GDKerr, "42\n\nTrans 0.000 msec\nShred 0.000 msec\nQuery 0.000 msec\nPrint 0.000 msec\n"); } else { err = xquery_change_genType(ctx, genType); if (err == NULL) ======== > > Index: MapiClient.mx > =================================================================== > RCS file: /cvsroot/monetdb/clients/src/mapiclient/MapiClient.mx,v > retrieving revision 1.201 > retrieving revision 1.202 > diff -u -d -r1.201 -r1.202 > --- MapiClient.mx 29 Mar 2010 18:45:15 -0000 1.201 > +++ MapiClient.mx 13 Apr 2010 12:43:04 -0000 1.202 > @@ -332,10 +332,11 @@ > static void > timerEnd(void) > { > + stream_flush(toConsole); > t1 = gettime(); > if (mark && specials == NOmodifier) { > - stream_printf(toConsole, "%s %7ld.%03ld msec %s\n", mark, (long) ((t1 - t0) / 1000), (long) ((t1 - t0) % 1000), mark2 ? mark2 : ""); > - stream_flush(toConsole); > + fprintf(stderr, "%s %7ld.%03ld msec %s\n", mark, (long) ((t1 - t0) / 1000), (long) ((t1 - t0) % 1000), mark2 ? mark2 : ""); > + fflush(stderr); > } > } > > > > ------------------------------------------------------------------------------ > Download Intel® Parallel Studio Eval > Try the new software tools for yourself. Speed compiling, find bugs > proactively, and fine-tune applications for parallel performance. > See why Intel Parallel Studio got high marks during beta. > http://p.sf.net/sfu/intel-sw-dev > _______________________________________________ > Monetdb-checkins mailing list > Monetdb-checkins(a)lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/monetdb-checkins -- | Dr. Stefan Manegold | mailto:Stefan.Manegold@cwi.nl | | CWI, P.O.Box 94079 | http://www.cwi.nl/~manegold/ | | 1090 GB Amsterdam | Tel.: +31 (20) 592-4212 | | The Netherlands | Fax : +31 (20) 592-4199 |

2 1

[Monetdb-developers] select data type to store variable length string
by kss ram 12 Apr '10

12 Apr '10

Hello, Thank you everyone for being supportive! I have a brief query about the data types, according to the documentation http://monetdb.cwi.nl/SQL/Documentation/Data-Types.html, for storing string/char type date, we have following options 1.CHAR[ACTER] (L) - character string with length L 2.VARCHAR (L) | CHARACTER VARYING (L) - string with atmost length L CLOB | CHARACTER LARGE OBJECT BLOB | BINARY LARGE OBJECT DECIMAL(P,S) | NUMERIC(P,S) My question is would CLOB would be the right choice to store variable length/ dynamic string/char data, varchar(L) is out of the picture as i wont know the length of the string. Thank you for your time and help!

2 1

[Monetdb-developers] regarding debug output in query execution
by anshul gangwar 12 Apr '10

12 Apr '10

hello Is there a cluster(column consists of many clusters) based joining. If so than is there different algo used for each cluster also. What is the difference between (b=tmp_2607) and b=tmpr_2610 Is there any big differece between them. And any time real BAT ids of (id given at time of creation) columns are passed to BAT_fetchjoin i.e algos or the bat id passed are only of the clusters. In which file we can find the code where mil code(generated physical plan from pf) code given as input and algo finding process starts Send free SMS to your Friends on Mobile from your Yahoo! Messenger. Download Now! http://messenger.yahoo.com/download.php

2 1

[Monetdb-developers] Fw: regarding debug output in query execution
by anshul gangwar 12 Apr '10

12 Apr '10

Does anybody have any idea about our problem as asked before hello Is there a cluster(column consists of many clusters) based joining. If so than is there different algo used for each cluster also. What is the difference between (b=tmp_2607) and b=tmpr_2610 Is there any big differece between them. And any time real BAT ids of (id given at time of creation) columns are passed to BAT_fetchjoin i.e algos or the bat id passed are only of the clusters. In which file we can find the code where mil code(generated physical plan from pf) code given as input and algo finding process starts Send free SMS to your Friends on Mobile from your Yahoo! Messenger. Download Now! http://messenger.yahoo.com/download.php -----Inline Attachment Follows----- ------------------------------------------------------------------------------ Download Intel® Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta. http://p.sf.net/sfu/intel-sw-dev -----Inline Attachment Follows----- _______________________________________________ Monetdb-developers mailing list Monetdb-developers(a)lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/monetdb-developers Send free SMS to your Friends on Mobile from your Yahoo! Messenger. Download Now! http://messenger.yahoo.com/download.php

1 0

[Monetdb-developers] regarding monetdb data type
by Tomas Lopez 12 Apr '10

12 Apr '10

Hi, i installed MonetDB server and client. it works fine with console to create table and insert data. but if insert data larger than 512 byte, i can not get the whole data back to excel. can someone tell me what are the things that may be going wrong. thank u.

3 2

[Monetdb-developers] How to get BLOB data from ODBC?
by Xiao Lei 12 Apr '10

12 Apr '10

Hi guys, I tried to insert and get BLOB data from ODBC with Monet. But I always get the error message as, ERROR [07006] [MonetDB][ODBC Driver 1.0]Restricted data type attribute violation Can anyone tell me what's the problem? Thx. -- Best Regards, Xiao Lei

1 0

Re: [Monetdb-developers] [Monetdb-checkins] MonetDB/src/gdk gdk_bat.mx, , 1.241, 1.242
by Niels Nes 11 Apr '10

11 Apr '10

On Sun, Apr 11, 2010 at 11:47:41AM +0200, Sjoerd Mullender wrote: > Should this have been changed on the Feb2010 branch? not sure, the problem of performance problems and leaking hash files was reported on the Head. I'll check the code in stable. Niels > > On 2010-04-11 11:30, Niels Nes wrote: > > Update of /cvsroot/monetdb/MonetDB/src/gdk > > In directory sfp-cvsdas-4.v30.ch3.sourceforge.com:/tmp/cvs-serv4560 > > > > Modified Files: > > gdk_bat.mx > > Log Message: > > make sure we use the correct hash file (use views bat id for hhash/thash) > > > > > > Index: gdk_bat.mx > > =================================================================== > > RCS file: /cvsroot/monetdb/MonetDB/src/gdk/gdk_bat.mx,v > > retrieving revision 1.241 > > retrieving revision 1.242 > > diff -u -d -r1.241 -r1.242 > > --- gdk_bat.mx 22 Feb 2010 10:10:48 -0000 1.241 > > +++ gdk_bat.mx 11 Apr 2010 09:30:13 -0000 1.242 > > @@ -3355,7 +3355,7 @@ > > > > hp->filename = GDKmalloc(strlen(nme) + 12); > > if (hp->filename != NULL) > > - sprintf(hp->filename, "%s.%chash", nme, b->batCacheid > 0 ? 'h' : 't'); > > + sprintf(hp->filename, "%s.%chash", nme, bv->batCacheid > 0 ? 'h' : 't'); > > if (hp->filename == NULL || > > (bv->H->hash = HASHnew(hp, tpe, yy + lim, HASHmask(lim))) == NULL) { > > GDKfree(hp->filename); > > > > > > ------------------------------------------------------------------------------ > > Download Intel® Parallel Studio Eval > > Try the new software tools for yourself. Speed compiling, find bugs > > proactively, and fine-tune applications for parallel performance. > > See why Intel Parallel Studio got high marks during beta. > > http://p.sf.net/sfu/intel-sw-dev > > _______________________________________________ > > Monetdb-checkins mailing list > > Monetdb-checkins(a)lists.sourceforge.net > > https://lists.sourceforge.net/lists/listinfo/monetdb-checkins > > > -- > Sjoerd Mullender > > ------------------------------------------------------------------------------ > Download Intel® Parallel Studio Eval > Try the new software tools for yourself. Speed compiling, find bugs > proactively, and fine-tune applications for parallel performance. > See why Intel Parallel Studio got high marks during beta. > http://p.sf.net/sfu/intel-sw-dev > _______________________________________________ > Monetdb-checkins mailing list > Monetdb-checkins(a)lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/monetdb-checkins -- Niels Nes, Centrum Wiskunde & Informatica (CWI) Science Park 123, 1098 XG Amsterdam, The Netherlands room L3.14, phone ++31 20 592-4098 sip:4098@sip.cwi.nl url: http://www.cwi.nl/~niels e-mail: Niels.Nes(a)cwi.nl

1 0

Re: [Monetdb-developers] Compiler Mapi.h on windows
by Ken Kao 11 Apr '10

11 Apr '10

Sorry, I confuse it . Mapi.h only includes stdio.h. ( Mapi.c does indeed include more headers.) My version is latest Feb2010-SP1 release of MonetDB. The really problem for me is that I can not Compiler my code on windows after include Mapi.c . There are many headers include in Mapi.c. (ex: monet_utils.h, stream.h, stream_socket.h) I Find them in < MonetDB-1.36.3\src\common > . I try to include them in my code as well. Unfortunately, the first problem is that I have to extended to include unistd.h in stream.h. I have no ideal to sovle it because unistd.h is a linux header. My compiler enviroment is visual studio 2008 . Could you give some guides to create a correct compiler enviroment for Mapi.c ? I does not understand that How can I create a similar build enviroment for my code as MonetDB on Windows .Could you give some suggestion ? I really hope to be able to compile Mapi.c on Windows myself. I also try to use static link to libmapi.dll to achieve my purpose . But I need libmapi.lib can be successfully. If I only have mapi.h and libmapi.dll . Can I successfully complete such a way ? If No. Can I have any way to get libmapi.lib ? I really need it to achieve the second method. Thank you very mch!!

1 0

[Monetdb-developers] help regarding installtion after changes
by anshul gangwar 10 Apr '10

10 Apr '10

hello If the installation fails , how to remove the changes that installation has done. We are doing the changes to file gdk_relop.c and its .mx one and then using supertar ball we are trying to install it. Is it also possible that it doesn't install fully and still say installed successfully as earlier we have installed mserver separately from synaptic manager in ubuntu and it says succesfullly installed Send free SMS to your Friends on Mobile from your Yahoo! Messenger. Download Now! http://messenger.yahoo.com/download.php

3 2