- developers-list - monetdb.org

[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

[Monetdb-developers] RDF
by Bellekens, P.A.E. 18 Apr '08

18 Apr '08

Dear MonetDB developers, I have a question related to RDF. In you PDF online, I see that it is possible to import RDF in the MonetDB system, although there is not much further information concerning RDF input. Could you please inform me a bit more about: - is it possible to upload RDF data? - is there any kind of RDF specific querying? - is it possible to run MonetDB as a backend in a triple store like for example in Sesame? - is there any performance measuring done for RDF querying? It is known that column stores have a good performance (c-store benchmarks) for RDF data as well... With kind regards, Pieter

2 2

Re: [Monetdb-developers] [Monetdb-checkins] MonetDB5/src/modules/mal rdf.mx, , 1.2, 1.3
by Lefteris 17 Apr '08

17 Apr '08

On Thu, Apr 17, 2008 at 5:24 PM, Marco Antonelli <marcoantonelli(a)users.sourceforge.net> wrote: > Update of /cvsroot/monetdb/MonetDB5/src/modules/mal > In directory sc8-pr-cvs16.sourceforge.net:/tmp/cvs-serv26814/src/modules/mal > > Modified Files: > rdf.mx > Log Message: > Added a T bat that stores the types of the RDF objects (IRI, integer, double, plain literal...). It still has to be sorted together with the S, P, O bats in all their possible combinations > > U rdf.mx > Index: rdf.mx > =================================================================== > RCS file: /cvsroot/monetdb/MonetDB5/src/modules/mal/rdf.mx,v > retrieving revision 1.2 > retrieving revision 1.3 > diff -u -d -r1.2 -r1.3 > --- rdf.mx 1 Apr 2008 14:48:08 -0000 1.2 > +++ rdf.mx 17 Apr 2008 15:24:15 -0000 1.3 > +static rdf_type RDFparseXsdDataType(char *uri) > +{ > + rdf_type ret; > + if (uri == NULL) > + ret = PLAIN; > + else if ( strcmp(uri, "http://www.w3.org/2001/XMLSchema#int") || > + strcmp(uri, "http://www.w3.org/2001/XMLSchema#integer")) > + ret = INTEGER; > + else if (strcmp(uri, "http://www.w3.org/2001/XMLSchema#decimal")) > + ret = DECIMAL; > + else if (strcmp(uri, "http://www.w3.org/2001/XMLSchema#double")) > + ret = DOUBLE; > + else if (strcmp(uri, "http://www.w3.org/2001/XMLSchema#boolean")) > + ret = BOOLEAN; > + else if (strcmp(uri, "http://www.w3.org/2001/XMLSchema#float")) > + ret = FLOAT; > + else if (strcmp(uri, "http://www.w3.org/2001/XMLSchema#string")) > + ret = STRING; > + else if (strcmp(uri, "http://www.w3.org/2001/XMLSchema#dateTime")) > + ret = DATETIME; > + else > + ret = UNKNOWN; > + > + raptor_free_memory(uri); > + return ret; > +} I am not sure if this is the correct way to identify the type, what happens if 2001 changes to 1998 or 2003? or if they use prefix instead of the uri? Does raptor gives you always the full uri back? it seems way to much hard wired. > > > ------------------------------------------------------------------------- > 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-checkins mailing list > Monetdb-checkins(a)lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/monetdb-checkins >

2 3

[Monetdb-developers] MonetDB clients asks for MonetDB common 1.22.0
by Marco Antonelli 17 Apr '08

17 Apr '08

Hello, I'm building Monet on my laptop, but the clients module during configure stops saying: checking whether MonetDB Common version 1.22.0 or newer is installed... no: found only version 1.20.0 configure: error: Clients requires at least version 1.22.0 of MonetDB, but only version 1.20.0 was found. I checked out from cvs as told here on the sourceforce page: http://sourceforge.net/cvs/?group_id=56967 how may I fix it? Marco

2 2

Re: [Monetdb-developers] [Monetdb-pf-checkins] pathfinder/compiler/mil mil_dce.c, , 1.19, 1.20 milgen.brg, , 1.142, 1.143
by Jan Rittinger 17 Apr '08

17 Apr '08

The answer is simple: kind=165 is your new tape operator and its printing is missing in print_statement(). Your tape is no expression with a return value but a statement with side effects. Thus the MIL grammar in milprint.c expects the handling of your tape in print_statement() instead of print_expression() -- see also the grammar at the top of milprint.c (as the error message suggests :)). Jan On 04/17/2008 09:27 AM, Jan Flokstra wrote with possible deletions: > Hi Jan, > > When I remove the if_() around my tj_docmgmt_tape() function in milgen.brg I > get the following error message. [...] > > Jan. > > # End rule: "Rel: std_sort(Rel)" > a0001 := a0006; > a0004 := a0001.project (str(nil)); > a0002 := a0007; > a0003 := a0001.project (str(nil)); > DocmgmTape (ws, a0004, a0002, a0003, a0001); > fatal error: I encountered problems while generating MIL output. > fatal error: This is possibly due to an illegal MIL tree, not conforming to > the grammar in milprint.c > fatal error: This is how far I was able to generate the script: > notice: node: ALG_tj_docmgmt_tape > fatal error: Illegal MIL tree. MIL printer screwed up (kind: 165). > # halted in milprint.c (print_statement), line 596 -- Jan Rittinger Database Systems Technische Universität München (Germany) http://www-db.in.tum.de/~rittinge/

1 0

Re: [Monetdb-developers] [Monetdb-checkins] MonetDB/src/gdk gdk_scanselect.mx, , 1.50, 1.51
by Sjoerd Mullender 14 Apr '08

14 Apr '08

On 2008-04-14 21:08, Niels Nes wrote: > Update of /cvsroot/monetdb/MonetDB/src/gdk > In directory sc8-pr-cvs16.sourceforge.net:/tmp/cvs-serv3941 > > Modified Files: > gdk_scanselect.mx > Log Message: > don't typemap dbl/flt to lng/int in scanselect bunfastins > (this is an attempt to fix the compilation problems with gcc 4.3 and up > which have very strict c99 aliasing rules ) > > > U gdk_scanselect.mx > Index: gdk_scanselect.mx > =================================================================== > RCS file: /cvsroot/monetdb/MonetDB/src/gdk/gdk_scanselect.mx,v > retrieving revision 1.50 > retrieving revision 1.51 > diff -u -d -r1.50 -r1.51 > --- gdk_scanselect.mx 6 Mar 2008 08:25:38 -0000 1.50 > +++ gdk_scanselect.mx 14 Apr 2008 19:08:34 -0000 1.51 > @@ -271,6 +271,8 @@ > @:T_BUNfastins(@1,@2,@3,sht,SIMPLE,sht)@ > @:T_BUNfastins(@1,@2,@3,int,SIMPLE,int)@ > @:T_BUNfastins(@1,@2,@3,lng,SIMPLE,lng)@ > + @:T_BUNfastins(@1,@2,@3,flt,SIMPLE,flt)@ > + @:T_BUNfastins(@1,@2,@3,dbl,SIMPLE,dbl)@ > @:T_BUNfastins(@1,@2,@3,fix,FIXEDATOM,(b)->ttype)@ > @:T_BUNfastins(@1,@2,@3,var,VARATOM,(b)->ttype)@ > @- > @@ -282,6 +284,8 @@ > @:H_BUNfastins(sht,SIMPLE,sht)@ > @:H_BUNfastins(int,SIMPLE,int)@ > @:H_BUNfastins(lng,SIMPLE,lng)@ > +@:H_BUNfastins(flt,SIMPLE,flt)@ > +@:H_BUNfastins(dbl,SIMPLE,dbl)@ > @:H_BUNfastins(vid,VID,vid)@ > @:H_BUNfastins(fix,FIXEDATOM,(b)->htype)@ > @:H_BUNfastins(var,VARATOM,(b)->htype)@ > @@ -296,21 +300,6 @@ > #define @2@1_bunfastins_nocheck(b, p, h, t, hs, ts) @3@1_bunfastins_nocheck(b, p, h, t, hs, ts) > #define @2@1_bunfastins(b, h, t) @3@1_bunfastins(b, h, t) > @h > -@:HT_bunfastins_typemap(void,flt,int)@ > -@:HT_bunfastins_typemap(chr,flt,int)@ > -@:HT_bunfastins_typemap(sht,flt,int)@ > -@:HT_bunfastins_typemap(int,flt,int)@ > -@:HT_bunfastins_typemap(lng,flt,int)@ > -@:HT_bunfastins_typemap(fix,flt,int)@ > -@:HT_bunfastins_typemap(var,flt,int)@ > -@:HT_bunfastins_typemap(void,dbl,lng)@ > -@:HT_bunfastins_typemap(chr,dbl,lng)@ > -@:HT_bunfastins_typemap(sht,dbl,lng)@ > -@:HT_bunfastins_typemap(int,dbl,lng)@ > -@:HT_bunfastins_typemap(lng,dbl,lng)@ > -@:HT_bunfastins_typemap(fix,dbl,lng)@ > -@:HT_bunfastins_typemap(var,dbl,lng)@ > - > #if SIZEOF_OID == SIZEOF_INT > @:HT_bunfastins_typemap(void,oid,int)@ > @:HT_bunfastins_typemap(chr,oid,int)@ > > > ------------------------------------------------------------------------- > 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-checkins mailing list > Monetdb-checkins(a)lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/monetdb-checkins I came up with the same fix today which I hadn't checked in yet. But I went one further: I also did oid and wrd. They may well cause the same warning/error. -- Sjoerd Mullender

1 0

Re: [Monetdb-developers] [Monetdb-pf-checkins] pathfinder/compiler/algebra builtins.c, , 1.84, 1.85 logical.c, , 1.93, 1.94
by Jan Rittinger 10 Apr '08

10 Apr '08

Hi Jan, your code looks quite good. Picky as I am I however have some comments :) Jan On 04/10/2008 03:38 PM, Jan Flokstra wrote with possible deletions: > [...] > U builtins.c > Index: builtins.c > =================================================================== > RCS file: /cvsroot/monetdb/pathfinder/compiler/algebra/builtins.c,v > retrieving revision 1.84 > retrieving revision 1.85 > diff -u -d -r1.84 -r1.85 > --- builtins.c 3 Apr 2008 09:42:11 -0000 1.84 > +++ builtins.c 10 Apr 2008 13:38:38 -0000 1.85 > @@ -44,6 +44,7 @@ > #include <assert.h> > #include <stdio.h> > > +#include "mem.h" ^^^^^ what do you need that for? > + > +/* > + * PFTIJAH defines. I decided not make a seperate include file for > + * pftijah so ensure that these defines are exactly the same as in > + * ../mil/milgen.brg > + */ that is ugly in my eyes! -- You could place your stuff in algebra.h like e.g., the axis steps instead. > + > +#define MYNODEKIND aat_pnode > +#define DOCMGMTTYPE aat_docmgmt > + > +#define PFT_FUN(F) (strncmp(F,"pftijah_",8)==0) > + > +#define PFT_QUERY_N_XX "pftijah_query_n_xx" > +#define PFT_QUERY_N_SX "pftijah_query_n_sx" > +#define PFT_QUERY_N_XO "pftijah_query_n_xo" > +#define PFT_QUERY_N_SO "pftijah_query_n_so" > +#define PFT_QUERY_I_XX "pftijah_query_i_xx" > +#define PFT_QUERY_I_SX "pftijah_query_i_sx" > +#define PFT_QUERY_I_XO "pftijah_query_i_xo" > +#define PFT_QUERY_I_SO "pftijah_query_i_so" > + > +#define PTF_QUERY_NODES(N) (N[14]=='n') > +#define PTF_QUERY_STARTNODES(N) (N[16]=='s') > +#define PTF_QUERY_OPTIONS(N) (N[17]=='o') Encoding all information in a single string is even more ugly!!! You have a context (ctx) where you can store arbitrary information. Just build your own struct (in algebra.h) and make it less error-prone. > + > +PFalg_schema_t pft_empty_schema(PFalg_simple_type_t item_t) { > + PFalg_schema_t schema; > + schema.count = 3; > + // INCOMPLETE, what about freeing this space ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ we use garbage collection. > +struct PFla_pair_t pft_query_param2(struct PFla_pair_t *p1, PFalg_simple_type_t itemType1, struct PFla_pair_t *p2, PFalg_simple_type_t itemType2) { > [...] > +struct PFla_pair_t pft_query_param3(struct PFla_pair_t *p1, PFalg_simple_type_t itemType1, struct PFla_pair_t *p2, PFalg_simple_type_t itemType2, struct PFla_pair_t *p3, PFalg_simple_type_t itemType3) { textwidth > 80 -- Jan Rittinger Database Systems Technische Universität München (Germany) http://www-db.in.tum.de/~rittinge/

1 0