Bugs item #1841727, was opened at 2007-11-30 14:14 Message generated for change (Comment added) made by nielsnes You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=482468&aid=1841727&group_id=56967
Please note that this message will contain a full copy of the comment thread, including the initial issue submission, for this request, not just the latest update. Category: SQL/Core Group: SQL CVS Head >Status: Closed >Resolution: Works For Me Priority: 5 Private: No Submitted By: Romulo Goncalves (romulog) Assigned to: Niels Nes (nielsnes) Summary: pkey in temp table Initial Comment: Dear developers if I run the following query against M5-SQL server: >CREATE TEMP TABLE PKTABLE (ptest1 int PRIMARY KEY); Program received signal SIGSEGV, Segmentation fault. [Switching to Thread 46912517404944 (LWP 10828)] 0x00002aaabc428bbf in sql_trans_create_kc (tr=0xb5c638, k=0x0, c=0x13edde8) at /ufs/goncalve/MonetDB/sql/src/storage/store.mx:3580 3580 int nr = list_length(k->columns); (gdb) Regards, Romulo ---------------------------------------------------------------------- >Comment By: Niels Nes (nielsnes) Date: 2008-01-05 19:22 Message: Logged In: YES user_id=43556 Originator: NO A simple test shows the following error (ie no crash) !CONSTRAINT: constraints on temporary tables are not supported ---------------------------------------------------------------------- Comment By: Romulo Goncalves (romulog) Date: 2008-01-05 19:12 Message: Logged In: YES user_id=1498628 Originator: YES The bug need to be open. Not all the situations were covered. The following query crashes the mserver: create global temporary table t2 (id int primary key); All the combinations should be tested. Regards, Romulo ---------------------------------------------------------------------- Comment By: Romulo Goncalves (romulog) Date: 2007-12-01 15:45 Message: Logged In: YES user_id=1498628 Originator: YES In the stable branch a proper warning message is sent to the user in case he tries to insert column constraints on temporary tables. That message informs him that we do not support constraints on temp tables. In the future such feature will be implemented on the current branch. So I will open a feature request for constraints on temp tables. It was fixed in: sql/src/server sql_schema.mx, SQL_2-20, 1.132.2.2, 1.132.2.3 The test to cover this bug is: src/test/BugTracker/Tests/pkey_in_temp_table.SF-1841727 Regards, Romulo ---------------------------------------------------------------------- Comment By: Martin Kersten (mlkersten) Date: 2007-12-01 14:48 Message: Logged In: YES user_id=490798 Originator: NO I don't see a reason to forbid it... but the standard may dictate differently. (most likely not) ---------------------------------------------------------------------- Comment By: Romulo Goncalves (romulog) Date: 2007-12-01 14:43 Message: Logged In: YES user_id=1498628 Originator: YES The problem is not the check but the key creation. Should we accept or not pkey creation in temp tables? If yes I remove the lines: if (isTempTable(t)) 3495 return NULL; if not I will have to warn the user that he cannot create pkey on temp tables. Regards, Romulo ---------------------------------------------------------------------- Comment By: Niels Nes (nielsnes) Date: 2007-11-30 21:16 Message: Logged In: YES user_id=43556 Originator: NO temp tables can do without key checking.. So make a nice warning. ---------------------------------------------------------------------- Comment By: Romulo Goncalves (romulog) Date: 2007-11-30 15:13 Message: Logged In: YES user_id=1498628 Originator: YES After some debug I found out that in MonetDB we do not allow temp tables with primary key. This line of code is in the SQL code: sql_trans_create_ukey (tr=0xf534b8, t=0x1152a98, name=0x2aaabc674020 "pktable_ptest1_pkey", kt=pkey) at /ufs/goncalve/MonetDB/sql/src/storage/store.mx:3487 if (isTempTable(t)) 3495 return NULL; Why? I would like to know if we want to continue like this (so I need to create warning message to avoid the segmentation fault) or I should I change it? I take the second option. Regards, Romulo ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=482468&aid=1841727&group_id=56967 ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ _______________________________________________ Monetdb-bugs mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/monetdb-bugs
