Michael Bell wrote:
you can simulate sequence stuff (with an own table or scripting, mysql has some abilities for scripting imho), otherweise, they have the ability to autoincrement numbered columns... which works normaly like a sequence...
I do exactly this with a seperate table. Nevertheless this is much more unclean (compared to other DBMS).at least, the last oracle version i worked with - does actually the same... they have tables and triggers for this... i mean - how clean is this? ;(... and ibm db4 too if i remeber right... at least behind the 'sequence' commands - so the difference ist quite minimalistic...
in mysql you have to do it your own at the moment... yes - so i think maybe one day mysql will provide some standard compilant way for this too, instead of just using autoincrement ;)...
i just checked the code, to make this mysql-sequence table safe, we should add some lock and unlock commands to it... to get it atomic, otherwise this could couse trouble - i will reread the actual mysql documentation - but i think there havn't much changed in this issues...
(or is this already encapsulated via a transaction or something?)
for what is this dummy column? i mean, we don't need/use it, so don't create it ;)
all: SEQUENCE_EXIST and SEQUENCE_NEXT are the same commands? for all dbs... - i thought exist checks if an sequence exists? or what does it mean?
and SQL statement concatenation with ";" as seperator does notthis is new, usaly this should work with mysql too...
The problem is that concatenation allows to perform several SQL operations especially inserts in a very efficient way. This is interesting for the batch stuff where we need performance. finally I removed this code from OpenCA::DBI to get better maintainable code.hmm, but maybe the speed would be more importend there? and i don't see a problem between speed and maintable code...
i will check some stuff - maybe the perl-module for mysql isn't able to handle concated queries, but this would be really strange...
The major issue which is not really clear for me is the transaction handling if you use more than one table. You can use different types of backends for tables in one database. So how should the transaction stuff work under such circumstances? This is the reason why I don't use mysql. If somebody can explain this to me then I would use mysql too.
this is a good questions ;)
greetings dalini
------------------------------------------------------- This SF.Net email is sponsored by: Sybase ASE Linux Express Edition - download now for FREE LinuxWorld Reader's Choice Award Winner for best database on Linux. http://ads.osdn.com/?ad_id=5588&alloc_id=12065&op=click _______________________________________________ OpenCA-Devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/openca-devel
