Modelling Oracle Sequences

[Julian North]

Hi,

I am in the process of porting a number of large MSSQL databases to
Postgres.

The current id generation model in SQL server is an oracle style sequence
generator i wrote using a pretty simple stored proc and table structure.

I now need to do the same thnig in postgres.

In MSSQL I was able to open an UPDATE cursor, which places a lock on the row
in the sequence table. I then update, release and return.

Has anyone done something similar or have any pointers on how to do this
safely, securely without an update cursor?

Any infor appreciated.

Cheers,

Julian.

_______________ _______________ _______________ _______________ ____________
This e-mail has been scanned for all viruses by Star Internet. The
service is powered by MessageLabs. For more information on a proactive
anti-virus service working around the clock, around the globe, visit:
http://www.star.net.uk
_______________ _______________ _______________ _______________ ____________

---------------------------(end of broadcast)---------------------------
TIP 9: the planner will ignore your desire to choose an index scan if your
joining column's datatypes do not match

[Pavel Stehule]

Hello

why can't use normal sequence? There isn't difference betwen Oracle and
PostgreSQL?

http://www.postgresql.org/docs/7.3/s...esequence.html
regards
Pavel

On Fri, 21 Nov 2003, Julian North wrote:

Hi,

I am in the process of porting a number of large MSSQL databases to
Postgres.

The current id generation model in SQL server is an oracle style sequence
generator i wrote using a pretty simple stored proc and table structure.

I now need to do the same thnig in postgres.

In MSSQL I was able to open an UPDATE cursor, which places a lock on the row
in the sequence table. I then update, release and return.

Has anyone done something similar or have any pointers on how to do this
safely, securely without an update cursor?

Any infor appreciated.

Cheers,

Julian.

_______________ _______________ _______________ _______________ ____________
This e-mail has been scanned for all viruses by Star Internet. The
service is powered by MessageLabs. For more information on a proactive
anti-virus service working around the clock, around the globe, visit:
http://www.star.net.uk
_______________ _______________ _______________ _______________ ____________

---------------------------(end of broadcast)---------------------------
TIP 9: the planner will ignore your desire to choose an index scan if your
joining column's datatypes do not match

---------------------------(end of broadcast)---------------------------
TIP 4: Don't 'kill -9' the postmaster

[Shridhar Daithankar]

Julian North wrote:

Hi,

I am in the process of porting a number of large MSSQL databases to
Postgres.

The current id generation model in SQL server is an oracle style sequence
generator i wrote using a pretty simple stored proc and table structure.

I now need to do the same thnig in postgres.

In MSSQL I was able to open an UPDATE cursor, which places a lock on the row
in the sequence table. I then update, release and return.

Has anyone done something similar or have any pointers on how to do this
safely, securely without an update cursor?

Are you trying to do this?

http://developer.postgresql.org/docs...-sequence.html

Sequence values are transaction safe in postgreql. You don't have to lock
anything to get a correct and unique sequence value.

HTH

Shridhar
---------------------------(end of broadcast)---------------------------
TIP 6: Have you searched our list archives?

http://archives.postgresql.org