Thread: Generating an ANSI compliant schema recreation script
I need to capture the schema on a postgres database and recreate it on another ansi compliant platform. Is it possible to generate a file (perhaps from pg_dump?) as a sequence of ansi compliant SQL commands which can be used to recreate the structure?
Thanks in advance!
J French <hikenboots@gmail.com> writes: > I need to capture the schema on a postgres database and recreate it on > another ansi compliant platform. Is it possible to generate a file > (perhaps from pg_dump?) as a sequence of ansi compliant SQL commands > which can be used to recreate the structure? If you had read the pg_dump manpage, you would have seen: -s, --schema-only dump only the schema, no data The schema dump will be close to ANSI-compatible, but you will probably have to edit it a bit. -Doug
I did read the page. Been there done that, ran the script. My question was if there was a canned script out there that I didn't have to clean up on the fly. This will be an cron job for a convoluted development process. Thanks though.
On 10/29/05, Douglas McNaught <doug@mcnaught.org> wrote:
J French <hikenboots@gmail.com> writes:
> I need to capture the schema on a postgres database and recreate it on
> another ansi compliant platform. Is it possible to generate a file
> (perhaps from pg_dump?) as a sequence of ansi compliant SQL commands
> which can be used to recreate the structure?
If you had read the pg_dump manpage, you would have seen:
-s, --schema-only dump only the schema, no data
The schema dump will be close to ANSI-compatible, but you will
probably have to edit it a bit.
-Doug
J French <hikenboots@gmail.com> writes: > I did read the page. Been there done that, ran the script. My question > was if there was a canned script out there that I didn't have to clean up > on the fly. This will be an cron job for a convoluted development > process. Thanks though. Yeah, that would be a bit trickier than a one-off edit. :) -Doug
J French <hikenboots@gmail.com> writes: > I did read the page. Been there done that, ran the script. My question was > if there was a canned script out there that I didn't have to clean up on the > fly. This will be an cron job for a convoluted development process. If your schema isn't using any non-standard features, I would think that the result of pg_dump would be pretty standard, with the exception of a few SET commands at the front and the ALTER OWNER commands. (The latter can be suppressed with --no-owner.) What exactly is giving you a problem? regards, tom lane
No problems. I am about to write a python script to cleanup the output of pg_dump. Before I do I just wanted to verify that there wasn't an already available script which I could feed directly into another ansi compliant database without modification. No sense reinventing the wheel. The output of pg_dump is pretty clean for this purpose already. just thought I'd ask beforehand.
On 10/29/05, Tom Lane <tgl@sss.pgh.pa.us> wrote:
J French <hikenboots@gmail.com> writes:
> I did read the page. Been there done that, ran the script. My question was
> if there was a canned script out there that I didn't have to clean up on the
> fly. This will be an cron job for a convoluted development process.
If your schema isn't using any non-standard features, I would think that
the result of pg_dump would be pretty standard, with the exception of a
few SET commands at the front and the ALTER OWNER commands. (The latter
can be suppressed with --no-owner.) What exactly is giving you a
problem?
regards, tom lane