Michael Fuhr wrote:
>On Fri, Jan 14, 2005 at 12:03:02PM +0000, Bradley Kieser wrote:
>
>
>
>>I have an admin script that drops a schema and then recreates it (quick
>>way to refresh from backup). However, if the schema doesn't exist, the
>>drop errors, as you would expect. The next SQL command, the schema
>>create, is never executed.
>>
>>
>
>How do you know the create is never executed? Are you looking for
>the schema and not finding it? How are you looking? Or is there
>an error message? What is it?
>
>
Looking for the schema and not finding it.
>Are you doing the drop and create in the same transaction? If so,
>
>
I have auto-commit on.
>then the drop statement's failure is preventing subsequent statements
>in that transaction from executing. If that's not the problem then
>please post a self-contained test case so we can see exactly what
>you're doing.
>
>
Hmm, maybe I need to explicitly call commit? Will try and get back to you.
>
>
>>I do not have ON_ERROR_STOP set so why is this happening?
>>
>>
>
>How do you know ON_ERROR_STOP isn't set? Could it be set in
>a .psqlrc file? What does \set show?
>
>
>