Jean-Pierre Pelletier wrote:
> Can the type cache be flushed programmatically ?
> Could a workaround be for createArrayOf() to be extended
> to accept schema qualified type name?
I think that would only be part of the fix, you'd still need to make the
type cache schema aware (even if the driver didn't track changes to
search_path). The JDBC javadoc for createArrayOf states that the
typeName parameter "is the value returned by Array.getBaseTypeName". So
you'd need to change that to return a schema qualified name and I'm not
sure what else would need to change.
> An unrelated problem is that I had to override toString()
> in testtype to avoid "malformed record literal", is this the
> recommended way to handle composite type?
Right now the JDBC driver doesn't know much about composite types, so
that does seem like the correct workaround.
In response to
pgsql-jdbc by date
|Next:||From: Kris Jurka||Date: 2008-09-11 17:28:21|
|Subject: Re: looking for a jdbc method.|
|Previous:||From: Kris Jurka||Date: 2008-09-11 17:19:14|
|Subject: Re: Found a problem with 8.2-508 JDBC 4|