Kenneth Hutchinson wrote:
> Also, every time I try to access my connection, it tells me that "This
> method is not yet implemented". It doesn’t tell me which method, though.
Don't you have a stack trace ?

> Can anyone shed some light on how to get PostgreSQL up and running with
> Crystal Reports v11? Should I switch to v10?

Did you try with the latest JDBC driver ?

> PostgreSQL db
Which version ?


---------------------------(end of broadcast)---------------------------
TIP 8: explain analyze is your friend

Kris Jurka wrote:
>
> On Thu, 10 Feb 2005, Xavier Poinsard wrote:
>
>
>>I read in the code that the Uninmplemented exception was designed to
>>reduce driver size, but would you consider a patch adding the name of
>>the unimplemented function to the message ?
>>
>
>
> I don't know, it doesn't seem all that useful. Anyone doing real
> development will surely know what function they are calling and will have
> the stacktrace handy. If we find out that Crystal Reports or other
> applications only return the error message with no context then we should
> probably do something about it. For now I believe time would be better
> spent actually implementing these methods.

Sure, but they are a lot of methods to implement and if we want to fix
quickly the functions used by Crystal reports (for example), without
stack trace, it is the only simple way.

>
> Kris Jurka
>
> ---------------------------(end of broadcast)---------------------------
> TIP 2: you can get off all lists at once with the unregister command
> (send "unregister YourEmailAddressHere" to majordomo@postgresql.org)
>


---------------------------(end of broadcast)---------------------------
TIP 2: you can get off all lists at once with the unregister command
(send "unregister YourEmailAddressHere" to majordomo@postgresql.org)

On Thursday 10 February 2005 11:05, Xavier Poinsard wrote:
> > I don't know, it doesn't seem all that useful. Anyone doing real
> > development will surely know what function they are calling and
> > will have the stacktrace handy. If we find out that Crystal
> > Reports or other applications only return the error message with
> > no context then we should probably do something about it. For now
> > I believe time would be better spent actually implementing these
> > methods.
>
> Sure, but they are a lot of methods to implement and if we want to
> fix quickly the functions used by Crystal reports (for example),
> without stack trace, it is the only simple way.

I haven't been following this thread very closely. Excuse me if I'm
totally off base here.

I assume you're talking about this method in org.postgresql.Driver:


public static SQLException notImplemented()
{
return new PSQLException(GT.tr("This method is not yet implemented."), PSQLState.NOT_IMPLEMENTED);
}

I agree with Kris that there is no need to include the method name in
the exception. It is already shown in the stack trace.

However, you can easily patch this method so that it includes the
caller's method name in its error message. All you have to do is peek
one level up the stack trace.


---------------------------(end of broadcast)---------------------------
TIP 5: Have you checked our extensive FAQ?

http://www.postgresql.org/docs/faq

Kris Jurka wrote:
>
> On Thu, 10 Feb 2005, Vadim Nasardinov wrote:
>
>
>>Out of curiousity though, what makes it error-prone?
>>
>
>
> Any ad-hoc parsing is easy to get wrong by forgetting corner cases. I'm
> not saying it can't be done correctly, just that we shouldn't bother.

Especially given that different JVMs format stack traces differently,
and Throwable.getStackTrace() is allowed to omit stack frames.

I think setting a DriverManager log writer is the simplest solution
here, closely followed by making notImplemented() take a method name arg.

-O

---------------------------(end of broadcast)---------------------------
TIP 7: don't forget to increase your free space map settings