Neil,
Will this allow you to pass bytea data as binary data in the parameters
section (ability to bind values to parameters) or will this still
require that the data be passed as a text string that the parser needs
to parse. When passing bytea data that is on the order of Megs in size
(thus the insert/update statement is multiple Megs in size) it takes a
lot of CPU cycles for the parser to chug through sql statements that
long. (In fact a posting to the jdbc mail list in the last couple of
days shows that postgres is 22 times slower than oracle when handling a
1Meg value in a bytea column).
thanks,
--Barry
Neil Conway wrote:
> On Thu, 11 Apr 2002 10:54:14 +0800
> "Christopher Kings-Lynne" <chriskl@familyhealth.com.au> wrote:
>
>>Out of interest, since the FE/BE protocol apprently doesn't support prepared
>>statements (bound variables), what does this patch actually _do_?
>
>
> It implements preparable statements, by adding 3 new SQL statements:
>
> PREPARE <plan> AS <query>;
> EXECUTE <plan> USING <parameters>;
> DEALLOCATE <plan>;
>
> I didn't write the original patch -- that was done by Karel Zak.
> But since that was several years ago, I'm working on cleaning it up,
> getting it to apply to current sources (which has taken a while),
> and fixing the remaining issues with it. Karel describes his work
> here:
>
> http://groups.google.com/groups?q=query+cache+plan&hl=en&selm=8l4jua%242fo0%241%40FreeBSD.csie.NCTU.edu.tw&rnum=1
>
> (If that's messed up due to newlines, search for "query cache plan"
> on Google Groups, it's the first result)
>
> Cheers,
>
> Neil
>
