On Mon, Oct 27, 2008 at 10:37:19PM +0000, Andy Wardley wrote: >M wrote: >> It all works fine apart form the uniquing stage which does nothing > >That's most puzzling. list.unique definitely works (for the >definition of "works" that I just tried):
I'm sure it does, if it didn't it would have been fixed ages ago.
> [EMAIL PROTECTED] ~] tpage
> [% list = [10 20 10 11 1 20 21 'foo' 'Foo' 'bar' 'foo'];
> list.unique.join
> %]
oooh I din't know you could do multiple lines inside a [% %] with ;s
handy.
>It's so trivially simple that I can't see anything that can
>go wrong with it. Unless I'm missing something obvious...
I assumed this to be the case, which is why I was baffled, I didn't know
it was
> sub list_unique {
> my %u;
> [ grep { ++$u{$_} == 1 } @{$_[0]} ];
> }
but assumed it was that simple.
> > I've solved my problem by adding a DISCRETE to the SQL
>
>That's the better solution, I think. But it's still very puzzling as
>to why list.unique wasn't working. Could it be that your data wasn't
>exactly the same? Perhaps some elements had leading or trailing
>whitespace that differentiated it enough to make it non-unique?
Possible, but the DISCRETE has worked and that wouldn't if there was
whitespace as you suggest.
I suspect it has something to do with the hash returned by the DBI
query, but that's all a bit beyond me tbh, and having a solution means I
don't need to think about it any more.
M
--
GPG key here. http://www.gatrell.org/gpg.public.key
signature.asc
Description: Digital signature
_______________________________________________ templates mailing list [email protected] http://mail.template-toolkit.org/mailman/listinfo/templates
