On Fri, Apr 16, 1999 at 12:09:32AM +0200, Axel Beckert wrote:
> there comes up another question: Shouldn't
> that be regarded as a bug? If mutt would parse the first of those two
> quoted commands correctly, it should regard
>
> index_format="%4C %Z %{%b %d} %-15.15L (%4l) %s"
>
> as one string, like e.g. tcsh interprets
>
> xyz" "zyx
>
> as one space-containing string. But instead mutt only interpets
>
> index_format="%4C
>
> as string and tries to interpret
>
> %Z
>
> as the next variable which should be set. This can't be proper!
No, you misunderstand. It parses the line two times.
When reading the send_hook it paeses it, and regards
%4C %Z %{%b %d} %-15.15L (%4l) %s
indeed as one string. However the parsing removes the quoting.
When the send-hook is executed, it now executes:
set index_format=%4C %Z %{%b %d} %-15.15L (%4l) %s
which (correctly) sets index_format to %4C and tries to interpret %Z.
Now, while talking about this, as send-hook's do not look at the
contents at definition time, maybe there is a way to stop the stripping
of any quotes. i.e. Handle the second argument as it was always included
in '
CU,
Sec
--
"Oderint dum metuant. Let them hate so long as they fear."
-- Lucius Accius (170-90 BC). One of history's earliest BOFHs.
