* David Kastrup (2006-03-27) writes:
> Ralf Angeli <[EMAIL PROTECTED]> writes:
>
> You are living one hour in the future. Daylight saving time is not
> that drastic.
It's more like I am being haunted by the past. I sent the mail you
replied to about a week ago.
>> * David Kastrup (2006-03-21) writes:
>>
>>> Maybe. If you want to try something, feel free to do so.
>>
>> The following patch fixes the bug for me. The way preview/auto.el is
>> being passed to `preview-make-package' (or not) might be considered
>> peculiar.
[...]
>> + $(ELCC) -l preview/prv-install -f preview-make-package
>> $(DESTDIR)$(packagedir) "$(PACKAGE)" \"`echo $(AUCTEXDATE)|sed
>> 's/\(.*\)-\(.*\)-\(.*\)/\1.\2\3/'`\" \"$(AUCTEXVERSION)\" tex-site.el `{
>> test "x$(preview_enabled)" != xyes || echo "preview/auto.el"; }` && \
>
> That does not sound like a good idea since I would not bet on all
> shells considering `true` to be nothing (namely, different from an
> empty string argument).
>
> Perhaps just use the duplication
>
> if test "x$(preview_enabled)" = yes; then \
^^^
xyes
> $(ELCC) ... tex-site.el preview/auto.el ; \
> else \
> $(ELCC) ... tex-site.el ; \
> fi && \
>
> ?
>
> Sure, more wordy, but not much so.
Does the if..else..fi statement play along with &&? That means, does
it always return `true' if the respective statement was executed (and
`false' likewise)?
--
Ralf
_______________________________________________
auctex-devel mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/auctex-devel
