Re: BUILD_MULE vs. BUILD_WITHOUT_MULE
>>>> Charles G Waldman writes:
In the XEmacs package sources, file XEmacs.rules has a variable
BUILD_MULE.
In the ChangeLog there is the following entry:
1999-04-17 Karl M. Hegbloom <psu25682(a)odin.cc.pdx.edu>
* Makefile (SUBDIRS): Don't put "mule" in
SUBDIRS unless BUILD_MULE=t
I comment out BUILD_MULE in the "Developer Configurable"
section of
XEmacs.rules, as there instructed:
# Build Mule? Comment for no Mule
#BUILD_MULE=t
But guess what, after commenting this and typing Make, I get a ton
of
errors from building the Mule packages!
Well, on further investigation I see this ChangeLog entry:
1999-12-14 Andreas Jaeger <aj(a)suse.de>
* Makefile (SUBDIRS): Build for mule unless
BUILD_WITHOUT_MULE is
set.
Why was the new variable BUILD_WITHOUT_MULE introduced, when we
already have BUILD_MULE? And why isn't it documented in XEmacs.rules?
Because
it didn't work. The main Makefile doesn't include
XEmacs.rules - and BUILD_MULE is defined in XEmacs.rules only. As a
quick fix (I mean hack), I added this. AFAIR Karl did send in December a
patch to correct this - but this patch was also broken and didn't work
with mule builds.
Whassup wit dat?
Feel free to clean up everything - or document
it properly. I didn't
have the time so far to do it right :-(.
Andreas
--
Andreas Jaeger
SuSE Labs aj(a)suse.de
private aj(a)arthur.rhein-neckar.de