Re: [patch] stephen's staging headers patch

[Phil Edwards <pedwards at disaster dot jaj dot com>]

On Mon, Jul 02, 2001 at 04:52:09PM -0700, Benjamin Kosnik wrote:
> working with this patch, I have to say that I'd be in favor of a
> directory of soft links, not copied files: I found it a bit difficult
> to mentally switch to having my include "source" directory be the
> build directory. This is also counter to existing practice in the rest
> of the GNU toolchain and gcc/libjava etc. I found two duplicate files
> a bit confusing.

I would also prefer soft links.  Among other things, I can look at a file
in the build dir and /know/ where it came from instantly.

We can replace the calls to cp with @LN_S@, and if the build system doesn't
support symbolic links, either hard links or straight copying (I forget
which) will be used instead.  There's also an autoconf test which would
need to be added, and I don't recall its name offhand.

Without having tried the patch, just looking over it, I must say this is
very nice.  All the comments about elegance and drinkability are in order.
One thing:

> + c_compatibility_srcdir = ${glibcpp_srcdir}/include/c_compatibility
> + c_compatibility_builddir = ./
> + c_compatibility_headers = \
> + 	${c_compatibility_srcdir}/assert.h \
[...]
> + stamp-c_compatibility: ${c_compatibility_headers}
> + 	@if [ ! -d "${c_compatibility_builddir}" ]; then \
> + 	  mkdir -p ${c_compatibility_builddir} ;\
> + 	fi ;\
> + 	cp $? ${c_compatibility_builddir} ;\
> + 	echo `date` > stamp-c_compatibility

What are the plans for '${glibcpp_srcdir}/include/c_compatibility'?
Is that another symlink, or another variable?


Phil

-- 
Would I had phrases that are not known, utterances that are strange, in
new language that has not been used, free from repetition, not an utterance
which has grown stale, which men of old have spoken.
                                     - anonymous Egyptian scribe, c.1700 BC