Date: Sun, 4 May 2003 17:03:56 +1000 (EST) From: Bruce Evans <bde@zeta.org.au> To: Ruslan Ermilov <ru@freebsd.org> Cc: Darren Reed <darrenr@reed.wattle.id.au> Subject: Re: misc/44148: installworld in 4.7-STABLE does not installIPFilter related header files Message-ID: <20030504164851.F2372@gamplex.bde.org> In-Reply-To: <20030503175034.GB97244@sunbay.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sat, 3 May 2003, Ruslan Ermilov wrote:
> On Sat, May 03, 2003 at 08:37:35AM +1000, Bruce Evans wrote:
> > On Sat, 26 Apr 2003, Ruslan Ermilov wrote:
> > > Here's an implementation (without ipfilter). Please review.
> >
> > I tested it lightly. It seems OK except for having lots of style bugs.
> >
> No surprise. ;)
>
> > % +compat:
> > % +.for i in ${LDIRS} ${LSUBDIRS} machine crypto
> > % + if [ -L ${DESTDIR}${INCLUDEDIR}/$i ]; then \
> > % + rm -f ${DESTDIR}${INCLUDEDIR}/$i; \
> > % fi
> >
> > New style bug: indent of only 4 in a shell script (it's only a continued
> > line at the level of the Makefile).
> >
> Moot, as this is not a shell script, but rather a string
> argument passed to a shell via -c. Strictly speaking,
> it should be formatted like follows:
>
> compat:
> .for i in ${LDIRS} ${LSUBDIRS} machine crypto
> if [ -L ${DESTDIR}${INCLUDEDIR}/$i ]; then \
> rm -f ${DESTDIR}${INCLUDEDIR}/$i; \
> fi
>
> But that would look uglier, so I take your advice. ;)
Nah. It needs to be read as a shell script, so it should follow the
normal indentation rules for a shell script. 8-char indents seem to be
normal for shell scripts too. There are some large examples in bsd.links.mk
and bsd.man.mk.
> > % .endfor
> > % mtree -deU ${MTREE_FOLLOWS_SYMLINKS} -f ${.CURDIR}/../etc/mtree/BSD.include.dist \
> > % -p ${DESTDIR}/usr/include
> ...
> > Maybe the compat stuff can be cleaned up a bit. mtree could be run more
> > globally as in all other Makefiles excrpt for the problem of removing the
> > symlinks to directories.
> >
> Not sure I understand what you mean. This is only to deal with
> the old directory-level symlinks; when we are sure we won't run
> this on such an old system, we can safely remove it.
(The above is part of the compat stuff). Yes, it may be necessary for
now. I just want to plan to remove it, or maybe go the other way and
push down installing of includes from ../Makefile.inc1 to here.
> > % +copies:
> > % +.for i in ${LDIRS} ${LSUBDIRS} crypto machine machine/pc
> > % +.if exists(${DESTDIR}${INCLUDEDIR}/$i)
> > % + cd ${DESTDIR}${INCLUDEDIR}/$i; \
> > % + for h in *.h; do \
> > % + if [ -L $$h ]; then \
> > % + rm -f $$h; \
> > % + fi; \
> > % + done
> >
> > More too-small indentations.
> >
> This case is more complicated. I'm not sure how to format it
> properly (I always wondered how to do it, BTW). I now have
> used the "make it look like a normal shell script" rule of
> thumb, this now at least looks sane.
We should also consider indenting nested make statements. Then the
command should probably be indented more than the usual 1 tab.
> Bruce, if you reply and get my ~/.vacation.msg back this
> would mean that I won't have a chance to commit it until
> I'm back, which 90% means it won't be committed up until
> code freeze in 5.x is over. I'd like to avoid it, so if
> you get my vacation message back, please take a moment
> and commit it for me, making any necessary style fixes.
OK. I haven;t got it yet :-).
Bruce
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20030504164851.F2372>