> Date: Fri, 01 Jun 2007 18:14:31 -0400
> From: Peter Memishian <peter.memishian at sun.com>
> Subject: Re: [osol-code] Re: [Fwd: [ksh93-integration-discuss]
ksh93-integration pre-reviewround"two" (webrev 2007-05-14)]
> To: Al Hopper <al at logical-approach.com>
> Cc: Peter Memishian <peter.memishian at sun.com>, Roland Mainz
<roland.mainz at nrubsig.org>, Roger.Faulkner at sun.com, Mike Kupfer
<Mike.Kupfer at sun.com>, Rod Evans <Rod.Evans at sun.com>, OpenSolaris Code
mailing
list <opensolaris-code at opensolaris.org>, Korn Shell 93 integration/migration
project discussion <ksh93-integration-discuss at opensolaris.org>, Bart
Smaalders
<bart.smaalders at sun.com>, Craig Mohrman <Craig.Mohrman at eng.sun.com>,
Casper Dik
<casper.dik at sun.com>, Stephen Hahn <stephen.hahn at sun.com>
>
>
> > I don't understand what the "problem" is with the way Roland has
> > implemented this? Is it a _big_ problem - or is is simply your
> > personal preference?
>
> Roger (cc'd on this email) is the right person to engage for the
> philosophical argument for it -- I am merely asking for consistency
> with the way the rest of our Makefiles work. The changes necessary
> to provide that consistency are minor and should take less than an
> hour (at most) to implement.
>
> Again, it is not about personal preference. It's about consistency and
> simplicity.
>
> --
> meem
Since I was mentioned, and at the risk of offending, the reason is this:
We want to do two things:
- Run lint over all of the lint-clean sources as part of
the nightly build, to keep those sources lint-clean.
- Be able to run lint over all of the sources, lint-clean or not,
in order to find and clean up lint-dirty sources as time permits.
The first part is achieved with $(SRC)/Makefile.lint
The second part is done by:
cd $SRC/lib ; make lint
cd $SRC/cmd ; make lint
So the next time we have a lint party (we used to do them every
couple of years), we can easily identify the lint-dirty sources.
Roger
