Date: Mon, 2 Aug 1999 20:20:02 -0700 (PDT) From: hoek@freebsd.org To: freebsd-bugs@FreeBSD.org Subject: Re: misc/6612 Message-ID: <199908030320.UAA02614@freefall.freebsd.org>
index | next in thread | raw e-mail
The following reply was made to PR misc/6612; it has been noted by GNATS.
From: hoek@freebsd.org
To: freebsd-gnats-submit@freebsd.org, sjr@home.net
Cc:
Subject: Re: misc/6612
Date: Mon, 2 Aug 1999 23:12:04 -0400 (EDT)
>
> bsd.man.mk can't handle man pages with ":"s, since it uses
> the ":T:S" constructs to extract elements.
No, that's not actually the reason.
The reason is because of loops such as this,
.for x in ${page}
$x: dependsonwhatever
create the $x file
.endfor
make(1) gets confused because ${page} is something like "a::b" and thinks
that "a" is the target with sources "b" and "dependsonwhatever".
Note that bsd.man.mk will work for the specific case where NOMANCOMPRESS
is defined, MANBUILDCAT is not defined, and MANFILTER is not defined.
It is broken for all other permutations of those settings.
The bsd.man.mk is an example of evolutionary design by committee. Fixing
it requires some rewriting...
Alternatively, make(1) itself could be modified to allow escape
characters in target specifications. I would not be opposed to this
type of modification to make(1) and it would be very trivial (see the
function src/usr.bin/make/parse.c::ParseDoDependency().
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-bugs" in the body of the message
home |
help
Want to link to this message? Use this
URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199908030320.UAA02614>