Date: Tue, 2 Oct 2007 08:00:08 GMT From: Kay Lehmann <kay_lehmann@web.de> To: freebsd-ports-bugs@FreeBSD.org Subject: Re: ports/113750: update science/kst to 1.4.0 Message-ID: <200710020800.l928082r039995@freefall.freebsd.org>
next in thread | raw e-mail | index | archive | help
The following reply was made to PR ports/113750; it has been noted by GNATS. From: Kay Lehmann <kay_lehmann@web.de> To: Max Brazhnikov <makc@issp.ac.ru> Cc: Stefan Walter <stefan@freebsd.org>, GNATS <FreeBSD-gnats-submit@freebsd.org> Subject: Re: ports/113750: update science/kst to 1.4.0 Date: Tue, 02 Oct 2007 09:54:05 +0200 So I made some tests again and found that the problems with kst are quite similar to my problems with mbdyn (pr: ports/114805). In this case the problem is quite weird: To disable automatic recognition of cdf and netcdf, when the user installs kst without enabling the corresponding options, the Makefile.in of the datasources-dir gets patched to not include the subdir of the named feature. So if we want to install both of them, we get the problem, that Makefile.in doesn't get patched and the original distributed Makrfile.in is older than Makefile.am and so gmake runs an automake-process which results in a Makefile which does not include all the subdirs of datasources and not all plugins/libs are installed =>pkg_plist gets broken, if both cdf and netcdf are enabled. So this is not a problem of cdf. I described a similar problem in the problem report for mbdyn and asked how somethiung like this should be solved, but I got no answer. I can think of some dirty work arounds but I think there must be an elegant/right way to fix this. Do you have any suggestions? Maybe I should open a thread on the ports-mailinglist? Thanks and greetings, Kay
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200710020800.l928082r039995>