svn commit: r324154 - head/astro/aa

[Alexey Dokuchaev]

On Fri, Aug 02, 2013 at 02:54:53PM +0400, Boris Samorodov wrote:
> 02.08.2013 12:33, Alexey Dokuchaev п©п╦я┬п╣я┌:
> > New Revision: 324154
> > URL: http://svnweb.freebsd.org/changeset/ports/324154
> > 
> > Log:
> >   - Convert Makefile header to the new two-line form
> >   - Remove 404'ing URL from MASTER_SITES
> >   - Do not explicitly set EXTRACT_AFTER_ARGS, use NO_WRKSUBDIR instead
> >   - Avoid patching makefile: pass correct values via MAKE_ARGS; this also
> >     remove double whitespace before -Wall from the build logs (nice)
> >   - NOPORTDOCS -> PORT_OPTIONS:MDOCS, get rid of .for loops, unmute the
> >     installation commands, make pkg-message display consistent with package
> >   - Reformat port description for better readability while here
> [...]
> >  MAINTAINER=	frank at dynamical-systems.org
> 
> Hm, I thought that maintained ports should go though PRs...
> Am I wrong here?

Normally yes.  However: 1) these are the ports which have been failing for
a while on the build cluster, and no one stepped to fix it; 2) miwi@ asked
me if I can help to offload the pile of error logs, so I guess I can have
an informal blessing from portmgr@ side; 3) I try not to touch/change any
user-visible logic of the port; 4) I have several hundreds logs to read,
analyze, fix, test (several times on different CPU arches), and commit.
I know what I'm doing, I try my best to make sure my fixes are correct, I
have time to invest in ports now, and I don't see that being bureaucratic
is worth anything -- I emphasize -- in this particular case.

That said, for some of these, I nonetheless ask for explicit approval, esp.
if maintainer hangs out on IRC.

./danfe