Subject: Re: pkg/31205 (Patch for dillo 0.8.5 to handle NetBSD releases without re-en
To: None <pkg-manager@netbsd.org, gnats-admin@netbsd.org,>
From: David H.Gutteridge <dhgutteridge@sympatico.ca>
List: pkgsrc-bugs
Date: 01/09/2006 03:55:02
The following reply was made to PR pkg/31205; it has been noted by GNATS.
From: David H.Gutteridge <dhgutteridge@sympatico.ca>
To: Tim Rightnour <root@garbled.net>
Cc: <gnats-bugs@netbsd.org>
Subject: Re: pkg/31205 (Patch for dillo 0.8.5 to handle NetBSD releases without re-en
Date: Sun, 8 Jan 2006 22:54:13 -0500
> This works and I have committed it. Can you see that this gets into the
> hands of the dillo maintainers?
Sure, I'm contacting them about something else I noticed in the code too. We'll
see what they say.
I had another thought: pkgsrc supports other OSes, so I had a look at FreeBSD,
OpenBSD, and DragonFlyBSD to see if they had reentrant resolver functions, and
I found that OpenBSD doesn't (their ports repository has a patch that simply
comments out the G_DNS_THREADED macro we were dealing with), FreeBSD apparently
doesn't before 5.3 (and gethostbyname() still isn't reentrant anyway), and from
looking at DragonFly's CVS repository, it doesn't seem like they do either.
So perhaps I should amend the patch to add checks for these three OSes as well?
I'm not sure what the pkgsrc policy would be on this though (aside from DragonFly,
that is) in terms of how much support to provide? These features could be moving
targets.
Anyway, the simplest addition would just be a line that checks for these OSes and
does nothing if it encounters one of them, viz.:
#elif defined(__DragonFly__) || defined(__FreeBSD__) || defined(__OpenBSD__)
Dave