Date: Fri, 03 Jun 2011 14:35:59 +0200 From: Matthias Andree <mandree@FreeBSD.org> To: freebsd-ports@freebsd.org Subject: Re: GPC 2006 (Pascal) -- deprecated or "expired"?? Message-ID: <4DE8D52F.3010207@FreeBSD.org> In-Reply-To: <BANLkTi=Qn6j4eVTxTcG0PdN1RjjvdEuQSQ@mail.gmail.com> References: <BANLkTimmQ7UYubpe0R9NYCVDqXhApcSdGA@mail.gmail.com> <20110603001251.GA66356@server.vk2pj.dyndns.org> <BANLkTi=Qn6j4eVTxTcG0PdN1RjjvdEuQSQ@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Am 03.06.2011 06:56, schrieb Rugxulo: > Hi, > > On Thu, Jun 2, 2011 at 7:12 PM, Peter Jeremy <peterjeremy@acm.org> wrote: >> On 2011-Jun-01 17:23:37 -0500, Rugxulo <rugxulo@gmail.com> wrote: >>> It seems somebody isn't very knowledgeable about GPC. :-( >> ... >>> the real deal: *BSD hates GPL, esp. GPLv3, >> >> Please stop spreading FUD. > > Sorry, not trying to be rude here (honestly), but it is GPLv3, which > clearly is not popular with *BSD. Please don't act like licenses don't > matter. The question is "base system" or "ports" - and for ports, it's much less of an issue than for the base system. > The whole point is, even if GPLv3 isn't your favorite, you shouldn't > throw away GPC without a suitable replacement. Sure, FPC is > semi-related (barely), but it's a whole different dialect, so the code > is NOT compatible! So there. FPC is at least something, even if not the same, and nobody besides you stated that GPC was removed because of the license. > In other words, if you can temporarily get GPC up to speed, you can > then compile Scott Moore's (ISO 7185) P5 with it, which is public > domain. (FPC won't compile it as-is.) That's better than just dumping > everything in favor of FPC only (incompatible), IMO. So - are you willing to maintain it? Or are you going to call others unknowledgeable? >> Even if there was another release 20070904, >> that is 3½ years ago - which seems to justify the "development >> has ceased" statement in the removal message. > > Development has ceased, yes. There were only a handful of maintainers > over the years. Plus the real dealbreaker is that GCC has backend bugs > (and incompatibilities) which the GPC devs either didn't understand or > couldn't afford the time to fix. So it stagnated. Hence "latest" is > only unofficial patches against GCC 3.4.4 (stable) and 4.1.2 (buggy). So, the world has moved on, and you're apparently the first person to come forward and say that GPC is worthwhile. Let me ask you a question: which software would you compile with GPC on FreeBSD? Just to figure out if there's just frustration on your end or need for a project. If there's need for a project, we can probably bring you up to speed with porting a newer version, providing you're willing to invest some time learning the ports conventions and actually maintain the beast for your foreseeable future (which should spread to more than a few weeks). > There can't be any big maintenance except making sure it actually > builds, esp. with newer GCCs. Other than that, I see no problems here. Does that imply you are volunteering? > "No reason to keep it around" is silly. Practically speaking, I know > most of you probably don't need or want Pascal, but it exists. And The bare existence isn't sufficient or necessary to create a port. We've had these discussions before. > suggesting FPC as an alternative is not an answer since it lacks > support for ISO 7185 (original, Wirth) or ISO 10206 (Extended). That's > the big reason to keep it around (obviously?), though I admit FPC is > much easier to build / bootstrap. > > Don't take this all the wrong way, I'm trying to point you in the > right direction. Well, the simple answer is, without a volunteer to invest the resources to make (a) GPC compile, (b) packaged properly for FreeBSD, (c) maintained so that it correctly compiles typical Wirth source code projects, it won't happen, as much as people pity it. Pascal isn't special in this regard, it just happens to be less common, so it's in turn harder to find a volunteer. :) > How hard can it be to rebuild GCC 3.4.4?? Seriously, it can't be that > hard for you guys. I've (rarely) even rebuilt DJGPP on Windows, and > surely it's 1000x easier on *nix!! So go ahead, do it, and send the PR with the port. But you'll need to deal with bug reports, incompatibilites, adapt the port to changing environments, and make sure it works on FreeBSD 7.3 and 8.1 and soon enough 9.0 for the Tier-1 architectures. Documentation how to proceed is at: http://www.freebsd.org/doc/en_US.ISO8859-1/books/porters-handbook/
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4DE8D52F.3010207>