[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Mingw-cross-env-list] Proposal for new guidelines
From: |
Volker Grabsch |
Subject: |
Re: [Mingw-cross-env-list] Proposal for new guidelines |
Date: |
Thu, 28 Jan 2010 02:46:09 +0100 |
User-agent: |
Mutt/1.5.13 (2006-08-11) |
Mark Brand <address@hidden> schrieb:
> If I may try to summarize you, the prefix has the following benefits:
>
> It indicates that there is a least a possibility that the tool is
> platform specific.
>
> It indicates that the version of the tool is compatible with
> mingw-cross-env.
>
> It avoids interference with commands of the same name that might be
> in the PATH but not provided by mingw-cross-env.
Yes, these were exactly my points.
> I wonder though how much extra trouble it will be to make a package
> invoke, for example, "i686-pc-mingw32-flex". I'd be surprised if this
> "just works".
Whoops, that was a mistake on my side. The prefix issue won't
apply to flex and bison because we'll require them to be already
installed on the build system. Same for autoconf and automake.
The prefix question arises just for Qt, gSOAP, etc.
We already solved it for Qt. What's about gSOAP and others?
> I think I was thinking about patched Makefile.in generated from
> Makefile.am and whether admitting automake into the pantheon will affect
> this. Portaudio is the one case of this I find, so it's not much to
> worry about.
Okay, that's an important point. So we'll have to fix
src/portaudio-win32.patch,
too.
Anything else that needs discussion? Otherwise, let's start the
conversion.
Greets,
Volker
--
Volker Grabsch
---<<(())>>---
Administrator
NotJustHosting GbR