gnustep-dev
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Problem building corebase...


From: Stefan Bidi
Subject: Re: Problem building corebase...
Date: Sun, 29 May 2011 15:21:38 -0500

Well, I don't think it's that silly.  In reality the only real dependency here is the header.  Libuuid is already installed on all Linux systems since it's part of e2fsprogs.  I can't see the BSDs not having it readily available either, specially if they want to do anything needing ext2/3/4 access.  Windows has it's own implementation of UUID, but it shouldn't be too hard including some conditional code in there to use that instead of libuuid.  I think it's a reasonable dependency seeing as 100% of the platforms targeted by GNUstep already include a complete, reasonably bug-free implementation of this specification.

Stef

On Sun, May 29, 2011 at 2:57 PM, David Chisnall <address@hidden> wrote:
We looked at libuuid, but it seemed silly to add an external dependency for something that's a couple of dozen lines of code.  The algorithms are documented in the relevant RFCs, so a compatible implementation is not hard.

David

On 29 May 2011, at 20:54, Stefan Bidi wrote:

> The reason I used libuuid is because that's what Apple uses (http://en.wikipedia.org/wiki/Universally_unique_identifier).  It was easy enough to wrap it.  I guess we could have either fallback code for people without libuuid (in which case I'd need to create a configure script) or just completely get rid of the dependency.  Any opinions?
>
> Stef
>
> On Sun, May 29, 2011 at 2:30 PM, David Chisnall <address@hidden> wrote:
> On 29 May 2011, at 20:28, Stefan Bidi wrote:
>
> > Do you have the uuid library installed (libuuid1 + uuid-dev on Debian)?  This is a new dependency.  The uuid_t type is defined in uuid/uuid.h, there should be no reason why you're getting this error if the headers are correctly installed.
>
> Please take a look at the ETUUID code.  We have well-tested implementations of the UUID generation algorithms, without introducing an external dependency.
>
> David
>
> --
> This email complies with ISO 3103
>
>
> _______________________________________________
> Gnustep-dev mailing list
> address@hidden
> https://lists.gnu.org/mailman/listinfo/gnustep-dev


-- Sent from my Apple II



reply via email to

[Prev in Thread] Current Thread [Next in Thread]