[ Thread Index |
Date Index
| More lists.liballeg.org/allegro-developers Archives
]
On Wed, 2005-11-02 at 19:27 +0000, Peter Hull wrote:
> On Tuesday, November 1, 2005, at 01:26 am, Peter Wang wrote:
>
> > I would really like to migrate to Subversion ASAP, after 4.2.0 is out.
>
> Sounds good, I think we could get ourselves onto the beta programme by
> giving these reasons to SF.net:
> Allegro is an active project
> It has a substantial code-base but is not massive.
> Its developers are using several different platforms
> It's due for reorganisation which (moving and renaming), as I
> understand it, is the key advantage of SVN over CVS.
> It's not exactly mission-critical to anyone
There are many advantages. For me, the most important is the much
smaller bandwith (I'm on dialup). And then all the details. No more
flags to pass like -z9 or, -up to diff. Handling of binary files.
Moving/renaming/deleting of files and directories. Directories are
actually under version control. Then sane handling of
modules/branches/whatver (i.e., they don't exist anymore, I never could
understand how it works in CVS). Then the simple revision numbers,
nothing weird like with CVS. And if I think longer, I will find much
more still :)
> My only worry is the tools that we have. At the moment I use Emacs and
> ProjectBuilder for editing, both of which have good CVS integration.
> What's the situation with Subversion? It's hardly bleeding-edge but it
> does not seem to be as well supported. WIndows folks seem to like
> Tortoise CVS and Win CVS; is there an equivalent?
Tortoise SVN.
> Regarding the 4.2.x branch, I would suggest keeping it on CVS for
> continuity and safety. When Subversion is freely available on SF.net,
> and assuming we like it, we can migrate at that point. There will be
> enough to worry about (even writing new code!!!) without trying to
> maintain 2 trees on a beta VC system.
Subversion is not beta. Only SF will provide a beta version at first..
but I really don't see how even SF can mess up much there.
--
Elias Pschernig