[Scummvm-devel] FreeSCI merge

[Willem Jan Palenstijn]

On Fri, Feb 13, 2009 at 12:47:28PM +0200, Eugene Sandulenko wrote:
> Hi Team,
> 
> This is separate e-mail with couple of technical questions. About the
> FreeSCI integration reasons and my apologies for lack of
> comminucations in that area (and reason), see e-mail "ScummVM
> integration".
> 
> The plan is to merge FreeSCI code tree with all historical data. I.e.
> with every commit which took place there.
> 
> Unfortunately current darcs repository seems to be the only survived
> source of such information, and some of the bits, such as branching
> info is lost there. Sad, still I use that as a best source.

I'm willing to manually re-add the branching info to an svn dump before
importing it, if you want. It looks like the original glutton revision
("project @ 933" in darcs) is an identical copy of the stable ("project
@ 932") revision, so it would be relatively easy to do. I can make time
for it tonight or this weekend.
 
> Now the question is. Where to put it? I consider putting it into
> 
>   /vendor/freesci/glutton
>   /vendor/freesci/stable
>   /vendor/freesci/web
> 
> I haven't come with better place as I don't want to clutter the root
> of the repository. Any thoughts?
> 
> After the code will land our tree AS IS. There will be svn copy of the
> needed bits wich will go to /trunk/engines/sci.

That sounds very reasonable.


> Another question is if you are agree and see benefit in having all
> FreeSCI repo saved for historical reasons, what about doing same for
> two other projects which we merged in the past, that is Sarien (AGI
> engine) and Reinherit (SAGA engine). I have backups of their pre-merge
> repositories, so that is doable.

That might be useful from time to time when trying to determine the
reason for certain pieces of code while tracking down bugs. If it is
does not cost too much time and effort to import them (and assuming they
don't have too large an impact on the size of our repository), I would
be in favour of this.



-Willem Jan