future of HgKit

[Andrey Somov]

Marcus Dubois wrote:
>     - clean-room implementation is far too difficult (if possible)
> 
> 
> I understand that it's a huge undertaking but I can't see why it 
> shouldn't be possible
> and in the end I think that it would be great for IDE integration.
> 
> Have you completely droped the will to, let me phrase it, "translate" it 
> to Java and what
> are the thoughts from the community about such a beast? Matt Mackall?
> 
> Kind regards
> Marcus 
> 

with HUGE effort it is possible to rewrite let's say Mercurial 1.3. But 
by the time it is finished Mercurial will be at version 1.7.
Rewrite will be always well behind the original. And it requires 
considerable help form the python community (documentation).
While with port you can relatively easy copy the maintenance to the port.
So I do not see how rewrite may possibly work in the long run.

Main concerns for IDE:
- stable and complete interface (parsing stdout is not very convenient)
- bearable performance
- minimum dependencies (preferably - zero, like it is now for old-good CVS)