RFC: Cleaning up Trunk

[Mark Kretschmann]

On 9/4/07, Seb Ruiz <ruiz at kde.org> wrote:
> I'm wondering if it would be worthwhile to rip out all of the old 1.4
> amarokisms from trunk. By this, I mean everything which hasn't been
> ported yet.
> eg:
>  - Playlist
>  - PlaylistItem
>  - MetaBundle
>
> etc etc.
>
> I know, it's a HUGE amount of work, but maybe it would help us in
> clearing out the codebase and understanding the new development
> techniques we are using. Another advantage is that it would help make
> obvious the giant holes we have in trunk.
>
> Obvious disadvantages would be that we would need to keep an older
> revision checked out locally to do any comparisons etc.
>
> We have at least a start on most of the major components of Amarok 2:
> Interface, Context, Playlist, Collections, Collection browser,
> Playlist browser, services.
>
> I don't imagine that the rest of the components would be refactored,
> but probably written again from scratch (with previous knowledge or
> referencing 1.4 code).
>
> Comments?

Yeah, I'm all for it. As we just saw with the "two playlists fighting
for the engine" bug, having the old code interfere with the new code
has it's own share of problems. It sucks to waste any time on such
stupid bugs for code that will be removed anyway. So we should do that
rather sooner than later.

Of course the worst case scenario could be that trunk is in an
unusable state for days. We have to be a bit careful not to let this
happen.

-- 
Mark