Re: State of the CEDET merge

[David Kastrup]

Lluís <address@hidden> writes:

> David Kastrup writes:
>> Sounds like having to track two separately moving targets.
>
> In fact, I'm only tracking changes introduced in Emacs. But not those
> introduced after my last merge for a specific file (these will have to
> be manually re-checked).
>
> Changes in the cedet trunk should be automatically merged when I merge
> the branch.
>
>
>> Any idea how to make the respective developers aware of the problem
>> and move in a more synchronized fashion, so as to decrease the speed
>> with which the task you have focused on grows?
>
> There's no easy solution.
>
> On one hand, files in Emacs where introduced with modification wrt the
> cedet CVS, so some of them are hard to track.
>
> On the other hand, people won't be able to contribute all the fixes into
> cedet instead of emacs and expect emacs tu pull from cedet; not until I
> finish the file-rename branch.
>
> All this, added with my lack of elisp skill, knowledge on cedet
> internals and knowledge on what has been changed and why, make the task
> a tough one.

Let's assume that you get the task completed in the manner you envision
and you are working on right now, and the merge and synch happens as
planned.

Will that leave Cedet and Emacs in a state where future synchronizations
of Cedet to Emacs will be possible in a semi-automatic manner, like Gnus
is synchronized frequently right now?

Or will it mean that every future synchronization will require just as
much effort as your current work?

In short: if you manage to catch up with your target, will it be
reasonably easy to keep it from running off again?

-- 
David Kastrup