Re: Problems with (defvar foo) and Emacs 23

[Bastien <bzg@gnu.org>]

Hi Nick,

Nick Dokos <nicholas.dokos@hp.com> writes:

> Agreed, but the point is that each and every variable renaming will need
> to be checked in the light of these criteria. Bugs like this have the
> potential of creating havoc for a long time to come.
>
> It may be easier to start from a working state than to try to fix the
> current broken state piecemeal.

The thing is...  Org is working fine here.  Of course, I'm not a
"thousand eyes" by myself and some features might be broken in the
current state, I will continue trying to catch them.

> Besides, I'd rather have a single commit
> in the history that does the right thing: having one commit that creates
> the initial buggy state and then N commits at various times to fix the
> brokenness might create all sorts of problems (breaking bisectability
> e.g.)

Theoretically agreed.  But again: if you look at the diffs from the 
list of commits you sent, most of them are okay.  Let's try to fix the
ones that are *not* okay instead of going backward?  I will revert the 
commits if things are really broken.

Best,

-- 
 Bastien