> So it's time to create a few cvs-branches.
I disagree. Merging branches after a year or so of development in these
not-so-modular LyX sources is a major pain and I think it's a sure bet
to assume that most of the work spent in these branches will be lost
(for quite a while at least). You see it every day with your own work
(the 'new kernel'). Keeping branches in sync is an extra burden. *I*
certainly would not want to do it.
> As I understand the issue, we need two new branches:
> One for Hebrew, and one for the multi-byte encoding languages.
I see no reason why not to include the Hebrew patch immediately.
The performance penalty is minimal if I read it right.
Concerning the multi-byte encoding: We should have a discussion whether
it is sensible to use Unicode internally. The world is changing and last
years' arguments won't fit anymore. We could get rid of almost all of
the encoding stuff at the price of double sized buffers plus quite a bit
of work. But in the end thing would be much simpler....
The current state (between a mess and a vision) should be capable of
incorporating both patches. If they fit in, put them in and care about
everything else later.
Andre'
--
Andre' Poenitz ...................... [EMAIL PROTECTED]
