[Clfs-dev] Moving On to CLFS 1.2.0

[Ken Moffat]

On Sun, Sep 14, 2008 at 10:07:58AM -0400, Joe Ciccone wrote:
> Looking at the milestone for CLFS 1.2.0:
> 
> UTF-8, is anyone capable of tackling this? I probably could, But I don't
> have the time to do a lot of research anymore, it would take me months.
> 
 Hmm, I've been building UTF-8 clfs systems for some time now.
Unfortunately, as Alexander has showed on lfs-dev this weekend, I no
longer remember what I've changed, and the changes may prevent the
system working properly in a non-UTF-8 non-ascii locale!  Trouble
is, now that shadow has been re-homed at debian, there isn't much in
the way of UTF-8 man pages in a common build (in my total build, the
only thing I've noticed in the past which benefits from what I do
for UTF-8 is smartmontools (ISTR it has fancy quotes).

 One of the issues raised in the past was that upstream groff
doesn't yet support UTF-8.  AFAIK, that hasn't changed.

 The only other issue I remember was a reluctance to force this on
builders, so the instructions would need to be wrapped in something
like "if you intend to use a UTF-8 locale".  Probably then needs a
further page of why you might want to use (or not use) UTF-8, I'm
not sure I'm up to that (too much of an enthusiast, it is the "one
true path" and as a native English speaker the obstacles are few and
the benefits for reading/writing other languages are great).  Might
also need some sort of adjustment if anybody uses jhalfs - I don't
know whether that should default to 'on' (no adjustment of the xml)
or 'off (treat the extra instructions as comments).

 I'm not yet back into development (first tasks are to have another
go at firefox3 on ppc64, this time with separate nss - the version
in current ff3 is old, and broken on powerpc, and then sort out what
I should be building for my desktop - it's months since I last
upgraded packages).  After that, I plan to build an LFS system with
some of the changes from DJ's branch but without gcc-4.3 [ call me
chicken if you want! ] to understand any breakage.  Then, I'll be
repeating with gcc-4.3-latest.

 After that, I'm willing to do this (so, maybe Summer 2010 at my
current speed ;) - but, that assumes people agree that UTF-8 is a
good idea.  Anyone have any views on this ?

 BTW Joe, thanks for getting 1.1.0 out.

ĸen
-- 
das eine Mal als Tragödie, das andere Mal als Farce