"Florian Ragwitz" <rafl@debian.org>, "Zefram" <zefram@fysh.org>,"Father Chrysostomos" <sprout@cpan.org>,

Howdy all.

Bit of a slow week, as I didn't have net access for a couple of days. It was
awful, and doesn't leave me with that much to report; But on the other hand,
it was an awesome week - Midterm passed, the pad stuff got merged, and all
the bugs in that started popping out. Fun : ) I'll follow up on that in
another mail, so back to last week.

With the pad merge, getting our and the error messages with scalar
filehandles working done was only a few lines, which makes several TODO
tests to pass; And with prototypes cleanish (there are some issues with
latin-1 prototypes in error messages, I think), that's another bunch of
TODOs finished.

As per Father C's suggestion, I added a sv_sethek() function to sv.c (I
know, bad name. Suggestions welcome), and revamped pp_caller and sv_ref() to
use it. sv_ref() was also modified to take in a SV to modify (which can me
NULL, in which case it creates a new normal SV and works on that) - so it
can be used as (void)sv_ref(TARG, obj, TRUE) in pp_ref, but as
SvPV_nolen(sv_ref(NULL, obj, TRUE)) in sv_reftype (and similarly in
UNIVERSAL_DOES).

And, although not quite related to UTF-8 cleanliness, I also got kill "NAME"
and %SIG lookup nul-clean, as well as most magicalization in gv.c.

The plan for this week is basically the same as last week, but in addition
to that, I'll also be starting my incursion into toke.c and/or trying to get
attributes clean.

• "Florian Ragwitz" <rafl@debian.org>, "Zefram" <zefram@fysh.org>,"Father Chrysostomos" <sprout@cpan.org>, by Brian Fraser