Front page | perl.perl5.porters |
Postings from July 2011
GSOC Status Report, Week 6
Thread Next
From:
Brian Fraser
Date:
July 7, 2011 08:48
Subject:
GSOC Status Report, Week 6
Message ID:
CA+nL+nZaWcxm7nCdT6oMvc-YFjtOK7ghZt+1MbgCb+RUcZVmPw@mail.gmail.com
Howdy folks.
Stash/GV cleanup is.. Well, not done yet, but close enough! I'm mid-rebasing
and, while there's a couple of regressions popping up, except for the one
relating to HEK_HASH() that I mentioned in a previous mail, I should
hopefully be able to get things sorted out today-ish; I expect that by
tomorrow I'll be nagging rafl for a smoke-me branch. So huzzah in advance
for that.
Progress has been fairly steady this week; I spent most of it
checking/correcting the work so far, writing tests, and these last two days,
organizing and amending commits.
For the bad news:
- I noticed a couple of minutes ago that I was doing something incredibly
silly with ->DOES, so that's not yet done. Assuming I don't have another
blunder like that though, ->DOES should be done soon (again).
- Several tests in uni/gv.t remain as TODO, because they deal with
prototypes, which aren't yet clean - Unfortunately, I had no idea that you
could use a arbitrary string as a prototype, not just @$+_, etc, so this had
completely escaped me until now. From a quick look, it doesn't appear
terribly daunting to fix (the prototype is stored as an SV), but unless
someone thinks they are important and they should be done foremost, I'll
shove them into the TODO pile.
- On the rebasing front, I still have to split up one hugeish toke.c
commit into something more reasonable, so that's not altogether ready for
review yet.
And since we are at it, I'm missing tests for these warnings/exceptions, so
any help in finding how to trigger them would be quite welcome:
gv.c, Gv_AMupdate: (Stub found while resolving|Can't resolve) method "Foo"
overloading "Bar" in package "Baz"
gv.c, amagic_call: Operation "X": no method found ...
doio.c, do_openn: Warning: unable to close filehandle X properly.
ext/mro/mro.xs: mro_get_linear_isa_c3: Inconsistent hierarchy during C3
merge of class ...
The rest of this week will go into polishing the work so far; Moreso when
the reviews and smoke reports start piling in :) With some luck, today or
tomorrow I'll push the branch, so start preparing to do your worst?
And that's all for now, I think. Short report, but my struggle with git
rebase aside, it wasn't a particularly dramatic week.
Thread Next
-
GSOC Status Report, Week 6
by Brian Fraser