Mailing List Archive

Tomáš Chvátal posted on Mon, 10 Aug 2009 18:22:21 +0200 as excerpted:

> kde3 - Quo vadis mate.
> kde4.2 - Stable or not to stable that is the question there
> kde4.3 - oh yeah we are great 0 day 0 hour commit :D

FWIW, 4.3 killed a couple bugs here, so it may be better to focus on
stabilizing it than on having to cope with the same bugs in 4.2.x. Of
course, that might be the continuing case, the newest always killing
serious bugs in the old, for awhile longer...

4.3.0 is good enough to finally kill 3.5 today...

I may reply to the desktop list update I see waiting to be read, with a
bit more detail, later.

--
Duncan - List replies preferred. No HTML msgs.
"Every nonfree program has a lord, a master --
and if you use the program, he is your master." Richard Stallman

Tomáš Chvátal wrote:
> When -> 19:00 UTC @ 20.8.2009
> Where -> #gentoo-meetings


> Topics:
> kde3 - Quo vadis mate.
> kde4.2 - Stable or not to stable that is the question there
> kde4.3 - oh yeah we are great 0 day 0 hour commit :D
> qt4-tng eclass - we are better than startrek too :]
> Back to the future 2 - I was there, and brought back the plans for 4.4 we will
> have after this meeting (The longterm plans ;]).
>
> Looks like these could last us for a while. So thats it.

Our meetings are usually long and tiring. I would like to prevent that.
I propose we take this week to discuss the topics on list, so that we
only need to make final decisions/voting during the meeting, with
minimal discussion. Then we could be done in an hour and have more time
to chase girls or drink beers afterwards...

Cheers,
Ben

Dne čtvrtek 13 Srpen 2009 19:57:39 Ben de Groot napsal(a):
> Our meetings are usually long and tiring. I would like to prevent that.
> I propose we take this week to discuss the topics on list, so that we
> only need to make final decisions/voting during the meeting, with
> minimal discussion. Then we could be done in an hour and have more time
> to chase girls or drink beers afterwards...
Aye aye Ben. This make absolute sense.

Also When i am responding to this thread, Jorge asked me to add to the list:
- unversioned sets and what to do with set at all
My addition to this topic is:
I would remove the unversioned sets for now and if someone is interested he
would start working on the portage specification which would actualy do the
sets now. Currently they are huge PITA and i think life would be better
without them.

- kde3
Here i wrote nice mail to the desktop too, so read it, if we wont get
manpower we need to start to slowly remove packages for kde3 from main tree
and try to fix biggest holes. (I seriously hope someone will volunteer on the
meeting to help us with kde3).
- kde4.2
I would go pass the stable on 4.3.0 or 4.3.1 we still have tons of bugs open
and from what i can see the 4.3 didnt really open any major bugs so far.
- kde4.3
Status: done ; Issues: there are same issues for 4.3 as were for 4.2 so just
need fix them so we can stable
- qt4-tng eclass
Here i dunno what ever you plan guys, so enlighten me :P
- Back to the future 2
What do you think we should focus our team powers. Only fixing stuff is bad,
we need enhancements, so i want to hear constructive ideas, call this part
brainstorming.

Tomas

Tomáš Chvátal wrote:

> - qt4-tng eclass
> Here i dunno what ever you plan guys, so enlighten me :P

We have had qt4-edge.eclass in qting-edge overlay for quite some time
now. I think that most of the new functionality is ready to go to
portage. As we now define phases we did not before, it would break
qt.eclass. So we decided to go for a new eclass. We said we would still
brainstorm/bikeshed over the name for this eclass, but i haven't heard
anything better than qt4-tng so far. New ideas are welcome.

As far as I understand the langs handling still needs work, or are we
ready to commit that as well?

Is there anything else needed, or any other comments, before we propose
the new eclass to -dev ML?

Cheers,

Ben

On Saturday 15 August 2009 19:47:35 Ben de Groot wrote:
> Tomáš Chvátal wrote:
> > - qt4-tng eclass
> > Here i dunno what ever you plan guys, so enlighten me :P
>
> We have had qt4-edge.eclass in qting-edge overlay for quite some time
> now. I think that most of the new functionality is ready to go to
> portage. As we now define phases we did not before, it would break
> qt.eclass. So we decided to go for a new eclass. We said we would still
> brainstorm/bikeshed over the name for this eclass, but i haven't heard
> anything better than qt4-tng so far. New ideas are welcome.
>
> As far as I understand the langs handling still needs work, or are we
> ready to commit that as well?
Works for most packages so I think it is safe to introduce it as a separate
function. I would strongly suggest though not to call it inside src_install()
function but use it separately instead
>
> Is there anything else needed, or any other comments, before we propose
> the new eclass to -dev ML?
Please review it first since there are some "pretty" hacks like :
a) fallback to S="${WORKDIR}/{MY_P} when ${WORKDIR}/${P} doesnt exist
b) DOCS installation on src_install and a special variable DOCSDIR which
defines the path under which the docs are located on source dir
c) some other hacks I don't remember :)

Since I was the one who introduced those hacks, I am not really sure you 've
noticed them o:)

>
> Cheers,
>
> Ben

--
Markos Chandras (hwoarang)
Gentoo Linux Developer [KDE/Qt/Sound/Sunrise]
Web: http://hwoarang.silverarrow.org