1 |
Ok, so there are two fundamental ideas here: |
2 |
|
3 |
1) Keep the qt use flag, use it if a package offers qt3 or qt4 support. |
4 |
If both, then make it for the more recent version and add a local flag for |
5 |
qt3 support. |
6 |
|
7 |
A few of us like this one, including me. The downside to this is you get |
8 |
a USE that may look like "qt -qt3" which is a bit ugly. Upside is that it |
9 |
"just works". |
10 |
|
11 |
|
12 |
2) Remove qt use flag, and create qt3 and qt4 global flags. |
13 |
|
14 |
This is what a few others are behind. It's more descriptive of what's |
15 |
actually going on, but will disrupt ~30 packages that currently use the |
16 |
"qt" flag. |
17 |
|
18 |
|
19 |
I suppose I'm not really big on one versus the other. I was for #1 simply |
20 |
because it required the least amount of effort to implement, however the |
21 |
people who are in favor of #2 have volunteered to do the work to implement |
22 |
it as well as put qt3 into the use.defaults for 2006.1 so KDE will work |
23 |
"out of the box". I'm not inclined to go against them simply because I |
24 |
don't see a big downside to going to qt3/qt4 global flags as long as |
25 |
someone is willing to do the work. |
26 |
|
27 |
So, as long as nobody comes up with a major objection, consider this my |
28 |
recommendation to allow portage to move to the #2 scenario. The "it just |
29 |
works now" excuse isn't valid, either. And, if it turns out the change |
30 |
really sucks..well...we can always figure out something else. :) |
31 |
|
32 |
Comments and flames are most welcome. |
33 |
|
34 |
-- |
35 |
gentoo-dev@g.o mailing list |