| 1 |
On Sun, 08 Mar 2009 00:08:37 -0800 |
| 2 |
Josh Saddler <nightmorph@g.o> wrote: |
| 3 |
> Is there a reason why we should ram through a new EAPI for something |
| 4 |
> that *looks* like another "Paludis supports this so let's make it a |
| 5 |
> Portage standard" proposal? Is there some kind of time deadline here |
| 6 |
> that you all want? |
| 7 |
|
| 8 |
If we were sticking through "things Paludis has that're useful", the |
| 9 |
list would be an awful lot longer... And not everything on that list is |
| 10 |
even in Paludis... But hey, some things never go out of fashion. |
| 11 |
|
| 12 |
A few things on the list are required pretty soon. Bug 249166 and |
| 13 |
various existing screwups in the tree suggest that [use(+)] (or |
| 14 |
whatever syntax is decided upon) is going to crop up fairly frequently. |
| 15 |
And certain people are reluctant to nuke their prepalldocs calls |
| 16 |
despite Council decisions... I think the rest are just there because |
| 17 |
they're useful and easy. |
| 18 |
|
| 19 |
> Also, why the bannination of || ( foo? (.).) -- how is it error prone, |
| 20 |
> exactly. |
| 21 |
|
| 22 |
Last time I checked, every single use of foo? as a direct child of || |
| 23 |
in the tree was wrong, as were the Portage docs. Let's say you have the |
| 24 |
following: |
| 25 |
|
| 26 |
DEPEND="|| ( |
| 27 |
foo? ( cat/foo ) |
| 28 |
bar? ( cat/bar ) |
| 29 |
cat/baz |
| 30 |
)" |
| 31 |
|
| 32 |
Then this is wrong: |
| 33 |
|
| 34 |
if use foo ; then |
| 35 |
myconf="--enable-foo" |
| 36 |
elif use bar ; then |
| 37 |
myconf="--enable-bar" |
| 38 |
else |
| 39 |
myconf="--enable-baz |
| 40 |
fi |
| 41 |
|
| 42 |
Here's why: |
| 43 |
|
| 44 |
Say the user already has baz installed, and has USE="foo". The package |
| 45 |
manager will see that baz is already installed, and as per the rules |
| 46 |
for || will not look at the foo or bar deps. So if the user doesn't |
| 47 |
have foo installed, the --enable-foo will be wrong. |
| 48 |
|
| 49 |
So the 'correct' code would have to be something like this: |
| 50 |
|
| 51 |
if use foo && has_version cat/foo ; then |
| 52 |
myconf="--enable-foo" |
| 53 |
elif use bar && has_version cat/bar ; then |
| 54 |
myconf="--enable-bar" |
| 55 |
else |
| 56 |
myconf="--enable-baz |
| 57 |
fi |
| 58 |
|
| 59 |
Except this is *still* a mess, because a user who has USE="foo bar" is |
| 60 |
going to end up with a differently configured package depending upon |
| 61 |
what he happens to have installed up-front, which is something that's |
| 62 |
not supposed to happen. |
| 63 |
|
| 64 |
-- |
| 65 |
Ciaran McCreesh |
Archives