| 1 |
On Thu, 9 Jan 2014 17:52:16 -0800 |
| 2 |
Patrick McLean <chutzpah@g.o> wrote: |
| 3 |
|
| 4 |
> On Fri, 10 Jan 2014 02:19:03 +0100 |
| 5 |
> Tom Wijsman <TomWij@g.o> wrote: |
| 6 |
> |
| 7 |
> > Or rather: "What does it take to migrate parts of pkgcore into |
| 8 |
> > portage?" |
| 9 |
> |
| 10 |
> Why not just switch to using pkgcore as the default package manager. |
| 11 |
|
| 12 |
Has anyone switched to pkgcore already? |
| 13 |
|
| 14 |
It needs to work well and be wide tested before it can become a default. |
| 15 |
|
| 16 |
> radhermit has been doing a lot of work lately getting EAPI 5 support |
| 17 |
> added, and generally fixing bugs etc. |
| 18 |
|
| 19 |
Are we there yet? |
| 20 |
|
| 21 |
The thing about pkgcore is that it is perceived as running behind; and |
| 22 |
with the lack of interest in it due to that, it seems that having it |
| 23 |
will take some time before it lifts off the ground. Don't get me wrong, |
| 24 |
it eventually will if we pursue it; but, when? It can take time. |
| 25 |
|
| 26 |
> Since we no longer have anyone intimately familiar with the |
| 27 |
> portage code, we should just switch to a more modern and readable |
| 28 |
> system. |
| 29 |
|
| 30 |
Agreed, but we also should keep Portage alive and kicking until then. |
| 31 |
|
| 32 |
> Rather than having random people trying to learn the |
| 33 |
> convoluted portage code, let's concentrate on getting pkgcore to a |
| 34 |
> point where we can replace portage with it. |
| 35 |
|
| 36 |
Either way, people need to learn something; and if we can't guarantee |
| 37 |
the short term future of Portage before pkgcore becomes usable, the |
| 38 |
long term future could rather be out of reach before you know it. |
| 39 |
|
| 40 |
In the short term we should focus on Portage, but in the long term we |
| 41 |
should indeed look at one or another replacement; and indeed does |
| 42 |
pkgcore soon appear to be the most interesting option here. |
| 43 |
|
| 44 |
To satisfy QA and Portage teams at the same time; I'm going to start |
| 45 |
digging into repoman soon as there are 80+ bugs open for it, so in the |
| 46 |
short term (days / weeks), I have no plans for pkgcore myself. |
| 47 |
|
| 48 |
From there on I plan to do a software re-engineering style inspection on |
| 49 |
the Portage base to learn and understand its convoluted structure; at |
| 50 |
that point we can make more educated choices as to which way to go |
| 51 |
forward, but until then ... let's just fix as much bugs as time permits. |
| 52 |
|
| 53 |
Don't get me wrong; pkgcore is great, but it takes time for attention |
| 54 |
to shift to it as Portage's slowly runs into more legacy code problems. |
| 55 |
|
| 56 |
-- |
| 57 |
With kind regards, |
| 58 |
|
| 59 |
Tom Wijsman (TomWij) |
| 60 |
Gentoo Developer |
| 61 |
|
| 62 |
E-mail address : TomWij@g.o |
| 63 |
GPG Public Key : 6D34E57D |
| 64 |
GPG Fingerprint : C165 AF18 AB4C 400B C3D2 ABF0 95B2 1FCD 6D34 E57D |
Archives