1 |
-----BEGIN PGP SIGNED MESSAGE----- |
2 |
Hash: SHA256 |
3 |
|
4 |
On 14/08/15 06:43 PM, Johannes Huber wrote: |
5 |
> |
6 |
> |
7 |
> Am 08/15/15 um 00:19 schrieb Andrew Savchenko: |
8 |
>> Hi, |
9 |
> |
10 |
>> While I have no objections about EAPI 4 deprecation (except |
11 |
>> concerns mentioned above), I see no strong need for this also. |
12 |
>> Just declare EAPI 5 as recommended. Having legacy support for |
13 |
>> EAPI 4 will not hurt possible contributions. |
14 |
> |
15 |
> Just imagine that the Gentoo developer quiz doesn't contain |
16 |
> question about ancient stuff. Would reduce the question count at |
17 |
> least by some questions. Which could lower the barrier to step |
18 |
> forward for some people. Or do we have enough developers? |
19 |
> |
20 |
>> Best regards, Andrew Savchenko |
21 |
> |
22 |
> |
23 |
|
24 |
So first of all, yes i believe all eclasses support EAPI5 by now. |
25 |
|
26 |
Secondly, though, conversion to EAPI5 is not actually trivial, there |
27 |
are a couple of things, 'usex' related for instance, that also need |
28 |
to be taken care of. If it was just a matter of running a sed -e |
29 |
's/^EAPI=4/EAPI=5/' on all in-tree ebuilds this would have been done |
30 |
a long time ago. |
31 |
|
32 |
So although deprecation of EAPI4 is a nice thought, there is still |
33 |
some work to be done. |
34 |
|
35 |
Finally, the gentoo developer quiz -should- still contain questions |
36 |
about ancient stuff. There are still EAPI0 and EAPI2 ebuilds in the |
37 |
tree at least. |
38 |
|
39 |
-----BEGIN PGP SIGNATURE----- |
40 |
Version: GnuPG v2 |
41 |
|
42 |
iF4EAREIAAYFAlXOkvMACgkQAJxUfCtlWe1VkQD/cBeJW7Go12EkpSDL86MGzcNJ |
43 |
nHOBBHkdH9iQPCNfeo0BAO3v6rs7FHEIeJ7ze+JDFGqvJcZbsdcXZafRZaqbpwLE |
44 |
=bh9T |
45 |
-----END PGP SIGNATURE----- |