| 1 |
On Sun, 24 Jul 2016 10:17:24 +0200 |
| 2 |
Chí-Thanh Christopher Nguyễn <chithanh@g.o> wrote: |
| 3 |
|
| 4 |
> Michał Górny schrieb: |
| 5 |
> > So, to summarize we shouldn't fix existing code because people did |
| 6 |
> > assume accepting invalid parameters was fine. |
| 7 |
> |
| 8 |
> You are changing the API of versionator.eclass, even if it was unintentional API. |
| 9 |
|
| 10 |
There is no such thing as 'unintentional API'. API is what's described. |
| 11 |
If you rely on internals, random undefined behaviors or whatever, it's |
| 12 |
not a part of the API. |
| 13 |
|
| 14 |
Yes, I know that most of Gentoo developers are bright enough to assume |
| 15 |
'if it is not explicitly forbidden in words I can't find a way to bend |
| 16 |
to pretend it's allowed anyway, it's allowed even if it doesn't make |
| 17 |
any sense'. But that's a social problem that could easily solved by more |
| 18 |
proactive retirements, and not a good API design point. |
| 19 |
|
| 20 |
> > Then ebuilds will fail just the same |
| 21 |
> |
| 22 |
> No. Before, ebuilds would maybe display an upgrading message when they |
| 23 |
> shouldn't, or vice versa. Now the eclass dies on them. |
| 24 |
|
| 25 |
So what's the alternative? Design another eclass where ebuilds will |
| 26 |
fail just the same because people will ignore the more explicit |
| 27 |
requirement just the same as they do ignore the API? |
| 28 |
|
| 29 |
The problem is not 'there is a valid case to pass useless parameters |
| 30 |
to the function'. The problem is 'I make an invalid assumption about |
| 31 |
what will happen based on my limited knowledge which I believe is |
| 32 |
more correct than people who wrote package managers'. |
| 33 |
|
| 34 |
-- |
| 35 |
Best regards, |
| 36 |
Michał Górny |
| 37 |
<http://dev.gentoo.org/~mgorny/> |
Archives