| 1 |
On Sun, 24 Jul 2016 00:17:21 +0200 |
| 2 |
Chí-Thanh Christopher Nguyễn <chithanh@g.o> wrote: |
| 3 |
|
| 4 |
> Michał Górny schrieb: |
| 5 |
> > Ensure that proper number of parameters is passed to each versionator |
| 6 |
> > function; die otherwise. This prevents the functions from proceeding |
| 7 |
> > with undefined behavior when mis-called. |
| 8 |
> |
| 9 |
> You are making what versionator.eclass accepts stricter. That it used to work |
| 10 |
> when passed multiple versions may be unintentional, but I think you need to |
| 11 |
> introduce a new eclass or new function names in this case. |
| 12 |
|
| 13 |
So, to summarize we shouldn't fix existing code because people did |
| 14 |
assume accepting invalid parameters was fine. |
| 15 |
|
| 16 |
What we should do instead is create an almost identical copy of it, ask |
| 17 |
people to switch to new code with parameter checks. But... why? |
| 18 |
|
| 19 |
If people already pass valid parameters, then new and old code would be |
| 20 |
functionally identical. If they do not, then they can fix parameters |
| 21 |
and stay with the old code. But they shouldn't have been doing that in |
| 22 |
the first place... |
| 23 |
|
| 24 |
So in the end, we create another copy of the functions that's used |
| 25 |
inconsistently with the old copy, and only when people feel like |
| 26 |
switching. Then ebuilds will fail just the same because people wouldn't |
| 27 |
care to *ensure* they pass valid parameters while switching, as long as |
| 28 |
they won't trigger it on their limited test case... |
| 29 |
|
| 30 |
-- |
| 31 |
Best regards, |
| 32 |
Michał Górny |
| 33 |
<http://dev.gentoo.org/~mgorny/> |
Archives