| 1 |
Dnia 2014-06-11, o godz. 13:12:38 |
| 2 |
Chí-Thanh Christopher Nguyễn <chithanh@g.o> napisał(a): |
| 3 |
|
| 4 |
> > REQUIRED_USE=" |
| 5 |
> > curl_ssl_winssl? ( elibc_Winnt ) |
| 6 |
> > ssl? ( |
| 7 |
> > ^^ ( |
| 8 |
> > [...] |
| 9 |
> > ) |
| 10 |
> > )" |
| 11 |
> |
| 12 |
> I don't like this. If the user specifies several SSL providers in |
| 13 |
> make.conf, it should mean that any of these is fine and the ebuild can |
| 14 |
> choose an arbitrary one. The exactly-one-of operator would cause emerge |
| 15 |
> to complain in this case and possibly force the user to have complex |
| 16 |
> package.use setups. |
| 17 |
|
| 18 |
Your idea comes with three significant drawbacks: |
| 19 |
|
| 20 |
1. USE flag setups become unclear -- user sees five different SSL |
| 21 |
providers turned on and has no clue which one is actually used. |
| 22 |
|
| 23 |
2. You create 2^n-1 valid USE flag combinations which map to n |
| 24 |
different file sets. This means that there are 2^n-n-1 useless USE flag |
| 25 |
combination which make matching binary packages a PITA. |
| 26 |
|
| 27 |
3. There is no clean way of enforcing SSL provider match between |
| 28 |
packages. Wasn't this thread initially about curl and rtmpdump |
| 29 |
requiring matching flags? |
| 30 |
|
| 31 |
-- |
| 32 |
Best regards, |
| 33 |
Michał Górny |
Archives