1 |
On 04/05/2015 01:17 PM, Rich Freeman wrote: |
2 |
|
3 |
> If you're going to fork a library, and don't intend to keep the |
4 |
> packages API-compatible, then change the filenames. What is so hard |
5 |
> about this? LIbressl was even an outside fork, so it didn't come with |
6 |
> any of the baggage of "we're the real libssl team" or whatever. |
7 |
|
8 |
Libressl is (widely) API compatible with openssl. As said numerous times |
9 |
before: they broke the API when they thought it is security relevant. |
10 |
|
11 |
This is about the fact that they _added_ features which are not present |
12 |
in openssl. However, openntpd still compiles with openssl. |
13 |
|
14 |
> |
15 |
> Sure, we can do the USE=libressl route like we did with libav, but |
16 |
> since this is still new would it make more sense to just rename the |
17 |
> libressl files so that they can still go in /usr/lib but without being |
18 |
> called libssl? Then any package that wants to use them will need to |
19 |
> have their build logic changed accordingly. They aren't drop-in |
20 |
> replacements for each other anyway, as much as people would wish they |
21 |
> were, so we should resist the urge to pretend that they are. |
22 |
> |
23 |
|
24 |
By that you are effectively forking libressl and causing a huge mess |
25 |
downstream for both developers and users. It may even cause cross-distro |
26 |
breakage. I can name several examples of this happening due to debian |
27 |
going it's own way. Last of which affected openclonk (upstream merged a |
28 |
patch from a debian dev who expected any distro does the same hacks they |
29 |
do). |
30 |
|
31 |
So please, have a little sense for QA. Those ideas are just making it |
32 |
worse. This is something that has to be resolved upstream. If they don't |
33 |
cooperate long-term, then their fork will just die out for sure (and for |
34 |
good). However, I currently don't see strong signs for that. |