[kde-freebsd] PATCH: net/knemo-kde4 IPv6 support

[Anthony Jenkins]

On 02/27/16 03:55, Adriaan de Groot wrote:
> On Wednesday 24 February 2016 23:59:37 Anthony Jenkins wrote:
>> This (git-based) patch allows net/knemo-kde4 to compile the IPv6 stuff.
>>
>> Anthony
> Hi Anthony,
>
> Thanks for the patches. It seems none of the K-F people have IPv6 available to
> test the usefulness of the patches (for instance, I'm on one of the crappy
> Dutch ISPs that don't offer IPv6 at all). So we'll have to take your word for
> it that this is functionally correct. (Which is fine).
>
> But I've got a few questions about the patch itself, since it seems to do way
> more than just add IPv6 support:
>
>   - change to xmlstorage.h -- just removing an unused member?
>   - change to interfacestatistics.h -- just removing an unused member?
>   - change to stats_vnstat.cpp not BSD-specific, upstream it
>   - change to interfacestatistics.cpp not BSD-specific, upstream it
>   - change to configdialog.h -- just removing an unused member?
>   - change to utils.cpp, the usual way to mark something unused in KDE code is
> to use Q_UNUSED inside the function body, not __attribute__ ; this might go in
> an #ifdef
>   - change to bsdbackend.cpp is the meat of this patch. Just looking at the
> patch I'd say "ship it", but there's a couple of thingies:
>     - we should run this by our OpenBSD friends, it may be something we can
> upstream
>     - looks like int len should be size_t len, that would save you a cast or
> two
>     - two-step initialisation of const struct sockaddr_in6* sin6 looks weird
>     - sin6 could be used in the IN6_IS_ADDR_* macros, no?
>
> All-in-all I think that the whole thing should be upstreamed, not done in
> packaging. I can try to do that (it would mean I'd have to find my way through
> KDE's currently byzantine patch-submission schemes *or* act like a bull in a
> china shop and just commit it). Or, if you'd like to upstream it yourself,
> just say so.
>
> [ade]
Oh yeah, sorry... I forgot about the non-IPv6 changes.  I had the same 
upstream patch notion, I'll try to submit it up there after fixing your 
suggestions.

Thanks,
Anthony