[Slackbuilds-users] slackbuild(s) failing on S14.1-64 when compat32 installed
Jim Diamond
Jim.Diamond at acadiau.ca
Thu Jan 2 03:40:37 UTC 2014
On Wed, Jan 1, 2014 at 19:52 (-0600), Erik Hanson wrote:
> On Wed, 1 Jan 2014 18:46:46 -0400
> Jim Diamond <Jim.Diamond at acadiau.ca> wrote:
>> On Wed, Jan 1, 2014 at 11:54 (-0500), xgizzmo wrote:
>>> It breaks Handbreak.
>> Hmmm... so it does.
>> Can anyone comment on whether this is a bug in the HandBreak build
>> process or whether it is a fundamental problem?
> The issue with HandBrake is not relevant, it simply serves as an
> example of why we can't sed it into every script in the repo.
I think that sentence contradicts itself. :-)
To rephrase my question (which perhaps is more for my personal
education than of general interest, I don't know): can anyone familiar
with HandBrake comment on whether they think HandBrake is doing
something weird / unusual / invalid to cause this problem?
(I noticed that if I reran the line which crashes and burns, but moved
the -L/usr/lib to the end of the line, it happily compiles. I haven't
had time to investigate why this is so, but perhaps it is obvious to
someone, and if they wanted to share their wisdom I'd be happy to be
educated on this point.)
> Who wants to test all 3800+ builds to see which ones can't take this
> change?
No one, I'm (pretty) sure.
> For what it's worth, there are a few scripts in the repo that
> already have it. That's probably how this should stay: at the
> discretion of individual maintainers.
That seems reasonable. Particularly if individual maintainers are
aware of the issue. While no-one is likely to go off and test 3800
builds, a maintainer who is aware of the issue *may* be up to testing
his/her slackbuild on S32, S64 and S64+32.
But I guess the bottom line is that adding
LDFLAGS="-L/usr/lib$LIBDIRSUFIX"
is a non-starter.
Cheers.
More information about the SlackBuilds-users
mailing list