Need second look at the test suit to find out what tests do I turn on.

I wouldn't be surprised if I never checked passing anything beyond codec to mencoder. I can confirm the problem here with master and that this pr fixes the problem on my system, so +1 from me.

It would be nice if this could get our smoketests working....

I'll add that it doesn't look like our simple smoketest exercises any additional movie generation parameters, so it wouldn't have caught this.

Also, we don't even try to install mencoder on Travis--I'm not sure if that's an oversight or intentional.

Didn't we try to do that at one point, but gave up due to weird problems?

On Thu, Jan 15, 2015 at 1:05 PM, Ryan May notifications@github.com wrote:

I'll add that it doesn't look like our simple smoketest exercises any
additional movie generation parameters, so it wouldn't have caught this.

Also, we don't even try to install mencoder on Travis--I'm not sure if
that's an oversight or intentional.

—
Reply to this email directly or view it on GitHub
#4004 (comment)
.

All I can find is #2678 , which doesn't indicate that it's ever tried. I can throw up a PR and see what happens.

My only memory of that is memcoder was being difficult and went with the low hanging fruit of testing everything else.

Either we fixed something else sense then (doubtful) or I am/was confused (likely) as the tests seem to work now.

Either way, we can discuss anything else about the mencoder tests in #4005, and just leave this about the fixes for commandline args. Does anyone have any objections to this PR?

Probably should be back-ported to 1.4.x

cherry-picked as 09be64e