[R] RGL Problem

[Stefan Evert]

> On 24 Nov 2015, at 13:32, Duncan Murdoch <murdoch.duncan at gmail.com> wrote:
> 
>> Perhaps it would make sense always to use the well-known standard XQuartz paths on Mac and only consider other locations if explicitly asked for by the user?
> 
> If rgl is using non-standard features in its makefiles, let me know what they are.  I use GNU make, so it's easy for GNUisms to slip into it.

I didn't have any problems with GNUisms – I only use GNU make anyway.

> We already have special handling for X11 on the Mac.  I don't know why yours isn't working, but it's pretty well-known that homebrew breaks the usual R build system, so I'd blame it on that.

Yes, it turned out that I had a bogus xmkmf in /usr/local/bin, which I had inadvertently installed through homebrew. My fault.

>  If you are using homebrew, you should be prepared to manually configure rgl.  Or you can send me a patch to rgl's configure.ac that works around the differences, and I'll try it out.

My thought was to prefer the known standard location of XQuartz to any automatically discovered X11 configuration on Darwin, i.e. run a test for X headers and libs in this location first, before trying xmkmf.  I guess that normally someone who has xmkmf would want to use this configuration rather than the standard one, so let's quietly forget about this idea.

In case anybody else encounters a similar problem – no X11 found even they know they've installed XQuartz – the first thing you should do is to check "which xmkmf".  If you find one, it is likely to be the cause of the failure.

Best,
Stefan