My thought is that it's a packaging problem.

As it is, the GRASS source should run on El Cap, with a simple 
configure-make-install and no packaging.  It will use external libraries in 
their installed locations, which should work in El Cap.

The problem comes when some of those external libraries are bundled inside the 
app without any modification (ie wxPython).  Those libraries still have their 
embedded install locations which would normally break other software (and is 
not related to the El Capitan problem), but GRASS sets DYLD_LIBRARY_PATH to get 
around this (I don't know the original reason for this, but it makes the 
package moveable/renamable).  DYLD_LIBRARY_PATH is where the El Cap problem 
comes in - as part of new security features the system now ignores this.

2 possible solutions, both on the packaging end:

- compile the external libraries (before compiling GRASS) as if they are 
embedded in the GRASS app.  That is, configure the prefix to be inside the 
to-be-installed GRASS app.

- post-process the libraries when bundled to update the embedded library paths.

The first is simple, but has a couple drawbacks: the app is not moveable 
(minor) because it still depends on the full paths embedded in the libraries 
and GRASS, and this doesn't work with premade binaries of the libraries (like 
wxPython).

The second is harder, but can be done to any bundled library or tool, and the 
paths can be made relative to make the app movable/renamable.  Though the 
movablility part would be best supplemented by changes in the GRASS source 
makefiles to make GRASS have all relative paths (this is what I was originally 
looking at, that stalled me).

At a minimum, and the simplest, full paths should be changed in the bundled 
external libraries, by either or both methods.

> On Dec 4, 2015, at 2:09 PM, Brian Miles <[email protected]> wrote:
> 
> Michael, William,
> 
> I'd like to take a look at this to see if I can help.  I'm not an expert in 
> makefiles, or linkers, but I'm willing to put some time in to help move the 
> ball forward.  Is there a synopsis of the problem so I don't have to start 
> from square one?
> 
> Thanks,
> 
> Brian
> 
>> On Dec 4, 2015, at 2:47 PM, Michael Barton <[email protected]> wrote:
>> 
>> This is still not resolved. William Kyngesburye has suggested some ways 
>> forward, but we are not yet sure exactly what causes the failure. Help is 
>> appreciated.
>> 
>> Michael
>> ____________________
>> C. Michael Barton
>> Director, Center for Social Dynamics & Complexity 
>> Professor of Anthropology, School of Human Evolution & Social Change
>> Head, Graduate Faculty in Complex Adaptive Systems Science
>> Arizona State University
>> 
>> voice:  480-965-6262 (SHESC), 480-965-8130/727-9746 (CSDC)
>> fax: 480-965-7671 (SHESC),  480-727-0709 (CSDC)
>> www: http://www.public.asu.edu/~cmbarton, http://csdc.asu.edu
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>>> On Dec 4, 2015, at 12:36 PM, Brian Miles <[email protected]> wrote:
>>> 
>>> Michael,
>>> 
>>> Just wondering what the status is on this.  Is there a ticket describing 
>>> the problems with OS X 10.11 and GRASS, and hopefully tracking the progress 
>>> to a fix?  I tried searching the tickets (http://trac.osgeo.org/grass/), 
>>> but wasn't able to filter out enough false positives.
>>> 
>>> Thanks,
>>> 
>>> Brian Miles
>>> Research Scientist
>>> Institute for the Environment
>>> University of North Carolina at Chapel Hill
>>> 
>>>> On Oct 8, 2015, at 8:14 PM, [email protected] wrote:
>>>> 
>>>> From: Michael Barton <[email protected]>
>>>> Date: October 8, 2015 at 8:07:58 PM EDT
>>>> To: GRASS developers grass-developers <[email protected]>, 
>>>> "grass-user grass-user" <[email protected]>
>>>> Subject: [GRASS-dev] WARNING TO MAC USERS - WAIT UPDATING TO EL CAPITAN 
>>>> (OS X 10.11)
>>>> 
>>>> 
>>>> I've had reports of GRASS not running for users updating to El Capitan. 
>>>> There is a possible workaround, but it is cumbersome and not guaranteed to 
>>>> work. I recommend that you do wait to update while we try to solve this. 
>>>> 
>>>> I've also heard that El Capitan breaks R.
>>>> 
>>>> Michael
>>>> ____________________
>>>> C. Michael Barton
>>>> Director, Center for Social Dynamics & Complexity 
>>>> Professor of Anthropology, School of Human Evolution & Social Change
>>>> Head, Graduate Faculty in Complex Adaptive Systems Science
>>>> Arizona State University
>>>> 
>>>> voice:  480-965-6262 (SHESC), 480-965-8130/727-9746 (CSDC)
>>>> fax: 480-965-7671 (SHESC),  480-727-0709 (CSDC)
>>>> www: http://www.public.asu.edu/~cmbarton, http://csdc.asu.edu
>>> 
>> 
> 

-----
William Kyngesburye <kyngchaos*at*kyngchaos*dot*com>
http://www.kyngchaos.com/

All generalizations are dangerous, even this one.


_______________________________________________
grass-dev mailing list
[email protected]
http://lists.osgeo.org/mailman/listinfo/grass-dev

Reply via email to