MathGroup Archive 2012

[Date Index] [Thread Index] [Author Index]

Search the Archive

Re: Wickham-Jones ExtendGraphics for OS X?

  • To: mathgroup at smc.vnet.net
  • Subject: [mg125204] Re: Wickham-Jones ExtendGraphics for OS X?
  • From: Murray Eisenberg <murray at math.umass.edu>
  • Date: Wed, 29 Feb 2012 07:21:06 -0500 (EST)
  • Delivered-to: l-mathgroup@mail-archive0.wolfram.com
  • References: <201202271142.GAA29404@smc.vnet.net>
  • Reply-to: murray at math.umass.edu

I can start each of "image" and "mandelbrot" from the command line.

I can give each a link-name, as you showed.

But I still cannot get any of the ExtendGraphics packages delaunay.m, 
mandelbrot.m, Contour.m to load properly inside Mathematica -- in each 
case getting an error that the specified file "is not a MathLink 
executable".

The value of DYLD_LIBRARY_PATH is empty, and what I read says this 
variable is not supposed to be used in OS X anymore.

(I'm getting well beyond my depth here...)

On 2/27/12 6:42 AM, Patrick Scheibe wrote:
> Hi Murray,
>
> yes, the gcc complains about the code produced by mprep, but this is a
> warning only and no error [1]. On my MacBook I get the same warning,
> when I use gcc, still the MathLink program runs fine.
>
> Did you check to start the the "image" program from the command-line and
> give it a link-name like "blub" and try to install this link by
>
> lnk = Install[LinkConnect["blub"]]
>
> inside Mathematica? If this runs, then your DYLD_LIBRARY_PATH is maybe
> not usable inside Mathematica. In this case you have to create/adjust
> your ~/.MacOSX/environment.plist file so that GUI-programs like
> Mathematica have a correct DYLD_LIBRARY_PATH. See here [2] how to do
> that.
>
> Cheers
> Patrick
>
> [1] You may know the famous joke:
> A guy is standing on the corner of the street smoking one cigarette
> after another. A lady walking by notices him and says
> "Hey, don't you know that those things can kill you? I mean, didn't you
> see the giant warning on the box?!"
> "That's OK" says the guy, puffing casually "I'm a computer programmer"
> "So? What's that got to do with anything?"
> "We don't care about warnings. We only care about errors."
>
> [2]
> https://developer.apple.com/library/mac/#documentation/MacOSX/Conceptual/BPRuntimeConfig/Articles/EnvironmentVars.html
>
> On Sun, 2012-02-26 at 04:22 -0500, Murray Eisenberg wrote:
>> Compiling as you showed worked just fine to produce the MathLink
>> executables 'delaunay' and 'contour.'
>>
>> However, there were compiler warnings when I applied gcc for both
>> 'image' and 'mandelbrot', as follows (where i use "\" for continuation
>> of input because the long command line input won't fit on a single line
>> here:
>>
>>     $ $MLROOT/mprep image.tm>  image.tm.c
>>     $ gcc -o image -I$MLROOT -L$MLROOT -lMLi3 -lstdc++ -framework \
>>         Foundation -lm image.tm.c
>>      image.tm.c: In function =91sendimage=92:
>>      image.tm.c:308: warning: incompatible implicit declaration of
>>         built-in function =91malloc=92
>>      image.tm.c:321: warning: incompatible implicit declaration of
>>         built-in function =91sprintf=92
>>      image.tm.c: At top level:
>>      image.tm.c:344: warning: prototype for =91sendimage=92 follows
>>         non-prototype definition
>>
>>     $ $MLROOT/mprep mandelbrot.tm>  mandelbrot.tm.c
>>     $ gcc -o mandelbrot -I$MLROOT -L$MLROOT -lMLi3 -lstdc++ -framework \
>>         Foundation -lm mandelbrot.tm.c
>>       mandelbrot.tm.c:402: warning: prototype for =91sendimage=92 follows
>>          non-prototype definition
>>
>> And then, of course, when I load either ExtendGraphics`Image` or
>> ExtendGraphics`Mandelbrot`, I get an error:
>>
>>     LinkOpen::linke: Specified file is not a MathLink executable...
>>
>> Can you suggest how to fix this. Or, in the alternative, supply
>> Macintosh OS X versions of the executables 'mandelbrot' and 'image'
>> (compatible with current OS X versions).
>>
>> On 2/20/12 2:44 AM, Patrick Scheibe wrote:
>>> Hi Murray,
>>>
>>> did you try to compile this by yourself? Here on Linux it builds fine
>>> and it does not need any Solaris library. The paths in the "build" file
>>> are outdated. That's why you cannot simply use it. Try it on the
>>> commandline by yourself. What you have to find is the MathLink dev
>>> directory in your Mathematica installation. Here it is under
>>>
>>> /usr/local/Wolfram/Mathematica/8.0/SystemFiles/Links/MathLink/DeveloperKit/Linux-x86-64/
>>>
>>> I assume on your machine it will be under something like
>>>
>>> /Applications/Mathematica.app/SystemFiles/Links/MathLink/DeveloperKit/CompilerAdditions/
>>>
>>> Let's say this path is $MLROOT, then you have to do the following:
>>>
>>> $MLROOT/mprep delaunay.tm>   delaunay.tm.c
>>> gcc -o delaunay -I$MLROOT -L$MLROOT -lMLi3 \
>>>     -lstdc++ -framework Foundation -lm delaunay.tm.c
>>>
>>> (Remove the backslash and input the last two lines as one!)
>>> This should compile fine the delaunay-triangulation and should work for
>>> all the other programms too.
>>>
>>> Cheers
>>> Patrick
>>>
>>>
>>> On Sun, 2012-02-19 at 06:29 -0500, Murray Eisenberg wrote:
>>>> Many years ago, Tom Wickham-Jones wrote a book, "Mathematica Graphics:=
>>
>>>> Techniques&   Applications", which I own. It was accompanied by the
>>>> ExtendGraphics application, which included nice packages for drawing the
>>>> Mandelbrot set and Delaunay set, among other things.
>>>>
>>>> The application is also available at:
>>>>
>>>>      http://library.wolfram.com/infocenter/Books/3753/
>>>>
>>>> But they don't work with Mathematica 8, at least under Mac OS X. The
>>>> .zip download includes Windows .exe binaries; the .tar download includes
>>>> ingredients for compiling binaries (including a "build" file, but the
>>>> build fails under OS X due to missing libraries from Solaris).
>>>>
>>>> Anybody have a version of the requisite binary files that work with
>>>> current Mathematica under OS X 10.7 (Lion)?
>>>>
>>>
>>>
>>>
>>
>> --
>> Murray Eisenberg                     murray at math.umass.edu
>> Mathematics&  Statistics Dept.
>> Lederle Graduate Research Tower      phone 413 549-1020 (H)
>> University of Massachusetts                413 545-2859 (W)
>> 710 North Pleasant Street            fax   413 545-1801
>> Amherst, MA 01003-9305
>>
>>
>
>
>

-- 
Murray Eisenberg                     murray at math.umass.edu
Mathematics & Statistics Dept.
Lederle Graduate Research Tower      phone 413 549-1020 (H)
University of Massachusetts                413 545-2859 (W)
710 North Pleasant Street            fax   413 545-1801
Amherst, MA 01003-9305



  • Prev by Date: Re: Extensive replacement of trigonometric functions
  • Next by Date: Re: Position in the importing loop from a Web
  • Previous by thread: Re: Wickham-Jones ExtendGraphics for OS X?
  • Next by thread: Equation solving problem