octave-maintainers
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Testers needed for an Octave-3.6.1 app (X86_64 only) bundle for MacO


From: Ben Abbott
Subject: Re: Testers needed for an Octave-3.6.1 app (X86_64 only) bundle for MacOS
Date: Sat, 03 Mar 2012 18:43:11 -0500

On Mar 3, 2012, at 6:25 PM, Thomas Yengst <address@hidden> wrote:

> On Mar 3, 2012, at 2:01 PM, Ben Abbott wrote:
>> 
>> On Mar 3, 2012, at 4:51 PM, Thomas Yengst wrote:
>> 
>>> On Mar 3, 2012, at 1:36 PM, Ben Abbott wrote:
>>> 
>>>> On Mar 3, 2012, at 4:28 PM, Thomas Yengst wrote:
>>>> 
>>>>>>> Ben,
>>>>>>> 
>>>>>>> Sorry - I misinterpreted your original question. Here is 'path' from 
>>>>>>> the octave prompt:
>>>>>>> 
>>>>>>> /temp/Octave-3.6.1.app/Contents/Resources/share/octave/3.6.1/m/miscellaneous
>>>>>>> /temp/Octave-3.6.1.app/Contents/Resources/share/octave/3.6.1/m/optimization
>>>>>>> /temp/Octave-3.6.1.app/Contents/Resources/share/octave/3.6.1/m/path
>>>>>>> /temp/Octave-3.6.1.app/Contents/Resources/share/octave/3.6.1/m/pkg
>>>>>>> /temp/Octave-3.6.1.app/Contents/Resources/share/octave/3.6.1/m/plot
>>>>>>> /temp/Octave-3.6.1.app/Contents/Resources/share/octave/3.6.1/m/polynomial
>>>>>>> /temp/Octave-3.6.1.app/Contents/Resources/share/octave/3.6.1/m/prefs
>>>>>>> /temp/Octave-3.6.1.app/Contents/Resources/share/octave/3.6.1/m/set
>>>>>>> /temp/Octave-3.6.1.app/Contents/Resources/share/octave/3.6.1/m/signal
>>>>>>> /temp/Octave-3.6.1.app/Contents/Resources/share/octave/3.6.1/m/sparse
>>>>>>> /temp/Octave-3.6.1.app/Contents/Resources/share/octave/3.6.1/m/specfun
>>>>>>> /temp/Octave-3.6.1.app/Contents/Resources/share/octave/3.6.1/m/special-matrix
>>>>>>> /temp/Octave-3.6.1.app/Contents/Resources/share/octave/3.6.1/m/startup
>>>>>>> /temp/Octave-3.6.1.app/Contents/Resources/share/octave/3.6.1/m/statistics
>>>>>>> /temp/Octave-3.6.1.app/Contents/Resources/share/octave/3.6.1/m/statistics/base
>>>>>>> /temp/Octave-3.6.1.app/Contents/Resources/share/octave/3.6.1/m/statistics/distributions
>>>>>>> /temp/Octave-3.6.1.app/Contents/Resources/share/octave/3.6.1/m/statistics/models
>>>>>>> /temp/Octave-3.6.1.app/Contents/Resources/share/octave/3.6.1/m/statistics/tests
>>>>>>> /temp/Octave-3.6.1.app/Contents/Resources/share/octave/3.6.1/m/strings
>>>>>>> /temp/Octave-3.6.1.app/Contents/Resources/share/octave/3.6.1/m/testfun
>>>>>>> /temp/Octave-3.6.1.app/Contents/Resources/share/octave/3.6.1/m/time
>>>>>>> 
>>>>>>> Tom
>>>>>> 
>>>>>> Opps, my bad, I wanted to check to see if lookup.oct was in the 
>>>>>> loadpath. Unfortunately, I don't see exactly how to do that.
>>>>>> 
>>>>>> You can check to see of lookup.oct is in your  load path.
>>>>>> 
>>>>>>    file_in_loadpath lookup.oct
>>>>>> 
>>>>>> You can also check the exec path.
>>>>>> 
>>>>>>    strsplit (EXEC_PATH, pathsep)
>>>>>> 
>>>>>> I get the results below.
>>>>>> 
>>>>>> file_in_loadpath lookup.oct
>>>>>> ans = 
>>>>>> /Users/bpabbott/Desktop/Octave-3.6.1.app/Contents/Resources/lib/octave/3.6.1/oct/x86_64-apple-darwin11.3.0/lookup.oct
>>>>>> 
>>>>>> strsplit (EXEC_PATH, pathsep)
>>>>>> ans = 
>>>>>> {
>>>>>> [1,1] = 
>>>>>> /Users/bpabbott/Desktop/Octave-3.6.1.app/Contents/Resources/libexec/octave/3.6.1/site/exec/x86_64-apple-darwin11.3.0
>>>>>> [1,2] = 
>>>>>> /Users/bpabbott/Desktop/Octave-3.6.1.app/Contents/Resources/libexec/octave/api-v48+/site/exec/x86_64-apple-darwin11.3.0
>>>>>> [1,3] = 
>>>>>> /Users/bpabbott/Desktop/Octave-3.6.1.app/Contents/Resources/libexec/octave/site/exec/x86_64-apple-darwin11.3.0
>>>>>> [1,4] = 
>>>>>> /Users/bpabbott/Desktop/Octave-3.6.1.app/Contents/Resources/libexec/octave/3.6.1/exec/x86_64-apple-darwin11.3.0
>>>>>> [1,5] = /Users/bpabbott/Desktop/Octave-3.6.1.app/Contents/Resources/bin
>>>>>> }
>>>>>> 
>>>>>> As you have an .octaverc file, you should check that there are no 
>>>>>> contents that might break the app bundle by modifying the core loadpath, 
>>>>>> exec path, or core m-file path.
>>>>> 
>>>>> file_in_loadpath lookup.oct
>>>>>    ==> returns NULL
>>>>> 
>>>>> strsplit (EXEC_PATH, pathsep)
>>>>>    ==> returns errors, which I guess makes sense since lookup.oct is not 
>>>>> in the load path
>>>>> 
>>>>> EXEC_PATH = 
>>>>> /temp/Octave-3.6.1.app/Contents/Resources/libexec/octave/3.6.1/site/exec/x86_64-apple-darwin10.8.0:/temp/Octave-3.6.1.app/Contents/Resources/libexec/octave/api-v48+/site/exec/x86_64-apple-darwin10.8.0:/temp/Octave-3.6.1.app/Contents/Resources/libexec/octave/site/exec/x86_64-apple-darwin10.8.0:/temp/Octave-3.6.1.app/Contents/Resources/libexec/octave/3.6.1/exec/x86_64-apple-darwin10.8.0:/temp/Octave-3.6.1.app/Contents/Resources/bin
>>>>>    ==> so this means that strsplit is not working because it's not in the 
>>>>> EXEC_PATH either.
>>>>> 
>>>>> I moved my .octaverc out of the way and still got the same results. 
>>>>> Doesn't look like I get lookup.oct in the EXEC_PATH, which I think causes 
>>>>> all the other problems.
>>>>> 
>>>>> When I type restoredefaultpath; path, I get the same result as you do Ben.
>>>>> 
>>>>> weird.
>>>>> 
>>>>> Tom
>>>> 
>>>> Ok. After you type restoredefaultpath, then lookup and strsplit should be 
>>>> working, correct?
>>>> 
>>>> After starting Octave, and if it is still misbehaving, please check that 
>>>> the ~./octaverc is not present. If it is, then quit octave and then rename 
>>>> or delete it.
>>>> 
>>>> The next time you start Octave everything should be ok. If it is not, 
>>>> check that ~/.octaverc isn't present. If it is, you'll need to figure out 
>>>> why (do you have it open in an editor or something?).
>>>> 
>>>> Ben
>>>> 
>>> no joy - I have removed ~/.octaverc - see below. I've closed every open 
>>> program - correct that. I rebooted just for giggles.
>>> 
>>> ABZ117:~ try25996$ ls -al .oct*
>>> -rw-------  1 try25996  staff   9697 Mar  3 13:37 .octave_hist
>>> -rw-r--r--  1 try25996  staff  12763 Feb 29 20:35 .octave_packages
>>> -rw-r--r--@ 1 try25996  staff   1123 Mar  3 13:19 .octaverc-orig
>>> ABZ117:~ try25996$ 
>>> 
>>> I have two versions of the octaverc script - one in my Macports directory
>>> /opt/local/share/octave/site/m/startup/octaverc
>>> calls /opt/local/share/octave/3.6.1/m/startup/octaverc
>>> 
>>> and 
>>> /temp/Octave-3.6.1.app/Contents/Resources/share/octave/site/m/startup/octaverc
>>> calls 
>>> /temp/Octave-3.6.1.app/Contents/Resources/share/octave/3.6.1/m/startup/octaverc
>>> 
>>> I don't know where octave.app is grabbing path information from.
>>> 
>>> Tom
>> 
>> Very strange.
>> 
>> The problem is either esoteric, or the result of some silly bug.
>> 
>> Does the problem persist if you place Octave-3.6.1.app on your Desktop?
>> 
>> What if you deactivate octave-3.6.1 you installed using MacPorts ?
>> 
>>    sudo port deactivate octave
>> 
>> That assume the port name is "octave". You can verify the port name by ...
>> 
>>    $ port installed | grep -i octave
>>      octave-local @3.7.0+_1+atlas+fltk+gcc45
>>      octave-local @3.7.0+_2+atlas+fltk+gcc45
>>      octave36 @3.6.1_0+atlas+fltk+gcc44 (active)
>> 
>> As you can see, I also have a 3.6.1 version installed and active. I hadn't 
>> realized that. I'll remove it to be sure it isn't interfering with the app 
>> bundle on my end. That will have to wait until after I've finished a new 
>> build. I report back later.
>> 
>> Ben
>> 
>> 
>> 
> Got it - deactivating the octave-devel macports did it. I'm not a macports 
> expert, so I don't know what path magic goes on with a deactivate. Once 
> deactivated, I get
> 
> octave-3.6.1:1> strsplit (EXEC_PATH, pathsep)
> ans = 
> {
>  [1,1] = 
> /temp/Octave-3.6.1.app/Contents/Resources/libexec/octave/3.6.1/site/exec/x86_64-apple-darwin11.3.0
>  [1,2] = 
> /temp/Octave-3.6.1.app/Contents/Resources/libexec/octave/api-v48+/site/exec/x86_64-apple-darwin11.3.0
>  [1,3] = 
> /temp/Octave-3.6.1.app/Contents/Resources/libexec/octave/site/exec/x86_64-apple-darwin11.3.0
>  [1,4] = 
> /temp/Octave-3.6.1.app/Contents/Resources/libexec/octave/3.6.1/exec/x86_64-apple-darwin11.3.0
>  [1,5] = /temp/Octave-3.6.1.app/Contents/Resources/bin
> }
> octave-3.6.1:2> 
> 
> Tom

Great! 

Ben


reply via email to

[Prev in Thread] Current Thread [Next in Thread]