[R-SIG-Mac] perhaps change working directory in the plist? (was: R.app GUI crashes on startup...)

[Simon Urbanek]

On Jun 19, 2011, at 7:44 PM, Michael C Gilbert wrote:

> 
> On Jun 17, 2011, at 10:41 AM, Michael C Gilbert wrote:
> 
>>> The .RData might be in your R working directory.
>>> If you don't know what your setting was, you can find it in the ~/Library/Preferences/org.R-project.R.plist file.
>> 
>> Thank you, Bob. In fact, I did think to check that. The working directory in my prefs is still set to my home directory: ~.
> 
> This just occurred to me: If there is indeed some kind of problem with the workspace, but since it appears to be the default in the prefs, perhaps I can change it by editing the prefs? The restart R.app and see what happens?
> 
> Bob, can you or someone give me an example of the format (name/value pair) of the entry for the worspace in the plist?
> 

Just moving the prefs aside may do the trick as well ...

Cheers,
S


> 
> 
>>> On Jun 16, 2011, at 7:51 PM, Michael C Gilbert wrote:
>>> 
>>>>>> As my first post on this list, I'll try to be complete but succinct.
>>>>>> 
>>>>>> This is the symptom described in the R for Max OSX FAQ 12.16: "R.app GUI crashes on startup" This started for me on Version 2.13, running on Mac OS 10.6.7. (I had been away from R for a few months, so I can't ID a single trigger.) I followed the instructions in FAQ 12.16. Moved the history file. There was no Rdata workspace to move (not that I could find).
>>>>> 
>>>>> It is .RData (note the dot!) and it is a hidden file. You can move it aside with this (in Terminal):
>>>>> mv ~/.RData ~/myOldData.RData
>>>> 
>>>> Sorry, I didn't make that clear in my haste. I know this. There is no .Rdata file in my home directory.
>>>> 
>>>> 
>>>>> Also note that the history file is configurable so it could be .Rapp.history or .Rhistory - so make sure you checked both (ever since the configurability this issue is in fact rare).
>>>> 
>>>> The .Rhistory file was there. But following the instructions in 12.16, I moved it.
>>>> 
>>>> 
>>>>>> No problem running R on the command line at all. I upgraded to 2.14, to see if it would persist. Same thing, suggesting an environment issue. But I don't know what it is about my environment I should troubleshoot. I cannot get a crash log using the instructions in 12.16, because no crash menu appears. All I get is this in the system log:
>>>>>> 
>>>>>> Jun 16 12:24:48 mcgLaptop-5 com.apple.launchd.peruser.502[285] ([0x0-0x1c71c7].org.R-project.R[19629]): Exited with exit code: 2
>>>>>> 
>>>>>> I apologize if this report is in some way incomplete. Any advice on how to proceed?
>>>>>> 
>>>>> 
>>>>> It is almost entirely incomplete because you don't say what actually happens. We ask you to, please, report exact facts, not your interpretation thereof. If there was a crash, please send the crash report. If there was an error, please send the exact error.
>>>> 
>>>> Here is the only thing I left out, to the best of my judgment: The R.app UI pops up for just a moment. No interactive time at all. Then it's gone. (I assumed that this was what was being described in 12.16, but you're correct: I should not assume.) No crash menu presents itself. The record I have of the event is the one I copied above from my system log. Otherwise, I believe I have included the available details. Unless there is a way to trigger a crash report without the menu appearing.
>>>> 
>>>> — Michael
>>>> 
>>>> _______________________________________________
>>>> R-SIG-Mac mailing list
>>>> R-SIG-Mac at r-project.org
>>>> https://stat.ethz.ch/mailman/listinfo/r-sig-mac
>>> 
>> 
>> _______________________________________________
>> R-SIG-Mac mailing list
>> R-SIG-Mac at r-project.org
>> https://stat.ethz.ch/mailman/listinfo/r-sig-mac
> 
> _______________________________________________
> R-SIG-Mac mailing list
> R-SIG-Mac at r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-mac
> 
>