Re: [R-SIG-Mac] Graphics in R, version 4.3.2, does not work well in MacOS

2024-02-16 Thread Michael Hall
It also works for me on both my current m3 and also my old x86.

I also always assumed Quartz meant XQuartz. However with neither architecture 
did it seem to start XQuartz in order to render. 

Sonoma on both machines. M3 = R 4.3.2. X86 = R 4.3.1
___
R-SIG-Mac mailing list
R-SIG-Mac@r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-mac


Re: [R-SIG-Mac] R-SIG-Mac Digest, Vol 241, Issue 15 - symbol not found in flat namespace '___kmpc_critical'

2023-04-29 Thread Michael Hall


> On Apr 29, 2023, at 5:32 PM, r-sig-mac-requ...@r-project.org wrote:
> 
> symbol not found in flat namespace ‘___kmpc_critical’

Seems to suggest a workaround…

https://github.com/glmmTMB/glmmTMB/issues/588
Trouble installing with openMP support: Symbol not found: ___kmpc_critical · 
Issue #588 · glmmTMB/glmmTMB
github.com





___
R-SIG-Mac mailing list
R-SIG-Mac@r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-mac


Re: [R-SIG-Mac] R , RCPP unsing MACOS ventura

2023-03-14 Thread Michael Hall



> On Mar 14, 2023, at 3:50 PM, Simon Urbanek  
> wrote:
> 
> Well, what I saw in the searches what that it was a problem with Homebrew 
> tools breaking on that flag. Again, it's all speculation, because no one 
> posted the actual issue ... so it's anyone's guess what this is all about 
> until someone actually posts the details I asked for - I don't see that 
> problem myself, but I don't use Homebrew (at least not on the default PATH).
> 
> Cheers,
> Simon

As you said earlier.

> The error suggests your are using compilers that don't support your system.

It doesn’t seem to recognize a 13.2 system. This could well be because it is 
trying to use a homebrew gcc that isn’t playing well with the Xcode command 
line tools.

The xcode-select -version that I showed earlier was supposed to indicate what 
version of the command line tools you are using no matter with what Xcode.app 
although I’m not sure exactly how you are supposed to correspond the CLT 
version to the OS one.

gcc —version should I think tell you if you have a homebrew or Apple one.

If homebrew is the culprit following the homebrew suggestion of uninstalling 
and reinstalling CLT would still seem to me to be worth the attempt.

If something is hardcoded in PATH messing up search orders that would be a 
different issue. 

Uninstalling homebrew as I think you suggested seems to be throwing out the 
baby, the bath water and the kitchen sink. There should be a less drastic 
solution.
 
___
R-SIG-Mac mailing list
R-SIG-Mac@r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-mac


Re: [R-SIG-Mac] R , RCPP unsing MACOS ventura

2023-03-14 Thread Michael Hall



> On Mar 14, 2023, at 3:32 PM, Simon Urbanek  
> wrote:
> 
> If that is the case, can you actually post what the issue is? The original 
> post had no details and did not follow up when asked for details. When 
> googling the issue seems to arise in Homebrew, not the Apple tools so as I 
> noted removing Homebrew should work.
> 
> Cheers,
> Simon
> 
> 
> 
>> On 15/03/2023, at 12:52 AM, Michael Hall  wrote:
>> 
>> 
>> 
>>> On Mar 14, 2023, at 6:00 AM, r-sig-mac-requ...@r-project.org wrote:
>>> 
>>> [R-SIG-Mac] R , RCPP unsing MACOS ventura
>> 
>> Google shows others have had the issue.
>> 
>> Suggestions include...
>> 
>> Uninstall and reinstall command line tools.
>> 
>> ——
>> 
>> ...following the wisdom of the "homebrew doctor" solves it, i.e. remove and 
>> reinstall:
>> 
>> sudo rm -rf /Library/Developer/CommandLineTools
>> sudo xcode-select —install
>> 
>> ———
>> 
>> 
>> Rebuild with export SYSTEM_VERSION_COMPAT=1 
>> 
>> Searched with…
>> : error: unknown value ‘13.2.0’ of ‘-mmacosx-version-min’

The general suggestion seemed to be for some reason to remove and reinstall the 
command line tools. The example I posted was one where it was the home-brew 
doctor that made that suggestion. Not that homebrew itself was the problem.

The 
>> : error: unknown value ‘13.2.0’ of ‘-mmacosx-version-min’

Seemed to be what the underlying issue was? and was what I searched on

___
R-SIG-Mac mailing list
R-SIG-Mac@r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-mac


Re: [R-SIG-Mac] R , RCPP unsing MACOS ventura

2023-03-14 Thread Michael Hall



> 
> I actually do not use Rcpp, I was just curious to know whether it works on my 
> machine or not.  Apparently it does.
> 


Me either. Fwiw.  

xcode-select --version
xcode-select version 2396.

Also Ventura 13.2.1

> library(Rcpp)
> evalCpp( "1+1")
[1] 2
::sprintf(buff, "%02x", from);
  ^
/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/usr/include/stdio.h:188:1:
 note: 'sprintf' has been explicitly marked deprecated here
__deprecated_msg("This function is provided for compatibility reasons only.  
Due to security concerns inherent in the design of sprintf(3), it is highly 
recommended that you use snprintf(3) instead.")
^
/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/usr/include/sys/cdefs.h:215:48:
 note: expanded from macro '__deprecated_msg'
#define __deprecated_msg(_msg) __attribute__((__deprecated__(_msg)))
  ^
1 warning generated.
> 

___
R-SIG-Mac mailing list
R-SIG-Mac@r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-mac


Re: [R-SIG-Mac] R , RCPP unsing MACOS ventura

2023-03-14 Thread Michael Hall



> On Mar 14, 2023, at 7:40 AM, Calboli Federico (LUKE) 
>  wrote:
> 
> I know that posting a makevars is somehow iffy, but I installed GCC with 
> homebrew and my
>  
> ~/.R/Makevars
>  

It sounds like it is worth a try. However, I’m not sure it’s clear your OS and 
other versions match the original posters?

Did you build for the same issue?


[[alternative HTML version deleted]]

___
R-SIG-Mac mailing list
R-SIG-Mac@r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-mac


Re: [R-SIG-Mac] R , RCPP unsing MACOS ventura

2023-03-14 Thread Michael Hall



> On Mar 14, 2023, at 6:00 AM, r-sig-mac-requ...@r-project.org wrote:
> 
> [R-SIG-Mac] R , RCPP unsing MACOS ventura

Google shows others have had the issue.

Suggestions include...

Uninstall and reinstall command line tools.

——

...following the wisdom of the "homebrew doctor" solves it, i.e. remove and 
reinstall:

sudo rm -rf /Library/Developer/CommandLineTools
sudo xcode-select —install

———


Rebuild with export SYSTEM_VERSION_COMPAT=1 

Searched with…
: error: unknown value ‘13.2.0’ of ‘-mmacosx-version-min’




[[alternative HTML version deleted]]

___
R-SIG-Mac mailing list
R-SIG-Mac@r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-mac


Re: [R-SIG-Mac] Please test R 4.0.4 RC

2021-02-13 Thread Michael Hall



> On Feb 13, 2021, at 8:30 PM, Simon Urbanek  
> wrote:
> 
> Michael,
> Thanks, but please don't post 3rd party links that are just one-liners with 
> extra advertising, especially when they have been already mentioned.
> Cheers,
> Simon

I think my link expanded some on the original answer which it didn’t sound like 
worked.
‘
It included additionally…

sudo xcode-select —reset

and…

If even that fails, go to the Apple developer download section and download 
Xcode manually.

But I did just google it and have no personal experience with the error. Often 
though just google will find you someone else who has dealt with the issue.

___
R-SIG-Mac mailing list
R-SIG-Mac@r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-mac


Re: [R-SIG-Mac] Please test R 4.0.4 RC

2021-02-13 Thread Michael Hall



> On Feb 13, 2021, at 5:41 PM, r-sig-mac-requ...@r-project.org wrote:
> 
> xcrun: error: invalid active developer path 
> (/Library/Developer/CommandLineTools), missing xcrun at: 
> /Library/Developer/CommandLineTools/usr/bin/xcrun

https://ma.ttias.be/mac-os-xcrun-error-invalid-active-developer-path-missing-xcrun/


[[alternative HTML version deleted]]

___
R-SIG-Mac mailing list
R-SIG-Mac@r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-mac


Re: [R-SIG-Mac] R-SIG-Mac Digest, Vol 205, Issue 13

2020-04-02 Thread Michael Hall



> On Apr 2, 2020, at 6:10 AM, r-sig-mac-requ...@r-project.org wrote:
> 
>> Is there a way of (only) listing all user installed (additional)
>> packages, ie not the ones that come with R?

https://www.r-bloggers.com/list-of-user-installed-r-packages-and-their-versions/
 


Sys.getenv()
…
R_LIBS_USER ~/Library/R/3.6/library
…

list.files('~/Library/R/3.6/library')
 [1] "alphavantager" "BBmisc""brew"  "C50"  
 [5] "checkmate" "clisymbols""coda"  "commonmark”   
...
[[alternative HTML version deleted]]

___
R-SIG-Mac mailing list
R-SIG-Mac@r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-mac


Re: [R-SIG-Mac] .libPaths

2017-12-25 Thread Michael Hall

> On Dec 25, 2017, at 5:09 PM, Michael Hall <mik3h...@gmail.com> wrote:
> 
> 
>> On Dec 25, 2017, at 5:01 PM, David Winsemius <dwinsem...@comcast.net> wrote:
>> 
>> 
>>> 
>> 
>> My guess is that you are running the R.app GUI and have not unchecked the 
>> default settings in the Startup panel of Preferences:
> 
> Correct that this from the app. It is also consistent with what RStudio shows 
> for .libPaths.
> The preferences item, however, is unchecked.
> I just tried setting it. Exiting the app. Going back in, unchecking it. 
> Exiting the app. 
> After going back in the user lib was still there. 

What did work was this from Terminal…

rm -r /Users/mjh/Library/R/3.4/library

apparently eliminating it from .libPaths with this…
paths <- paths[dir.exists(paths)]
?

I had tried removing everything from the directory at one point but then I did 
a mkdir for the directory itself thinking it would do no harm at least having 
it there.
But apparently it can be a problem if it even exists it then seems to become 
the default period?

___
R-SIG-Mac mailing list
R-SIG-Mac@r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-mac

Re: [R-SIG-Mac] istate CRAN mirror not working - unable to use .Rprofile to change

2017-06-08 Thread Michael Hall

> On Jun 8, 2017, at 5:22 PM, Duncan Murdoch  wrote:
> 
> 
> The documentation about .Rprofile is documentation about R.  R.app is a 
> separate project that acts as a front end to R.  Front ends can change things 
> that R does.  R.app uses its own Mac OS ways of storing persistent 
> configuration information.

OK, seems a reasonable split in support. I’m not sure how far you can get away 
from what R does though.
 
> 
>> 
>> But again, for my purposes yes, the preference is fine to change the default 
>> mirror and get me working normally. Thanks again.
>> 
>> I am still a little curious if there are any thoughts on why the iastate 
>> mirror stopped working?
> 
> You would likely have to figure out who was running it, and ask them. Mirrors 
> are coordinated by CRAN, but are run by the local people who run them.

I think for one mirror I did notice a contact person name, but I’m not sure I 
saw any contact information. Iowa is close but not close enough to go and try 
to track a responsible individual down.

Thanks again

___
R-SIG-Mac mailing list
R-SIG-Mac@r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-mac

Re: [R-SIG-Mac] istate CRAN mirror not working - unable to use .Rprofile to change

2017-06-08 Thread Michael Hall

> On Jun 8, 2017, at 5:53 AM, Duncan Murdoch  wrote:
> 
> If you are using R.app, you can also set the mirror in the Preferences menu.  
> Perhaps this setting overrides the .Rprofile setting.
> 

I was using the app and wasn’t aware of the preference, so thanks.

There was no .Rprofile before I added one. 

I wasn’t sure what the app was basing this on. I didn’t see a Rprofile. 
site anywhere in the framework either. Just a csv file 
that seemed to include all the mirrors turned up in a search.

Whether or not being unable to do this with .Rprofile is a concern I don’t 
know. Looking again at help(Startup) it shows doing this as an example for 
Rprofile.site and not .Rprofile. I haven’t tried with .Rprofile.site.
But I can’t exactly say it is documented that this should work. Although you 
might assume that you should be able to do in .Rprofile whatever might be done 
in Rprofile.site.

But again, for my purposes yes, the preference is fine to change the default 
mirror and get me working normally. Thanks again. 

I am still a little curious if there are any thoughts on why the iastate mirror 
stopped working? 


[[alternative HTML version deleted]]

___
R-SIG-Mac mailing list
R-SIG-Mac@r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-mac

[R-SIG-Mac] istate CRAN mirror not working - unable to use .Rprofile to change

2017-06-08 Thread Michael Hall
I started getting errors installing packages using my default repository 
instate…

Warning: unable to access index for repository 
http://streaming.stat.iastate.edu/CRAN/src/contrib:
  cannot open URL 'http://streaming.stat.iastate.edu/CRAN/src/contrib/PACKAGES'
Warning: unable to access index for repository 
http://streaming.stat.iastate.edu/CRAN/bin/macosx/el-capitan/contrib/3.4:
  cannot open URL 
'http://streaming.stat.iastate.edu/CRAN/bin/macosx/el-capitan/contrib/3.4/PACKAGES'
Warning message:
package ‘RWeka’ is not available (for R version 3.4.0) 

Googling, I was able to do this using a different mirror selected with 
chooseCRANmirror(), I chose Kansas.

I googled how to make this the new default and was pointed to .Rprofile.
This doesn’t seem able to do this, with…

message(".Rprofile")
local({r <- getOption("repos")
  r["CRAN"] <- "https://rweb.crmda.ku.edu/cran;
  options(repos=r)})

I get…

Type 'demo()' for some demos, 'help()' for on-line help, or
'help.start()' for an HTML browser interface to help.
Type 'q()' to quit R.

.Rprofile
[R.app GUI 1.70 (7338) x86_64-apple-darwin15.6.0]

[History restored from /Users/mjh/.Rapp.history]

Which shows the “.Rprofile” message indicating it has run.
However…

> r <- getOption("repos")
> r
CRAN 
"http://streaming.stat.iastate.edu/CRAN; 

Still shows iastate.
Then doing…

> source(".Rprofile")
.Rprofile
> r <- getOption("repos")
> r
CRAN 
"https://rweb.crmda.ku.edu/cran” 

so the .Rprofile does work. R must be changing it to iastate after the profile 
is run?
I had googled a mailing list thread where it was said .Rprofile wasn’t working 
on OS X. It was suggested in there that this be posted to this list. I checked 
the archives for the last couple months but didn’t see any mention.

Is this a problem that needs addressing? Or has a workaround besides sourcing 
.Rprofile whenever I want to install packages?

 


[[alternative HTML version deleted]]

___
R-SIG-Mac mailing list
R-SIG-Mac@r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-mac