Skip to content

Problem with dot chart examples in my Mac

8 messages · David Winsemius, Tom Hopper, Gregory Coats +3 more

#
I?m trying to see the result of the dotchart example, i.e. dotchart(VADeaths, main = "Death Rates in Virginia - 1940").
When I enter that and hit return, a blank Quartz 2 [*] window comes up.  I?ve checked, and the VADeaths data are present.
What should I do to actually see the result of that call?
David
#
I'm seeing the same problem in R 4.0.3 GUI 1.73 Catalina build (7892) on
macOS 11.2 with the following commands:
I'm also seeing the following error:
view: (<NSButton: 0x7fd14b6248b0>) to be less than or equal to 30 but got a
height of 32.000000. This error will be logged once per view in violation.


Plots fine in Rstudio 1.3.1093.

Regards,

Tom


On Sun, Feb 7, 2021 at 1:27 PM Parkhurst, David F. <parkhurs at indiana.edu>
wrote:

  
  
#
Tom Hopper,
On Jan 8, 2021, I first reported that R on Mac gave the error message you reported today.
We R Mac users are all getting this error. It is now a month later. I am not aware of any fix.
Greg Coats

2021-01-07 22:58:42.997 R[8311:37566]
Warning: Expected min height of view: 
(<NSPopoverTouchBarItemButton: 0x7fcb6c592570>) 
to be less than or equal to 30 but got a height of 32.000000. 
This error will be logged once per view in violation.
I'm also seeing the following error:
2021-02-07 14:41:15.950 R[4408:267051] 
Warning: Expected min height of view: 
(<NSButton: 0x7fd14b6248b0>) 
to be less than or equal to 30 but got a height of 32.000000. 
This error will be logged once per view in violation.
#
MacOS 11.2
R version 4.0.3 Patched (2021-01-30 r79912)
XQuartz 2.8.0_beta3

I get the same result. If I use Command Left Arrow/Command Right Arrow, the
plot appears so it is getting to the graphics device's history.
Alternatively, maximizing the XQuartz window causes the plot to appear and
it remains when the window is un-maximized.

David Carlson
On Sun, Feb 7, 2021 at 1:52 PM Tom Hopper <tomhopper at gmail.com> wrote:

            

  
  
#
If you are not aware of the fix, then you are not reading your email...

The NSButton message was reported to be gone in RGui v.7903, on r-help by malone at malonequantitative.com on Jan. 18. 

Current patch versions (https://mac.r-project.org) are at v.7922. Next formal patch release (4.0.4) is on Feb 15, prerelease downloadable from said page.

- Peter Dalgaard

  
    
#
Peter,

Your comment might be tongue-in-cheek, but taking it at face value, I?d like to point out that if there?s a fix and users don?t know about it, that?s a developer communication problem; not a user problem. Most developers deal with this problem by releasing updated versions of their software, which has not happened (current CRAN version is 1.73, which is what I?m using); I know of no developers who have been able to solve this communication problem by publishing notes about beta versions of their software to email lists that have low signal to noise ratios.

Seems like there?s a better way to get updates out if the developers want to avoid users repeatedly pointing out the same bugs.

Thank you,

Tom
#
R is a collaborative project, not a commercial one. You may need to look around or ask questions if you need something special. For the present case, someone was complaining about a problem, for which the solution was in the very same mailing list thread that he had himself contributed to. 

We have mechanisms in place for releasing updates in both source and binary formats. The general procedure is outlined on CRAN. For source releases, there is a daily snapshots link on the front page of CRAN. 

For Mac and Windows binaries, you need to go via the relevant links and for Windows, select links under "Other builds". 

For Mac, the link is on the Mac page, where you need to find the line saying "Information, tools and most recent daily builds of the R GUI, R-patched and R-devel can be found at http://mac.R-project.org/". 

What we do not do is to rush out full releases for any minor problem. We run carefully orchestrated patch releases approximately 4 times per year, with alpha/beta/RC cycles every time, so as to minimize the risk of destabilizing the code.

- Peter D.

  
    
#
More importantly, the message is a spurious message from the OS, not an error, and does not affect the functionality - the FAQ mentions how to disable those if they bother you since R has not control over those. If there is an actual bug to report, then please use the bug reporting system.

Cheers,
Simon