Re: Number format for table results outut from R data.frame/tibble

[John Hendy]

On Fri, Aug 6, 2021 at 2:30 PM Berry, Charles <ccberry@health.ucsd.edu> wrote:
>
> John,
>
> > On Aug 6, 2021, at 8:38 AM, John Hendy <jw.hendy@gmail.com> wrote:
> >
> > Greetings,
> >
> > I'm wondering how to align the results from the R buffer (which I
> > like) vs. the results printed by Org-mode for table results. Here's a
> > toy example:
> >
>
> `tbl_df' objects come with their own print/show method. And it does things 
> like add color and text formats. This is nice in a terminal, but not when you 
> want to display them in org.
>
> Turning off the color might be enough to give you a usable result:
>
> #+begin_src R :session foo :results output drawer
> library(tibble)
> options( cli.num_colors=1 )
> tmp <- tibble(x=1:5, y=x/pi)
> tmp
> #+end_src
>
> #+RESULTS:
> :results:
>
> # A tibble: 5 x 2
>       x     y
>   <int> <dbl>
> 1     1   0.3
> 2     2   0.6
> 3     3   1.
> 4     4   1.
> 5     5   2.
> :end:
>

Interesting, and thanks for taking a look. I don't think I specified
sufficiently that I'm actually aiming for latex/pdf output and
therefore want the org table. I can just add my own header row and get
booktabs formatting free upon export. If I keep with :results output,
the color options trick does remove a bunch of garbage from the
output... but :results value still yields:

#+RESULTS:
| 1 | 0.318309886183791 |
| 2 | 0.636619772367581 |
| 3 | 0.954929658551372 |
| 4 |  1.27323954473516 |
| 5 |  1.59154943091895 |

I was hoping there was a way to either (a) have org post-format or (b)
take the results as printed. I'm not sure on the exact mechanism used
which creates the mismatched output in the session vs. the table. I
get that behind the scenes I still have long, irrational numbers
here... it's just that if org can print them with x decimal places, is
there some way to get Org to intercept *those* values?

John

>
> If this is not close to what you need, I suggest writing your own formatting 
> function. If you have limited emacs-lisp skill, I suggest doing this in R.
>
> HTH,
>
> Chuck