Data Entry Editor

Prof Brian D Ripley wrote:

I would be particularly interested in comments on the new data entry
editor (used by fix/edit on data frames).  This is still work in progress
(in parallel with changes on X11).
[...]
A data entry editor is now implemented.  This is similar to the Unix
one (see ?dataentry) with the additional feature that right-clicking
in a cell turns it into an editable field (entered as a new value
by navigating away from the cell).

data.entry on Unix has lots of quirks.  Thanks for the comments: as I
said, this is work in progress, and based on some already flaky code.

I know. My intention was to point out some of these "quirks".
The Editor will be very useful and already is a _great_ tool for simple
data editing.

Looks like I should tell something like this more gracefully, but my
english too bad...

- Using data.entry() on a existing matrix, list or data.frame, unnamed
columns after that are named "var1", "var2", etc.

I don't see that here.

Example:

 x <- matrix(1:6, nrow=2)
 x

     [,1] [,2] [,3]
[1,]    1    3    5
[2,]    2    4    6

 data.edit(x)
 ## Closing the editor without modifications:
 x

     var1 var2 var3
[1,]    1    3    5
[2,]    2    4    6

- data.entry() does not work with all data.frames, e.g.:

   x <- data.frame(1:2, c("A", "B"))
   data.entry(x)
 Error in de.ncols(sdata) : wrong argument to dataentry

You should use edit and not data.entry with data frames.  That one is wrong
on Unix too.

So have a look at this:
 
 x <- data.frame(1:2, c("A", "B"))
 edit(x) ## same as:  edit.data.frame(x)
 ## Click into [3, 1]

R crashes...