[ESS] Bugs (?) is ESS 12.09 (under Windows)

[Jarmo Hurri]

Greetings.

>   >    ?function
>   >    followed by Enter hangs Emacs.
> Indeed, this one was a bit of an unfinished business in 12.09. It has
> been solved in the development version.

Ok, thanks.

>   > 2. Also, typing
>   >    ls (
>   >    followed by Enter (that is, leaving a command uncompleted), and then
>   >    pressing M-p when the "+" prompt is shown hangs Emacs.
>
> Nobody reported this one before. I cannot reproduce it on linux. What is
> M-p bound to in your case?

Sorry, it is bound to comint-previous-input (as indicated by the reply
by Josh below).

>   > The third problem is that under some (not so uncommon conditions,
>   > because I have seen this happen several times), sending a piece of R
>   > code from an .R-file to be evaluated via C-c C-b leaves the cursor in
>   > the R interaction buffer in the middle of the buffer, not at the bottom
>   > where the prompt is. This is very annoying, because you always need to
>   > go to the end of buffer "manually" (the error seems to persist if it
>   > happens once). But I can not reproduce this right now.
>
> Well, comint (the mode on which iESS is based) is designed to keep the
> cursor where it is. Particularly, if the cursor is after the prompt it
> keeps it there and inserts the output before the point.
>
> So my guess is that you just left you point somewhere in the middle of
> the process buffer before evaluating commands from a script buffer.

I don't think this is the case. The thing I have described has happened
to a large number of students, and we have tried to move the point to
the prompt, but the problem persists. That is, even when the point is at
the prompt, C-c C-b takes the cursor (point) to the middle of the
buffer. But I think the problem begins to appear when the point is in
the middle of the buffer.

I will try to reproduce the last problem at some time. (Have other work
on the table right now.)

All the best,

Jarmo