[ESS] ess-julia.el

Jay Kerns gjkernsysu at gmail.com
Sat Feb 23 15:46:30 CET 2013


Dear Vitalie,

On Sat, Feb 23, 2013 at 8:48 AM, Vitalie Spinu <spinuvit at gmail.com> wrote:
>
> Hm, I thought you have done with that and I was so much in a hurry that
> I have already merged it :)
>
> Thanks.

You are right - thanks goes to you.
>
>
> Echoing in process buffer is a bit different than "echoing" when sending
> code form a script buffer. The first one is just a logical indicating
> whether the process actually echoes it's input. For example,
>
>    > print("foo")
>
> would produce in *R* buffer
>
>    > print("foo")
>    print("foo")
>    [1] "foo"
>
> There are some processes that do that, for example S+. Julia and R
> don't, so there is (in principle) no need for that. This being a main
> idea, the story is not that simple with ESS, which intentionally sets
> comint-process-echoes dynamically depending on the value of
> ess-eval-visibly. This is in order to match the behavior that you get
> when evaluating directly in *R* buffer, and when sending code from
> script buffers.
>
> Your concern has to do with something else though. Correct me if I am
> wrong, but it is the thing that we call "(in)visible evaluation" that
> bothers you. That is, the evaluated code from the script buffer is not
> echoed in the inferior buffer.
>
> Currently ess-julia completely ignores ess-eval-visibly and that is
> historical. Back in May, julia was not supporting accumulation of
> commands. So if you send an incomplete commmand, julia would not wait
> for the terminal input and would throw an error.
>
> I think this has been solved on julia side in meanwhile and it is quite
> likely that it is possible to make ess-julia behave pretty much like
> ess-R does. I will look into it once my installation of julia is
> complete.
>

Thanks for the insightful reply; yes, the bulk of my comments were a
long-winded way to say, "I'm trying to get julia to act the same way R
does."

I appreciate the time you've taken to look into this.  I had a feeling
there had to be a reason why 2194 was set the way it was.  :-)

-- 
Jay



More information about the ESS-help mailing list