[ESS] Emacs taking over CPU

[Vitalie Spinu]

I will document this properly before the release. Will post here as well.

  Vitalie

 >>> Ross Boylan on Mon, 16 Sep 2013 16:17:31 -0700 wrote:

 > On Mon, 2013-09-16 at 15:48 -0700, Vitalie Spinu wrote:
 >> >> Ross Boylan <ross at biostat.ucsf.edu>
 >> >> on Mon, 16 Sep 2013 15:31:12 -0700 wrote:
 >> 
 >> [...]
 >> 
 >> 
 >> > If I go C-c C-c for this block it hangs up emacs while evaluating
 >> > something_slow().  C-g gets emacs back and also shows intermediate
 >> > output from something_slow().  But the second date() is not executed.
 >> 
 >> > Is this  the expected behavior?
 >> 
 >> Yes, you are using the default ess-eval-visibly = t. ESS just waits for
 >> process after each line of your code. 
 >> 
 >> You have two options. Either change it to any other value. Then emacs
 >> will not freeze and all your commands will be executed. But with the
 >> price of not seeing the sent commands
 >> 
 >> Second option is to set ess-inject-source to t. This way you will see
 >> all the command and it will not hand emacs. 
 > hang emacs?
 >> The drawback is that you
 >> won't be able to accumulate multiple paragraphs of code. 
 > Even after reading the help on the ess-inject-source variable I'm not
 > sure what you mean by "accumulating multiple paragraphs."  I guess it
 > has something to do with the docs statement that this only works for
 > valid R expressions.
 >> 
 >> This last part is a relatively new feature and as yet undocumented in
 >> the docs. I will have to spend some time and document all the ways ESS
 >> can evaluate. It is very unlikely that there will be any further
 >> improvement on that front in the year or so to come.
 >> 
 >> In any case the docs of ess-eval-visibly and ess-inject-source should be
 >> enough to completely understand ESS evaluation.
 > Not for me.  Aside from the confusion above, it looks as if this command
 > is only relevant to ess-tracebug.  It's not clear to me in what contexts
 > the setting is relevant.

 > Ross