Re: [babel] python session speed vs non session

From: "Eric Schulte" <schulte.eric@gmail.com>
To: Seth Burleigh <wburle@gmail.com>
Cc: emacs-orgmode@gnu.org
Subject: Re: [babel] python session speed vs non session
Date: Tue, 07 Sep 2010 16:41:41 -0600	[thread overview]
Message-ID: <87bp89xi9j.fsf@gmail.com> (raw)
In-Reply-To: AANLkTi=bXug1xn=CAy7q3O-rg+=dkKSNmLXE-d7CNAmw@mail.gmail.com

Hi Seth,

This is probably a combination of both the use of comint-mode which
indirects interaction with the python process through a temporary Emacs
buffer, in combination with the fact that Emacs must explicitly wait for
the interactive python process to return output until a special
"end-of-output" string has been received.

Dan has recently changed the session based evaluation for R to use a
different technique which doesn't require these explicit checks for an
"end-of-output" string which results in less clutter for the session
buffer and /potentially/ faster evaluation.  I would like to look at the
possibility of converting some other languages to this new schema.

Sorry I can't point to a clear culprit.  Best -- Eric

Seth Burleigh <wburle@gmail.com> writes:

> I have recently been trying out the src executing capabilities of orgmode. I
> have been confused why a python session takes significantly longer to
> execute than a non python session. For example
>
> #+begin_src python
> return 2+2
> #+end_src
>
> executes pretty quickly, but
> #+begin_src python :session test
> 2+2
> #+end_src
>
> executes twice as long. Looking at the command shell for python, it seems
> that the time is being taken to actually send the source to the command
> shell buffer. Once its sent (or once i can see it displayed in the command
> shell buffer) it immediately returns. Any ideas?
> _______________________________________________
> Emacs-orgmode mailing list
> Please use `Reply All' to send replies to the list.
> Emacs-orgmode@gnu.org
> http://lists.gnu.org/mailman/listinfo/emacs-orgmode