Grokbase Groups R ess-help April 2007
FAQ
I am using ESS-emacs, current versions, with R-devel 2.5.0 and in R have
options(error=recover) set for debugging. When I hit an error, instead
of giving me the usual menu of levels to look at in R, everything just
hangs until I hit cntl-g. As some of my programs run a long time anyway,
it is impossible for me to tell if I have hit an error or just haven't
been patient enough.

Suggestions?

Mark
Mark W Kimpel
Indiana University

Search Discussions

  • Richard M. Heiberger at Apr 23, 2007 at 8:27 pm
    Yes, ESS is looking for a "> " prompt. recover() is issuing
    a "Selection: " which is not the signal that ESS is looking for.

    If we made ESS sensitive to "Selection: "
    would there be a downside? Could we change the prompt conditional
    on something? For example, we could trap the statement in the *R* buffer
    options(error=recover)
    and set the inferior-ess-primary-prompt to recognize "Selection: ".
    When would we set it back?

    A better plan for the user, but harder on ESS, would be to monitor
    the options()$error

    Rich
  • Seth Falcon at Apr 24, 2007 at 4:17 am

    "Richard M. Heiberger" <rmh@temple.edu> writes:

    Yes, ESS is looking for a "> " prompt. recover() is issuing
    a "Selection: " which is not the signal that ESS is looking for.
    Hmm, options(error=recover) seems to be working fine for me inside
    ESS.

    recover() calls the menu() function which requests user input
    before dropping into a browser. Although menu appears to be internal
    it should be similar to readline. I suspect something else is going
    on -- do these functions work for others or am I just lucky ;-)

    + seth

    --
    Seth Falcon | Computational Biology | Fred Hutchinson Cancer Research Center
    http://bioconductor.org
  • Deepayan Sarkar at Apr 24, 2007 at 6:14 am

    On 4/23/07, Seth Falcon wrote:
    "Richard M. Heiberger" <rmh@temple.edu> writes:
    Yes, ESS is looking for a "> " prompt. recover() is issuing
    a "Selection: " which is not the signal that ESS is looking for.
    Hmm, options(error=recover) seems to be working fine for me inside
    ESS.

    recover() calls the menu() function which requests user input
    before dropping into a browser. Although menu appears to be internal
    it should be similar to readline. I suspect something else is going
    on -- do these functions work for others or am I just lucky ;-)
    They work for me too.

    I do get ESS freezes (that need C-g to interrupt) sometimes, but that
    seems unrelated (they are mostly triggered by C-c C-r, but never C-c
    C-n).

    -Deepayan
  • Dan Davison at Apr 24, 2007 at 6:30 am

    On Mon, Apr 23, 2007 at 09:17:02PM -0700, Seth Falcon wrote:
    "Richard M. Heiberger" <rmh@temple.edu> writes:
    Yes, ESS is looking for a "> " prompt. recover() is issuing
    a "Selection: " which is not the signal that ESS is looking for.
    Hmm, options(error=recover) seems to be working fine for me inside
    ESS.

    recover() calls the menu() function which requests user input
    before dropping into a browser. Although menu appears to be internal
    it should be similar to readline. I suspect something else is going
    on -- do these functions work for others or am I just lucky ;-)
    Normal use of options(error=recover) has always been OK for me in general, but I do see this behaviour in the following situation: ess-load-file (C-c C-l) hangs if it encounters a parsing problem and options(error=recover) is set.

    e.g. contents of file tmp.R:

    f <- function() {
    if(
    6
    }
    options(error=recover)
    ## C-c C-l in tmp.R buffer now results in a hang, requiring C-g before one can get to the R buffer where hitting <return> gets you the "Selection: " prompt [not that that's what you want; you never get the "Errors: Use C-c ` to find error" message, which of course you do get without options(error=recover) ]

    Dan


    R.version
    _
    platform i686-redhat-linux-gnu
    arch i686
    os linux-gnu
    system i686, linux-gnu
    status
    major 2
    minor 4.1
    year 2006
    month 12
    day 18
    svn rev 40228
    language R
    version.string R version 2.4.1 (2006-12-18)


    + seth

    --
    Seth Falcon | Computational Biology | Fred Hutchinson Cancer Research Center
    http://bioconductor.org

    ______________________________________________
    ESS-help at stat.math.ethz.ch mailing list
    https://stat.ethz.ch/mailman/listinfo/ess-help
  • Martin Maechler at Apr 24, 2007 at 7:56 am

    "Seth" == Seth Falcon <sfalcon@fhcrc.org>
    on Mon, 23 Apr 2007 21:17:02 -0700 writes:
    Seth> "Richard M. Heiberger" <rmh@temple.edu> writes:
    Yes, ESS is looking for a "> " prompt. recover() is issuing
    a "Selection: " which is not the signal that ESS is looking for.
    Seth> Hmm, options(error=recover) seems to be working fine for me inside
    Seth> ESS.

    are you using the unreleased development version of ESS ?

    There, we have fixed one of the problems of options(error=recover)

    Yes, I know we should release ESS-5.2.4 : The only problem is
    that I had promissed to add things there, that haven't been done
    yet...

    Martin

    Seth> recover() calls the menu() function which requests user input
    Seth> before dropping into a browser. Although menu appears to be internal
    Seth> it should be similar to readline. I suspect something else is going
    Seth> on -- do these functions work for others or am I just lucky ;-)

    Seth> + seth
  • Seth Falcon at Apr 24, 2007 at 2:07 pm

    Martin Maechler <maechler@stat.math.ethz.ch> writes:

    "Seth" == Seth Falcon <sfalcon@fhcrc.org>
    on Mon, 23 Apr 2007 21:17:02 -0700 writes:
    Seth> "Richard M. Heiberger" <rmh@temple.edu> writes:
    Yes, ESS is looking for a "> " prompt. recover() is issuing
    a "Selection: " which is not the signal that ESS is looking for.
    Seth> Hmm, options(error=recover) seems to be working fine for me inside
    Seth> ESS.

    are you using the unreleased development version of ESS ?
    Last Changed Rev: 3696
    Last Changed Date: 2007-02-11 13:47:10 -0800 (Sun, 11 Feb 2007)


    --
    Seth Falcon | Computational Biology | Fred Hutchinson Cancer Research Center
    http://bioconductor.org
  • Martin Maechler at Apr 25, 2007 at 3:50 pm

    "Seth" == Seth Falcon <sfalcon@fhcrc.org>
    on Tue, 24 Apr 2007 07:07:13 -0700 writes:
    Seth> Martin Maechler <maechler@stat.math.ethz.ch> writes:
    "Seth" == Seth Falcon <sfalcon@fhcrc.org>
    on Mon, 23 Apr 2007 21:17:02 -0700 writes:
    >>
    Seth> "Richard M. Heiberger" <rmh@temple.edu> writes:
    Yes, ESS is looking for a "> " prompt. recover() is issuing
    a "Selection: " which is not the signal that ESS is looking for.
    >>
    Seth> Hmm, options(error=recover) seems to be working fine for me inside
    Seth> ESS.
    >>
    are you using the unreleased development version of ESS ?
    Seth> Last Changed Rev: 3696
    Seth> Last Changed Date: 2007-02-11 13:47:10 -0800 (Sun, 11 Feb 2007)

    Ok. Thank you, Seth.
    Indeed that explains that you don't have problems.
    The following revision was indeed fixing these (problems):

    ------------------------------------------------------------------------
    r3693 | maechler | 2007-01-22 10:49:29 +0100 (Mon, 22 Jan 2007) | 1 line
    Changed paths:
    M lisp/ChangeLog
    M lisp/ess-cust.el
    M lisp/ess-inf.el

    use try(names(.)) in object completion!
    ------------------------------------------------------------------------

    Regards,
    Martin

Related Discussions

Discussion Navigation
viewthread | post
Discussion Overview
groupess-help @
categoriesr
postedApr 23, '07 at 8:03p
activeApr 25, '07 at 3:50p
posts8
users6
websiteess.r-project.org
irc#r

People

Translate

site design / logo © 2022 Grokbase