[ESS-bugs] ess-mode 5.3.0; i/o-encoding

Martin Maechler maechler at stat.math.ethz.ch
Wed Oct 10 20:04:39 CEST 2007


Hi Jarle,

would you consider upgrading to the current version of ESS,
5.3.6, and then see if the problems still remain?

I know that Ubuntu (and Debian in general) are very much behind,
since their maintainer of 'ess' is not too avid on making newer
version of ESS available as Debian (and hence Ubuntu) pacakges.

Also, you need to mention what version of R you are using,
and even better the full output of  
    sessionInfo()  from R
    ~~~~~~~~~~~~~
[which includes Sys.getlocale() that is relevant here]

Can you please post to  'ess-help' instead of 
'ess-bugs'.  That is a mailing list of several ESS users,
and I'd really like to have other people hear your problem and
possibly the solution.

Best regards,
Martin Maechler, ETH Zurich


>>>>> "JT" == Jarle Tufto <jarle.tufto at math.ntnu.no>
>>>>>     on Tue, 09 Oct 2007 15:58:40 +0200 writes:

    JT> Hi, I'm having a few problems with norwegian (danish)
    JT> letters and encoding of input and output to R via ess.
    JT> If I choose utf8 as encoding in the R session running in
    JT> the *R*-buffer with

    >> options(encoding="utf8")

    JT> or just keep the default "native.enc" and then change
    JT> the i/o-encoding of the *R*-buffer to "utf-8" using the
    JT> command

    JT> M-x set-buffer-process-coding-system

    JT> everything works fine, for example

    >> cat("ø\n")
    JT> ø
    >> 

    JT> However, if I change the encoding in R to "latin1" and i/o in *R* buffer 
    JT> to "latin-1" I get an error:

    >> cat("ø\n")
    JT> +

    JT> that is, the above command is interpreted as an
    JT> incomplete expression (which it is not).  More
    JT> complicated things like functions (pasted into *R* with
    JT> C-c C-f) are typically interpreted with similar strange
    JT> errors.

    JT> I would have thought that this should have worked since
    JT> the encodings using by both the sender and receiver of
    JT> the actual by sequence are the same...?

    JT> If it is of any importance, this happens on an
    JT> ubuntu-7.04 box with the setup attached below.

    JT> Jarle Tufto


    JT> jarlet:~/prosjekt/fiskerispillet:$ locale
    JT> LANG=en_US.UTF-8 LC_CTYPE="en_US.UTF-8"
    JT> LC_NUMERIC="en_US.UTF-8" LC_TIME="en_US.UTF-8"
    JT> LC_COLLATE="en_US.UTF-8" LC_MONETARY="en_US.UTF-8"
    JT> LC_MESSAGES="en_US.UTF-8" LC_PAPER="en_US.UTF-8"
    JT> LC_NAME="en_US.UTF-8" LC_ADDRESS="en_US.UTF-8"
    JT> LC_TELEPHONE="en_US.UTF-8" LC_MEASUREMENT="en_US.UTF-8"
    JT> LC_IDENTIFICATION="en_US.UTF-8" LC_ALL=

    JT> Emacs : GNU Emacs 21.4.1 (i486-pc-linux-gnu, X toolkit,
    JT> Xaw3d scroll bars) of 2007-08-28 on terranova, modified
    JT> by Debian Package: ess-mode 5.3.0

[...............]

    JT> _______________________________________________ ESS-bugs
    JT> ESS-bugs at stat.math.ethz.ch
    JT> https://stat.ethz.ch/mailman/listinfo/ess-bugs

    JT> _______________________________________________ ESS-core
    JT> list: https://stat.ethz.ch/mailman/listinfo/ess-core



More information about the ESS-bugs mailing list