headers
Marius Hofert | 21 May 2013 14:35
Picon
Favicon

ess-history-file nil is not respected anymore

Hi,

I have

(setq ess-history-file nil)

in ~/.emacs to avoid that .Rhistory files are written. This worked well
for me for quite some time until I updated to 13.05-1 [rev. 5586
(2013-05-17)]. Since then it shows .Rhistory and not nil (maybe this is
due to
https://github.com/emacs-ess/ESS/commit/710f4e462f0ede6e9f09ae9bd46f83ebaa3cfb77
?).

The question is, how can I get 'nil', so really no .Rhistory being
written?

Thanks & cheers,

Marius

______________________________________________
ESS-help <at> r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/ess-help
Permalink | Reply | 4 comments |
headers
Michael Lawrence | 20 May 2013 18:03

ess-boolean-command incorrectly assumes case-fold-search is 't'

Sorry to keep pestering you guys, but I really wanted to try the
(relatively) new developer mode. Got this error message:

ess-toggle-developer: Could not source ess-developer.R. Please investigate
the output of *ess-command-output* buffer for errors

There was no ess-command-output buffer, because the source actually
succeeded. Instead, the culprit appears to be ess-boolean-command, which
tries to match TRUE/true in a case-insensitive fashion. However, it calls
(re-search-forward "true" nil t), and the case-sensitivity of
re-search-forward depends on case-fold-search being 't'. This then fails
when case-fold-search is 'nil', for R at least.

Hope that helps,
Michael

	[[alternative HTML version deleted]]

______________________________________________
ESS-help <at> r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/ess-help
Permalink | Reply | 1 comment |
headers
Michael Lawrence | 20 May 2013 15:12

suggestion for R: indicating symbol scope / existence

This might be a lot of work, but it would be cool. Not completely though
out yet. I'll volunteer to help out, on the R side at least. The codetools
package should help a lot.

Via font-locking or similar, indicate the scope of every symbol in the
buffer. This would distinguish symbols found in the:
* Current scope (e.g., function)
* Package imports (in developer mode) -- and which package by tooltip?
* Base namespace (in developer mode)
* Search path (when buffer has a session) -- and which element?

The goal here would be to make it more obvious when the wrong symbol is in
scope, or if a symbol is unbound. Would be cool if this updated
continuously, in a separate thread or something, but for now cold updating
by key-press would work.

A simpler first shot might be to implement this for the session buffer:
does what I'm typing exist? Is it in my workspace, or a package?

Michael

	[[alternative HTML version deleted]]

______________________________________________
ESS-help <at> r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/ess-help
Permalink | Reply | 0 comments |
headers
Michael Lawrence | 18 May 2013 15:19

suggestion: eldoc support for S4 methods

I really like the new eldoc that is turned on in 13.04, and it would be
cool to make it work on S4 methods. Right now, it accesses only the
generic, which is often insufficient, because S4 methods will add arguments
as allowed by the presence of "..." in the generic formals.

Having no idea of how this works inside ESS, I thought that if you could
somehow get the text of the call, you could use selectMethod() to find the
actual method. I threw together the function below just for fun. It takes a
string representation of an R call, parses it, evaluates the arguments,
matches them to elements of the generic signature, and resolves the method.
It even handles the ".local" delegation that happens when methods add
formal arguments as mentioned. The evaluation of the arguments might be
expensive, so this should be on-demand only.

parseToSelectMethod <- function(expr) {
    call <- parse(text = expr)[[1]]
    generic <- match.fun(call[[1]])
    matched.call <- match.call(generic, call)
    args <- lapply(as.list(matched.call[-1]), eval, parent.frame())
    classes <- vapply(args, class, character(1))
    classes.sig <- classes[generic <at> signature]
    method <- selectMethod(generic, classes.sig)
    method.body <- body(method)[[2]]
    has.local <- identical(quote(`<-`(.local)), method.body[1:2])
    if (has.local) {
        method <- method.body[[3]]
    }
    method
}
(Continue reading)

Permalink | Reply | 3 comments |
headers
Michael Lawrence | 18 May 2013 15:07

ess-describe-object-at-point: how to cycle?

Hi guys,

When I issue ess-describe-object-at-point, it brings up the electric
buffer, as expected, and instructs to type 'm' to cycle. The ESS docs say
that 'e' cycles. But neither one of them seems to work, because as soon as
I type anything, the electric buffer disappears (hence the 'electric'
adjective). How do I cycle it? This might be something weird about my
setup.

Thanks,
Michael

	[[alternative HTML version deleted]]

______________________________________________
ESS-help <at> r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/ess-help
Permalink | Reply | 3 comments |
headers
Michael Lawrence | 18 May 2013 14:17

eval-line-and-step broken in org mode buffer

Hi,

If I have an org-mode buffer with these contents:

# Skips to here

#+begin_src R
library(tools) # eval-line-and-step issued here
library(MASS)
#+end_src

And I issue eval-line-and-step at the library(tools) line, ESS (13.05 at
least) will "skip" up to the first commented line ABOVE the current line,
instead of just going down to the next line.

This error is emitted in the mini-buffer: Wrong type argument: stringp, nil.

Thanks,
Michael

	[[alternative HTML version deleted]]

______________________________________________
ESS-help <at> r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/ess-help
Permalink | Reply | 2 comments |
headers
Steve Lianoglou | 17 May 2013 20:53

Getting eldoc to offer argument suggestions quicker

Hi folks,

I just upgraded to 13.05. So far so good.

Even though the enhanced eldoc/auto-complete stuff isn't new to this
release (I think it was in the last one) I actually just turned it on
here and *wow*, I totally dig it. Thanks for that!

I was curious if something is hosed w/ my setup, though, because in
order for the argument suggestions for a function to pop up in the
minibuffer, I need to make the point move in some way w/in the parens
of a function before it activates.

For instance, say I'm in the R buffer below and the `|` is the point.
After I type the open paren to `cor.test`, we're looking like so:

R> cor.test(|)

At this point, my minibuffer is silent. If, I hit space:

R> cor.test( |)

Then the args pop in my minibuffer. Or, I can just move the point left
one then right one to land back in between the parens to trigger the
args also show in the minibuffer.

I get the impression that it's not supposed to happen this way and the
args should show as soon as the point starts in between the parens
after I've opened them, but I'm not sure what's cooking on my setup.
I've commented out all of my own customizations to ESS that have been
(Continue reading)

Permalink | Reply | 1 comment |
headers
Frede Aakmann Tøgersen | 16 May 2013 09:44

Get an wrong type argument: processp, nil when quittin ESS/R

Hi

When quitting R from an R script file using C-c C-q I get an error message from Emacs. It is not a critical error though.

Here is the content of the *Messages* buffer:

Loading /gpfs/home/frtog/.recentf...done
Cleaning up the recentf list...done (0 removed)
Loading /gpfs/home/frtog/ESS/lisp/ess-site.el (source)...
enabling speedbar support
Loading /gpfs/home/frtog/ESS/lisp/ess-site.el (source)...done
For information about GNU Emacs and the GNU system, type C-h C-a.
Type C-h m for help on ESS version 13.05
ess-tracebug mode enabled
Fontifying *R*... (regexps....................)
ess-tracebug mode enabled
using process '*R*'
ansi-color-process-output: Wrong type argument: processp, nil
Quit

****** end of buffer

ansi-color-process-output seems to be the culprit.

I'm using R-2.14.0 (also happens with R-3.0.0)

My emacs: GNU Emacs 23.3.1 (x86_64-unknown-linux-gnu, GTK+ Version 2.10.4) of 2012-02-02 on tesla339

Content of *ESS* buffer:
(Continue reading)

Permalink | Reply | 0 comments |
headers
Matthew Fidler | 15 May 2013 21:10
Picon
Gravatar

smarter underscore

All,

I don't know if this is the right place to put this in, but I have changed
the ess-smarter-underscore package to allow smarter underscore in ESS. In
particlar, I have allowed ggplot functions to be recognized.  See:

https://github.com/mlf176f2/ess-smart-underscore.el

If there is interest, I can submit a patch to the ESS github code base.

Matt.

	[[alternative HTML version deleted]]

______________________________________________
ESS-help <at> r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/ess-help
Permalink | Reply | 0 comments |
headers
Martin Maechler | 15 May 2013 09:18
Picon
Favicon
Gravatar

ESS 13.05 released

A few moments ago, we have relased  ESS 13.05.

This is "big step" release  in the following sense:
There are several small improvements and also *NON* backcompatible 
changes of defaults on purpose  (all of which can be reverted if you need):

1) "ess-tracebug" has been turned on by default, which influences
  and facilitates debugging .. but also needs some "getting used to".

2) ess ElDoc is more visible, more "helping", but therefore
   potentially surprising before you're getting used to.
   OTOH, it should be more transparent to the R user as "ESSR"
   now is part of search().

3) The default ess-style has been changed from years of 'DEFAULT'
   to 'RRR',  see the recent e-mails on ESS-help, for the main
   rason that a close cousin of RRR has been in use in Vincent
   Goulet's Windows and MacOS distros for ESS. 

Below is the first half of the ANNOUNCE file part of the ESS docu.
Also note at the very bottom the cited e-mail from Rodney
Sparapani who gives links into the online version of the ESS
manual for the most important changes.

In the name of the ESS Core Team,

Martin Maechler, 
ETH Zurich

----------------------------
(Continue reading)

Permalink | Reply | 0 comments |
headers
Paul Johnson | 14 May 2013 17:49
Picon

S-Enter to send line to R, not C-Enter. Please?

Good morning.

This is my "random ESS question" day.

Am I the only one who wishes that send to R would be bound to S-Enter, not
C-Enter?

In cua mode, C-Enter is bound to their very handy rectangle selection tool.

For that reason, my .emacs has S-Enter set as the "send line to R buffer"
functionality.  Could you make it possible for me to assign that within ESS
(Rather than scripting up my own recipe, which I have been doing for 10
years).

This is not an entirely new thing, I did not invent it.  In the olden days
of Emacs for Windows, at the turn of the century, Shift-Enter was pre-set
by the packager, and I liked it.  Memory tells me the packager was C.
Curly.  We had that setting in the first RPM packages for ESS, in Fedora 6
or such.

pj

--

-- 
Paul E. Johnson
Professor, Political Science      Assoc. Director
1541 Lilac Lane, Room 504      Center for Research Methods
University of Kansas                 University of Kansas
http://pj.freefaculty.org               http://quant.ku.edu

	[[alternative HTML version deleted]]
(Continue reading)

Permalink | Reply | 3 comments |

Gmane