R SIG Debian - Oct 2006 - Poll: Does R_PAPERSIZE in /etc/R/Renviron matter?

A heuristic runs via the r-base-core package postinst. It is intended
to use the result from Debian's paperconf(1) toool, and to update the R
environment variable R_PAPERSIZE accordingly.  This apparently broke some
time ago when R changed to a dual variable setup, and my regexp didn't
notice. 

Does anybody use this, or is it ok if I simply disable this in Debian?  In
that case we simply do what everybody else does, so you all get to live by
the 'letter' setting that comes from my compile-time choice.  It is easy
enough to override either in /etc/R/Renviron or in ~/.Renviron.

Opinions either way? Let me know. For now, I intend to simply comment this
out in the next release (presumably a rc for 2.4.1).

Thanks, Dirk

-- 
Hell, there are no rules here - we're trying to accomplish something. 
                                                  -- Thomas A. Edison

I use it.

Thanks,
Robert.

On Tue, 24 Oct 2006 06:36, Dirk Eddelbuettel wrote:
> A heuristic runs via the r-base-core package postinst. It is intended
> to use the result from Debian's paperconf(1) toool, and to update the R
> environment variable R_PAPERSIZE accordingly.  This apparently broke some
> time ago when R changed to a dual variable setup, and my regexp didn't
> notice.
>
> Does anybody use this, or is it ok if I simply disable this in Debian?  In
> that case we simply do what everybody else does, so you all get to live by
> the 'letter' setting that comes from my compile-time choice.  It is
easy
> enough to override either in /etc/R/Renviron or in ~/.Renviron.
>
> Opinions either way? Let me know. For now, I intend to simply comment this
> out in the next release (presumably a rc for 2.4.1).
>
> Thanks, Dirk
-- 
Robert King, Statistics, School of Mathematical & Physical Sciences,
University of Newcastle, Australia
Room V133  ph +61 2 4921 5548
Robert.King at newcastle.edu.au   http://tolstoy.newcastle.edu.au/~rking/

I had not noticed it was broken, so don't seem to use the papersize 
setting much.

Markus

Dirk Eddelbuettel wrote:
> A heuristic runs via the r-base-core package postinst. It is intended
> to use the result from Debian's paperconf(1) toool, and to update the R
> environment variable R_PAPERSIZE accordingly.  This apparently broke some
> time ago when R changed to a dual variable setup, and my regexp didn't
> notice. 
> 
> Does anybody use this, or is it ok if I simply disable this in Debian?  In
> that case we simply do what everybody else does, so you all get to live by
> the 'letter' setting that comes from my compile-time choice.  It is
easy
> enough to override either in /etc/R/Renviron or in ~/.Renviron.
> 
> Opinions either way? Let me know. For now, I intend to simply comment this
> out in the next release (presumably a rc for 2.4.1).
> 
> Thanks, Dirk
> 

-- 
Markus Jantti
Abo Akademi University
markus.jantti at iki.fi
http://www.iki.fi/~mjantti

Dirk Eddelbuettel <edd <at> debian.org> writes:
> A heuristic runs via the r-base-core package postinst. It is intended
> to use the result from Debian's paperconf(1) toool, and to update the R
> environment variable R_PAPERSIZE accordingly.  This apparently broke some
> time ago when R changed to a dual variable setup, and my regexp didn't
> notice. 
> 
> Does anybody use this, or is it ok if I simply disable this in Debian?  In
> that case we simply do what everybody else does, so you all get to live by
> the 'letter' setting that comes from my compile-time choice.  It is
easy
> enough to override either in /etc/R/Renviron or in ~/.Renviron.
> 
> Opinions either way? Let me know. For now, I intend to simply comment this
> out in the next release (presumably a rc for 2.4.1).
What it is actually used for? Living in Europe I "hate" letter, but
never
noticed any effect of that setup. Uff, I have papersize="a4" in
options() in
~/.Rprofile so that is probably the reason that I do not care.

Gregor

On 23 October 2006 at 15:36, Dirk Eddelbuettel wrote:
| 
| A heuristic runs via the r-base-core package postinst. It is intended
| to use the result from Debian's paperconf(1) toool, and to update the R
| environment variable R_PAPERSIZE accordingly.  This apparently broke some
| time ago when R changed to a dual variable setup, and my regexp didn't
| notice. 
| 
| Does anybody use this, or is it ok if I simply disable this in Debian?  In
| that case we simply do what everybody else does, so you all get to live by
| the 'letter' setting that comes from my compile-time choice.  It is
easy
| enough to override either in /etc/R/Renviron or in ~/.Renviron.
| 
| Opinions either way? Let me know. For now, I intend to simply comment this
| out in the next release (presumably a rc for 2.4.1).

My thanks to those who answered already. For those who are perplexed, I
should have add the following few paragraphs.

Run

	$ echo 'print(options("papersize"))' | R --slave

to see your current setting. 

R 2.4.0 has a new heuristic based on the locale:

	The way option("papersize") is set has been changed.  On
	platforms which support the LC_PAPER locale category, the
	setting is taken first from the R_PAPERSIZE environment
	variable at run time, then from the LC_PAPER category
	("letter" for _US and _CA locales and "a4" otherwise).	On
	other platforms (including Windows and older Unixen), the
	choice is unchanged.

which should get it right for most people.  If not, the easiest per-user
override is in ~/.Renviron, and system-wide could be done in
/etc/R/Renviron. 

Thanks, Dirk

-- 
Hell, there are no rules here - we're trying to accomplish something. 
                                                  -- Thomas A. Edison

On Mon, 23 Oct 2006 15:36:02 -0500,
Dirk Eddelbuettel <edd at debian.org> wrote:

[...]
> Does anybody use this, or is it ok if I simply disable this in Debian?
> In that case we simply do what everybody else does, so you all get to
> live by the 'letter' setting that comes from my compile-time
choice.  It
> is easy enough to override either in /etc/R/Renviron or in ~/.Renviron.
I also have to use 'letter' here (but feel much like Gregor ;-) ), so
hadn't noticed the change.

I think it would be good to rely on paperconf, to keep the Debianness.


Cheers,

-- 
Seb