clean wheezy, R intrigue, squeeze-cran3 requires liblzma2?

This is a completely clean, fresh install.  I've not installed any peculiar
libraries or packages, except R, the installation of which still puzzles
me.

I configured a CRAN mirror for packages using squeeze-cran3 folder, and
that's where the first funny thing happened.  I noticed same when R-2.15.3
came out, but I thought I had this trouble because my system was all junked
up with unusual libraries.  But now, on a clean system, I get this error,
which indicates I need a package liblzma2, which does not exist in Debian:

The following packages have unmet dependencies:
 r-base-core : Depends: liblzma2 (>= 4.999.9beta+20091116) but it is not
installable
               Recommends: r-recommended but it is not going to be installed
               Recommends: r-base-dev but it is not going to be installed
               Recommends: r-doc-html but it is not going to be installed

Thanks Charles, but ... (below)


On Sat, May 11, 2013 at 6:28 PM, Charles Plessy
<charles-r-nospam at plessy.org

wrote:

Le Sat, May 11, 2013 at 04:25:14PM -0500, Paul Johnson a ?crit :



Dear Paul,

the packages in the squeeze-cran3 folder are built for the Debian
"Squeeze" (6.0) distribution.  The liblzma2 package exists in Squeeze,
but not in Wheezy
(Debian 7.0, the distribution you used for your fresh install). 
Downgrades are not supported, as exemplified by your problem.

Debian Wheezy was released just one week ago, so I guess it is only a
matter of
time before a "wheezy-cran" folder will be provided.

Best regards,

I agree to your interpretation of the packaging problem. I can post the R
packages for Wheezy on my website if anybody wants them.

Now, back to the "no glyph" problem. Does

demo(plotmath)

work for you on Debian 7?

Here's a re-producible example that generates bad on-screen output for me.

plot(0~0, type = "n")
text(.4, expression(group(lceil,x, rceil)), cex=0.5)

The glyphs for the left and right braces look like "dominoes".

I found a thread from 2010 about a similar problem, but that case blamed
the Wine fonts. Now I suspect some other bad symbol font has found its way
to the front of my path.

https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14355

That thread points out that using the Xlib version of the X11 device SOLVES
the problem, but at the price of slightly lower quality of output. If I
want to use the default Cairo device, apparently I have to fiddle
something.

I am experimenting with the debugging and fontconfig adjustments that might
be used to avoid the trouble.

To the non-full-time-font administrator like me, this is a bit overwhelming
because there are so many different types of fonts.

pj

--
Charles Plessy