Skip to content

Compiling R-devel under Windows

2 messages · apjaworski@mmm.com, Brian Ripley

Original
apjaworski@mmm.com
# 
Hello there,

I have a habit of compiling R-patched and R-devel every few days.
Recently, I noticed the following problem:

I downloaded both versions with Revision: 40421, Last Changed Date:
2007-01-09.  R-patched compiles with no problem.  R-devel compiles fine but
when it starts to make PDF documentation it generates the following (tail
of the "make distribution" output):

! pdfTeX warning (dest): name{Rfn.survival} has been referenced but does
not ex
ist, replaced by a fixed one

<C:/local/MiKTeX/fonts/type1/bluesky/cm/cmti10.pfb>{8r.enc}<C:/local/MiKTeX/fon
ts/type1/urw/times/utmbi8a.pfb><C:/local/MiKTeX/fonts/type1/bluesky/ams/cmbsy7.
pfb><C:/local/MiKTeX/fonts/type1/bluesky/cm/cmbx10.pfb><C:/local/MiKTeX/fonts/t
ype1/bluesky/cm/cmmib10.pfb><C:/local/MiKTeX/fonts/type1/bluesky/cm/cmr5.pfb><C
:/local/MiKTeX/fonts/type1/urw/courier/ucrro8a.pfb><C:/local/MiKTeX/fonts/type1
/bluesky/cm/cmmi5.pfb><C:/local/MiKTeX/fonts/type1/bluesky/cm/cmr7.pfb><C:/loca
l/MiKTeX/fonts/type1/bluesky/cm/cmex10.pfb><C:/local/MiKTeX/fonts/type1/bluesky
/cm/cmmi7.pfb><C:/local/MiKTeX/fonts/type1/bluesky/cm/cmr10.pfb><C:/local/MiKTe
X/fonts/type1/bluesky/cm/cmsy7.pfb><C:/local/MiKTeX/fonts/type1/bluesky/cm/cmmi
10.pfb><C:/local/MiKTeX/fonts/type1/bluesky/cm/cmsy10.pfb><C:/local/MiKTeX/font
s/type1/urw/times/utmri8a.pfb><C:/local/MiKTeX/fonts/type1/urw/courier/ucrr8a.p
fb><C:/local/MiKTeX/fonts/type1/urw/courier/ucrb8a.pfb><C:/local/MiKTeX/fonts/t
ype1/urw/times/utmr8a.pfb><C:/local/MiKTeX/fonts/type1/urw/helvetic/uhvr8a.pfb>
<C:/local/MiKTeX/fonts/type1/urw/times/utmr8a.pfb><C:/local/MiKTeX/fonts/type1/
urw/times/utmb8a.pfb>
Output written on refman.pdf (1447 pages, 6487470 bytes).
Transcript written on refman.log.
make[2]: *** [refman.pdf] Error 1
make[1]: *** [manuals] Error 2
make[1]: Leaving directory `/home/andyj/R-devel/src/gnuwin32'
make: *** [distribution] Error 2

It looks like the first version of refman.pdf gets generated fine and
something happens when it needs to run makeindex.  I tried to look at the
Makefile.win in doc/manuals but I cannot see anything obvious there.  In
any case, the makefiles are quite elabortae and interlinekd, so I probably
would not know how to fix them.

I am on Win2000.  I compile R using the Cygwin environment, but I have
MinGW installed and its executables are first on the search path.  I am
using MiKTeX and it is up-to-date.

Thanks in advance,

Andy

PS.  I seem to remember that this happened already a few versions ago.


__________________________________
Andy Jaworski
518-1-01
Process Laboratory
3M Corporate Research Laboratory
-----
E-mail: apjaworski at mmm.com
Tel:  (651) 733-6092
Fax:  (651) 736-3122
Brian Ripley
# 
You have an old snapshot.  R-devel r40430 seems fine.

You could have looked at the refman.log for informative output (no need to 
speculate: it was a help file that was provisional at the time).  Please 
don't expect all aspects of R-devel to work on all platforms at all times: 
It is

    R version 2.5.0 Under development (unstable)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
On Wed, 10 Jan 2007, apjaworski at mmm.com wrote: