[Rd] bug in texi2dvi? (PR#13833)

Prof Brian Ripley ripley at stats.ox.ac.uk
Thu Jul 16 07:19:44 CEST 2009

On Wed, 15 Jul 2009, Duncan Murdoch wrote:

> On 15/07/2009 6:00 AM, paola at idsia.ch wrote:
>> I am trying to call texi2dvi on a very simple tex-file (see below) and it 
>> fails with a memory access error. I originally tried to call it on another, 
>> meaningful tex file with the same error occuring. I am attaching a 
>> screenshot.
>> Thank you very much for your help.
> I couldn't view your screenshot, but I wonder why you think this is an R bug. 
> texi2dvi is a separate program, unrelated to R, except that R uses it for 
> some of the documentation.
> (texinfo is designed to process .texi files, not .tex:  your file looked like 
> latex to me, so I'm not surprised texinfo failed, but I'm somehwat surprised 
> it had a memory access error.)

Maybe, but both implementations of texi2dviI am aware of do handle 
most tex dialects.  The shell script that is part of the texinfo 
distribution says

     Run each Texinfo or LaTeX FILE through TeX in turn until all cross-ref-
     erences are resolved, building all indices.  The  directory  containing
     each  FILE  is searched for included files.  The suffix of FILE is used
     to determine its language (LaTeX or Texinfo).

and in MiKTeX texi2dvi is an alias for 'texify'.  We exploit this to 
build vignettes.

I too did not bother to try to unencode the screenshot, and all the 
basic information in a R-bugs report (including the version of R, the 
OS and the exact commands used) were missing.  Here the OS and hence 
implementation of texi2div were crucial data, as was which program had 
the 'memory access error' (most likely it was not R itelf)

Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595

More information about the R-devel mailing list