[Rd] R documentation -- buffer overflow? (PR#3400)

Dirk Eddelbuettel edd at debian.org
Fri Jul 4 12:58:47 MEST 2003


On Fri, Jul 04, 2003 at 06:04:07PM +0200, maechler at stat.math.ethz.ch wrote:
> I don't think so.  When I use cut and paste to recreate your
> file,
> and do (in a {t}csh; (ba|k)?sh needs another syntax)
> 
>      R CMD Rdconv --debug -t txt xswms2d.Rd >& xswms2d.txt
> 
> The resulting file becomes
> ----------------------------------------------------------
> -- read file 'xswms2d.Rd';
> 
> -- mark_brackets:............................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................!
>  .............................
> 
> ******* Syntax error: mismatched or missing brackets in
> /-----
> 
> \name{xswms2d}
> \alias{xswms2d}
> \title{Modelling of water flux and solute transport}
> 
> <..........>
> <..........>
> 
> --------------------------------------------------------------
> 
> and is even a bit larger than the original.
> I can hardly believe that
>   ``Syntax error: mismatched or missing brackets in ...''
> is caused by a (perl) buffer overflow.

I played with it to, and found that commenting out a single line was enough
to trigger the fault.  And that line had no issues with brackets.

Something fishy happening here ...

Dirk

-- 
Don't drink and derive. Alcohol and analysis don't mix.



More information about the R-devel mailing list