[Rd] bzfile() in R

[Prof Brian Ripley]

On Sat, 31 Jan 2004, Roger D. Peng wrote:

> Yes, that's true, but I was under the impression that since the 
> bzip2 sources are included with the R sources, that the bzip2 
> capability is always compiled in (either using the included 
> sources or a system library).  Also, the NEWS file for 1.9.0 says:
> 
>      o   configure no longer lists bzip2 and PCRE as `additional
>          capabilities' as all builds of R have had them since 		 
> 1.7.0.

Roger is correct.  Capabilities has not been changed, since earlier 
versions did not have bzfile (and even earlier ones did not have gzfile).


> 
> -roger
> 
> Dirk Eddelbuettel wrote:
> 
> > On Sat, Jan 31, 2004 at 08:37:05PM -0500, Roger D. Peng wrote:
> > 
> >>I'm putting together a data package and am finding that I get 
> >>enormous savings in space by using bzip instead of the usual gzip 
> >>in save().  Is it safe to assume that for R versions, say >= 
> >>1.7.1, that the function bzfile() will always be available?
> > 
> > 
> > I don't think that is correct as this an optional compile-team feature.
> > 
> > help(bzfile) says it 'may' be present whereas gzfile is guaranted (from a
> > 1.9.0 snapshot):
> > 
> >      All platforms support (gz)file connections and 'url("file://")'
> >      connections.  The other types may be partially implemented or not
> >      implemented at all.  (They do work on most Unix platforms, and all
> >      but 'fifo' on Windows.)
> > 		    
> > but then goes on to say under See Also:
> > 
> >      'capabilities' to see if 'gzfile', 'url', 'fifo' and
> >      'socketConnection' are supported by this build of R.
> > 
> > which suggests that even gzfile may not be guaranteed.
> > 
> > You could of course test via capabilities() and abort with a clear error.
> > 
> > Dirk
> >
> 
> ______________________________________________
> R-devel at stat.math.ethz.ch mailing list
> https://www.stat.math.ethz.ch/mailman/listinfo/r-devel
> 
> 

-- 
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