Full_Name: Antonio Possolo Version: 1.3.0 OS: Linux Submission from: (NULL) (192.35.44.3) If one attempts to read a character string from a file using scan, and the string comprises more than 8190 characters, then the string is truncated to this length, and no warning is issued. I've observed this both in the Linux and Win implementations of R 1.3.0. S-PLUS (v. 6.0 rel. 1 for Sun SPARC0 does not suffer from this behavior. -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html Send "info", "help", or "[un]subscribe" (in the "body", not the subject !) To: r-devel-request@stat.math.ethz.ch _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
On Fri, 3 Aug 2001 possolo@crd.ge.com wrote:> Full_Name: Antonio Possolo > Version: 1.3.0 > OS: Linux > Submission from: (NULL) (192.35.44.3) > > > If one attempts to read a character string from a file using scan, > and the string comprises more than 8190 characters, then the string > is truncated to this length, and no warning is issued. > > I've observed this both in the Linux and Win implementations of R 1.3.0. > S-PLUS (v. 6.0 rel. 1 for Sun SPARC0 does not suffer from this behavior.Yes, that's an R limitation. Why is it a problem? I think you would do better using readLines if it is a line-oriented file. AFAIK that is unlimited. It would be easy to issue a warning, quite a bit harder to remove the limit (which dates from the days when character strings were in most places limited to 8191 chars). -- Brian D. Ripley, ripley@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 272860 (secr) Oxford OX1 3TG, UK Fax: +44 1865 272595 -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html Send "info", "help", or "[un]subscribe" (in the "body", not the subject !) To: r-devel-request@stat.math.ethz.ch _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
On Fri, 3 Aug 2001 possolo@crd.ge.com wrote:> Full_Name: Antonio Possolo > Version: 1.3.0 > OS: Linux > Submission from: (NULL) (192.35.44.3) > > > If one attempts to read a character string from a file using scan, > and the string comprises more than 8190 characters, then the string > is truncated to this length, and no warning is issued. > > I've observed this both in the Linux and Win implementations of R 1.3.0. > S-PLUS (v. 6.0 rel. 1 for Sun SPARC) does not suffer from this behavior.Although readLines is normally to be preferred (in R and S-PLUS 6: it's a lot faster for a start), we've fixed this so that scan() in R 1.3.1 will give a warning R 1.4.0 will allow unlimited lines (unlimited by R, that is). As ever, previews are in r-patched and r-devel (tar bundles/rysnc) respectively. The 1.4.0 version needed quite a lot of re-writing of code, too much for a patch release. -- Brian D. Ripley, ripley@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 272860 (secr) Oxford OX1 3TG, UK Fax: +44 1865 272595 -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html Send "info", "help", or "[un]subscribe" (in the "body", not the subject !) To: r-devel-request@stat.math.ethz.ch _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._