savehistory directories and quitting R (PR#2038)

dmurdoch@pair.com dmurdoch@pair.com
Thu, 19 Sep 2002 12:55:10 +0200 (MET DST)


>Because I work in different directories, but always want to save my
>.Rhistory in the same place, I have changed the system function savehistory
>to this:
>
>function (file = "D:/R50/.Rhistory") 
>invisible(.Internal(savehistory(file)))
>
>When I use q() to quit R, and it asks me whether I want to "save current
>workspace", a response of YES means that the .Rhistory is saved in the
>current working directory, rather than D:/R50/.Rhistory. I take this to mean
>that the quit-on-prompt sequence makes a direct call to .Internal(
>savehistory...) rather than going via the actual savehistory() function. It
>would be nicer if prompt-on-quit would honour the user's definition of
>savehistory.

What happens on shutdown is specific to the platform you're working
on, but in both Windows and Unix versions, the filename is taken from
the R_HISTFILE environment variable.  I don't know if this works on
the Mac.

This would be more obvious if the R savehistory() function were
changed to check the R_HISTFILE environment variable too.

The problem with using the user's savehistory() function is that this
violates the idea of namespaces that we're working towards:  users
should be able to create their own function named "savehistory"
without worrying about changing important internal behaviour.

Duncan Murdoch

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._