Skip to content

R Function Guide

5 messages · Brett Magill, Uwe Ligges, Duncan Murdoch +1 more

Original
Brett Magill
# 
I was wondering if there was anything for R like the function guide 
provided by insightful for S-Plus at the url below:

http://www.insightful.com/resources/fguide.html

What makes if particularly appealing is the organization of 
functions by class.

If there is not something like this for R, I think it would be a great 
addition to the R documentation.  It would provide an easy way to 
identify an available function for a particular category of analysis 
and thus facilitate viewing the help pages for the appropriate 
function or functions--particularly if this were available for not just 
base packages, but contributed packages as well.  Personally, I 
find the biggest weakness of the current help system is the fact 
that I must know the name of a function in advance to request help 
on some analysis or proceedure.  Would be nice to start with a 
general idea, category, class, or set of proceedures and move to a 
list of related functions. 

Extending the original idea, wouldn't it be great if there were a list 
as described above that, in addition to listing functions by 
category, also provided package information, availability, version, 
etc for the listed functions.

I recognize that there is probably some overlap with the S-Plus 
guide, and that it in itself is probably a good reference, or at least a 
nice starting point for a simliar R guide.  However, I also suspect 
that quite a lot of the S-Plus functionality listed is not yet 
implemented in R.  Perhaps I am wrong.  
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help 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-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
Uwe Ligges
#
freud at starpower.net wrote:
Use help.start() and go to "Search Engine & Keywords".

Uwe Ligges
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help 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-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
Brian Ripley
#
On Tue, 5 Jun 2001 freud at starpower.net wrote:

            

      
      
      
    

        
Yes, there is.

            

      
      
      
    

        
That's the same as done by their help.start(), and is very similar to the
html search page in R, which does group by category (`keyword').
Even the categories are very similar.

However, given the number of functions involved (of the order of 2000)
I do find searches more useful (in either system).  There are quite a lot
of functions in S-PLUS which are not in what is to me the obvious category,
but they will be found by a search.   Thre have been lots of postings to
R-help recently where help.search() would have given the answer, so it
looks like an under-used tool.

Also that guide only covers functions in base S-PLUS: R's equivalents
cover all installed packages.  Nothing covers all 120+ packages on CRAN,
let alone those which are not on CRAN.  We do get requests for the latter,
but my own view is that we need better search mechanisms first (both faster
and using approximate matching).

[...]

  
    

      
      
    

  


  


      

    

    

      
      

        


  

        

      Duncan Murdoch
    

    

      
      #
    

  


  

      
On Wed, 6 Jun 2001 07:23:14 +0100 (BST), Prof Brian D Ripley
<ripley at stats.ox.ac.uk> wrote in message
<Pine.GSO.4.31.0106060712410.19713-100000 at auk.stats>:

            

      
      
      
    

        
What about making ?topic try help.search(topic) if help(topic) fails?

It's a very small modification to the help() source, basically just
replacing two "stop(some message)" calls with "cat(some message)", and
adding the help.search calls.  I'd probably add another parameter
"try.help.search" that defaults to FALSE when help() is called, and is
set to TRUE when ? is used.

Duncan Murdoch

       
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help 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-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

  
  


  


      

    

    

      
      

        


  

        

      Brian Ripley
    

    

      
      #
    

  


  

      
On Wed, 6 Jun 2001, Duncan Murdoch wrote:

        

            

      
      
      
    

        
One issue in the past (including 1.2.3) is that the first time help.search
is called in a session it builds a database which can take minutes (lots
of packages, NFS disks). Looks a lot better under 1.3.0.

I think I would still rather say `try help.search(topic)'.

Brian