[R] Newbie problem with read.table

[Jan Conrad]

Thank you all for your answers. Yes, indeed it was only a printing
problem (in fact I had tried
the \t option before without sucess).

My first interaction with R (and the R help), I must say I am impressed.


Best,

Jan

-----Original Message-----
From: Marc Schwartz [mailto:MSchwartz at mn.rr.com] 
Sent: Wednesday, October 12, 2005 3:14 PM
To: Roger.Bivand at nhh.no
Cc: Jan Conrad; r-help at stat.math.ethz.ch
Subject: Re: [R] Newbie problem with read.table


On Wed, 2005-10-12 at 14:56 +0200, Roger Bivand wrote:
> On Wed, 12 Oct 2005, Jan Conrad wrote:
> 
> > Hi R,
> >  I have a seemingly simple problem. I have a table in following 
> > format (tab seperated)
> > 
> >    Njets NBjets	NElec	  NMuon   Meff	HT	  HT3j	HE
Aplan
> > Plan	       
> > 1  4     3	 	2       0       366.278 253.642 87.7473   1385
> > 0.0124566   0.376712       
> > 2  3     1      	1       0       235.19  157.688 18.2852
> > 574.253 0.00064187  0.00528814
> > 
> > I read in with:
> > 
> > > ttbar<-read.table("test2.dat",header=TRUE)
> > 
> > 
> > > ttbar
> >   Njets NBjets NElec NMuon    Meff      HT    HT3j       HE
Aplan
> > 1     4      3     2     0 366.278 253.642 87.7473 1385.000
0.01245660
> > 2     3      1     1     0 235.190 157.688 18.2852  574.253
0.00064187
> >         Plan
> > 1 0.37671200
> > 2 0.00528814,
> > 
> >  i.e.. the table is split after 9 variables. How come ?
> 
> > options("width")
> $width
> [1] 80
> 
> says what the width of your console is. Columns beyond this get 
> wrapped
> gently (not each row by itself) - it can be set different values if
you 
> choose - try:
> 
> ow <- options("width")
> options(width=40)
> options("width")
> ttbar
> options(ow)
> options("width")
> 
> So this is just the print function for data.frame objects doing its 
> unsung job. A very useful function for looking at things when they 
> don't seem to be what you think is str(), which concisely says what 
> the structure of an object is, so str(ttbar)  should tell you that it 
> is a data frame of 10 variables and 2 observations.

Thanks to Roger for this clarification. I took the splitting of the
variables to be a consequence of the delimiter and not just a benign
consequence of the printed output (at least I presume this is the proper
interpretation of Jan's problem.)

The tab character is of course included in "whitespace"....using "\t"
explicitly would be helpful if there is embedded whitespace (other than
a tab) within a field.

Marc
<Off to get another cup of coffee....>