[R] Rearranging long dataframe for printing in Sweave (and recoding a factor)

[Jean-Louis Abitbol]

Dear R-Helpers,

I have to print a dataframe with >1000 rows and 2 columns for a report
done with Sweave.

I could use Hmisc latex function with longtable option.
However it is a waiste of space and paper given that I have only 2 (or
sometime 3) columns in the dataframe.

So it came to my mind that I could maybe  rearrange the dataframe and
create several new variables from the original 2 in a tabular way for
printing in less space.

I started with this:
library(Hmisc)
test <- data.frame(conc=trunc(runif(1000, 1,
1000)),nam=rep(c("A","B"),500))
test$groups <- cut2(test$conc,g=10) 

The idea being to use cut2 to get 10 groups with approx the same number
of rows and then subset on the groups to create 10 new variables.

I am stuck right away as cut2 produces test$groups with the levels:

levels(test$groups)
 [1] "[  1,100)" "[100,215)" "[215,312)" "[312,411)" "[411,500)"
 "[500,598)" "[598,700)" "[700,797)" "[797,890)" "[890,997]"

and so my first question is: how do I recode the levels in a more or
less automatic way (ie not having to spell them) so that they are
replaced by a factor with values 1 to 10 in this case or more generally
in 1 to g levels ? (and then I can subset in a loop and create conc1,
conc2, conc3 etc..) 

My second question is: is there a function around or an options in one
of the latex functions to do what I have explained at the beginning of
this mail ! I did not find that in Hmisc latex or in xtable.

Thanks to all for your help.

Best wishes, JL


-- 
  Jean-Louis Abitbol
  abitbol at sent.com
  Tel 33 (0)6 24 47 59 34
  1 ter Rue Céline, 92160 
  Antony, France