Grouping and/or splitting

Yes. I was missing the DROP argument.
But now the problem is splitting is causing some weird ordering of groups.

See below:

DF <- read.table(text="
Houseid,Personid,Tripid,taz
1,1,1,4
1,1,2,7
2,1,1,96
2,1,2,4
2,1,3,2
2,2,1,58
3,1,5,7
", header=TRUE, sep=",")
aa <- split(DF, DF[, 1:2], drop=TRUE)

Now the result is aa[3] should is (3,1) and not (2,2). Why? How can I
preserve the ascending order?

aa[3]

$`3.1`
 Houseid Personid Tripid taz
7       3        1      5   7

aa[4]

$`2.2`
 Houseid Personid Tripid taz
6       2        2      1  58


On Wed, Apr 4, 2012 at 6:29 AM, Rui Barradas <rui1174 at sapo.pt> wrote:

Hello,


Ashish Agarwal wrote

I have a dataframe imported from csv file below:

Houseid,Personid,Tripid,taz
1,1,1,4
1,1,2,7
2,1,1,96
2,1,2,4
2,1,3,2
2,2,1,58

There are three groups identified based on the combination of first and
second columns. How do I split this data frame?

I tried
aa <- split(inpfil, inpfil[,1:2])
but it has problems.

Output desired is

aa[1]
Houseid,Personid,Tripid,taz
1,1,1,4
1,1,2,7
aa[2]
Houseid,Personid,Tripid,taz
2,1,1,96
2,1,2,4
2,1,3,2
aa[3]
Houseid,Personid,Tripid,taz
2,2,1,58

     [[alternative HTML version deleted]]

Any of the following three works with me.


DF <- read.table(text="
Houseid,Personid,Tripid,taz
1,1,1,4
1,1,2,7
2,1,1,96
2,1,2,4
2,1,3,2
2,2,1,58
", header=TRUE, sep=",")

DF

split(DF, DF[, 1:2], drop=TRUE)
split(DF, list(DF$Houseid, DF$Personid), drop=TRUE)
with(DF, split(DF, list(Houseid, Personid), drop=TRUE))

The argument 'drop' defaults to FALSE. Was that the problem?

Hope this helps,

Rui Barrada

	[[alternative HTML version deleted]]