Re: [R] Reading hierarchical data

Here is a further simplification. We use the colClasses= argument with "NULL" for the columns we do not want so we do not have to later remove those columns. # record type ("1" or "2") rectype <- substr(input, 7, 7) # read in record type "1" input1 <- input[rectype == "1"] DF1 <- read.fwf(textCo

Re: [R] Reading hierarchical data

Try this. It uses input defined in Jim's post and defines the rectype of each row ("1" or "2"). It then reads the rectype "1" records into DF1 using read.fwf and the rectype "2" records into DF2 also using read.fwf. ix is defined to have one component per personal record giving the row number in

Re: [R] Reading hierarchical data

Will this do it for you: > input <- readLines(textConnection("06470 1 1 + 1 232 0 + 2 230 1 + 07470 1 0 + 1 240 1 + 08470 1 0 + 1 227 0 + 09470 1 0 + 1 213 1 + 2 222 0 + 3 224 1 + 10470 1 1 + 1 220 0 + 2 211 1 + 11470 1 0 + 1 217 0 + 2 210 1 + 3 226

[R] Reading hierarchical data

I would like to read the following hierarchical data set. There is a family record followed by one or more personal records. If col. 7 is "1" it is a family record. If it is "2" it is a personal record. The family record is formatted as follows: col. 1-5 family id col. 7"1" col. 9