Question

Take this dataset…

test <- data.frame(
    t1=c(2,3,5,6,7,10,10),
    t2=c(3,4,6,7,8,11,12),
    id=1:7
)

…which looks like this.
To clarify, each row is a previously identified link of two cases that have to stay bound together.

  t1 t2 id
1  2  3  1
2  3  4  2
3  5  6  3
4  6  7  4
5  7  8  5
6 10 11  6
7 10 12  7

I am hoping to identify the continuous sequences based on t2 == t1 recursively so that the links are:

link1 -  2-3,3-4
link2 -  5-6,6-7,7-8
link3 -  10-11
link4 -  10-12

The end result i’m looking for is this:

  t1 t2 id matchid
1  2  3  1       1
2  3  4  2       1
3  5  6  3       2
4  6  7  4       2
5  7  8  5       2
6 10 11  6       3
7 10 12  7       4

I have experimented with match(test$t2,test$t1) to get the initial links but am now getting stuck on how to continue the linking process.
My thoughts keep falling back to using a loop and that sounds like a terrible way to go.

Answer 1

Here is one way to do it:

test$matchid <- c(1, 1 + cumsum(tail(test$t1, -1) != head(test$t2, -1)))