CODEWITHSUNDEEP
rvector
mmendina
mmendina

Reputation: 11

Sequentially update a vector in R

Hello I have a vector such as

x<-c(**131144**,**1311605**,1311766,1312289,1312804) in R

And then another data frame like:

v1     , v2
**131144,1283758**
**1283758,19527672**
**1311605,19950311**
198151,37268685
**19950311,35307140**
11281862,11292508
35261079,26296073
625349,37306860
84255273,84259752

I would like to end up with a final vector like this one

x<-c(**19527672**,**19950311**,1311766,1312289,1312804)

Is like to iteratively searching for a value and when a match is found updating it and then keep searching for the updated value until no match found.

Thks in advance.

Upvotes: 1

Views: 77

Answers (2)

ThomasIsCoding
ThomasIsCoding

Reputation: 101247

An option with igraph

g <- graph_from_data_frame(df)
v <- membership(components(g))
tb <- by(names(v), v, function(x) x[degree(g, x, mode = "out") == 0])
m <- unname(v[as.character(x)])
ifelse(is.na(m), x, as.numeric(tb[m]))

gives

[1] 19527672 35307140  1311766  1312289  1312804

where plot(g) shows

enter image description here

Upvotes: 0

Onyambu
Onyambu

Reputation: 79208

Alittle tweak to the solution found here and we have the following:

relation <- function(vec, dat){
  .relation <- function(x){
    k = unique(c(dat[dat[, 1] %in% x, 2], x, dat[dat[, 2] %in% x, 1]))
    if(setequal(x,k)) tail(k, 2)[1] else Recall(k)
  }
  y <- unique(vec)
  sapply(y, .relation)[match(vec, y)]
}

relation(x, df)
[1] 19527672 35307140  1311766  1312289  1312804

Upvotes: 0

Related Questions