How can I use loops for and if - r

Question

This is my example dataset
> a
V1 V2
1 A1 5437
2 A1 5437
3 A1 5437
4 A2 1819
5 A2 1758
6 A2 1212
7 A2 1212
8 A3 1456
I want to compute unique values for column V2, so the result will be:
A1 1
A2 3
A3 1
I have started write my code, but I don't have idea - how should this look like:
old_id <- a[1,2]
old_art <- a[2,1]
for (i in nrow(a)){
if (old_id == a[1,i+2] && old_art == a[i+2,1]){
new_id[i] <- old_id[1,i+2]
new_art[i] <- i
}
i <- i+1
}
I know very simple solution like:
tapply(a[,2], a[,1], function(t) length(unique(t)))
but my task is to use loop function - probably for and if

Answer 1

This sounds like homework. But for loops run through all elements in the vector on the right hand side of in. This also means that your for loop will increment automatically and so you don't need i <- i+1.
Hence, your for loop should look like this
for (i in 1:nrow(a)) {
< your code >
}
# i <- i + 1 # No need for this!
Notice i in 1:nrow(a) and no i in nrow(a). I haven't checked your code, only your forsyntax.
Remember, for loops are just functions; so
for (i in 1:3) {
print(i)
}
#[1] 1
#[1] 2
#[1] 3
is the same as
`for`(i, 1:3, print(i))
#[1] 1
#[1] 2
#[1] 3
See ?"for".

Answer 2

Your question, specifically, concerns the ussage of for and if. Here is my approach:
You can define for in two forms: the "classic C style" and the "vector style".
The "classic C style" would be something like this:
for(i = 1; i <= nrow(a); i = i + 1) {
# Your code goes here
}
The "vector style" would be something like this:
for(i in 1:nrow(a)) {
# Your code goes here
}
Notice that, in both cases, the for statement is the one that increments the value of i. Also, remember that in R the starting index is one (unlike many C-like languages, where the starting index is usually zero).
As for your if statement, R uses only one & for and (and only one | for or), so your if statement should be something like this:
if(old_id == a[1,i+2] & old_art == a[i+2,1]) {
# More code here
}
Finally, if you want to debug your code, check this link.