data.table - R - How to add rows for missing values for unique group sequences?

Question

My problem is similar to this previous question Fastest way to add rows for missing values in a data.frame?

I can't figure out how to add rows padded with "NA" when the min/max is different by group.

> red<-data.frame(project = c(6, 6, 6, 6, 6, 9, 9, 9), period =c(1, 2, 5:7, 2, 4, 5), v3=letters[1:8], v4=c("red", "yellow", recursive = T))
> red
  project period v3     v4
1       6      1  a    red
2       6      2  b yellow 
3       6      5  c    red
4       6      6  d yellow
5       6      7  e    red
6       9      2  f yellow
7       9      4  g    red
8       9      5  h yellow

I want it to look like:

project period v3     v4
      6      1  a    red
      6      2  b yellow
      6      3 NA     NA
      6      4 NA     NA
      6      5  c    red
      6      6  d yellow
      6      7  e    red
      9      2  f yellow
      9      3 NA     NA
      9      4  g    red
      9      5  h yellow

When I used

library(data.table)
DT=as.data.table(red)
setkey(DT, project, period)

DT[CJ(unique(project), seq(min(period), max(period)))]

it made each project group have 7 periods; Project 6 should have periods 1-7, but Project 9 should have periods 2-5.

I've tried fiddling with .SD[ which.max(period)], by=project] but no cigar.

I thought it should be something simple in the seq(), but I tried seq(min(period, by=project)) with no luck

Thank you!

Answer 1

I don't know if this the idiomatic way or not, but I was able to achieve your desired output, by first creating an index and then subsetting the correct rows out of .SD per that index

DT[, indx := .GRP, project][, 
     .SD[CJ(unique(project), seq(min(period), max(period)))], indx]

#     indx project period v3     v4
#  1:    1       6      1  a    red
#  2:    1       6      2  b yellow
#  3:    1       6      3 NA     NA
#  4:    1       6      4 NA     NA
#  5:    1       6      5  c    red
#  6:    1       6      6  d yellow
#  7:    1       6      7  e    red
#  8:    2       9      2  f yellow
#  9:    2       9      3 NA     NA
# 10:    2       9      4  g    red
# 11:    2       9      5  h yellow