printing output if multiple criteria are met in data.table

Question

I'm trying to print out some output based on the existence of some numbers in other columns.

1) If the value of 0 exists in column V4 and if column V8 has a value greater than 30,

2) then print that line and 2 lines below it

3) but only if the same criteria is met for the line that had a value of 0, three lines above.

I can do steps 1 and 2 but I'm having trouble doing step 3. Here is the input:

> library(data.table)
> file
      V1   V2  V3 V4 V5      V6 V7 V8
 1:    0 -232 -77 -1  D     dog  0  0
 2:    1 -231 -77  0  C     cat  0  40
 3:    2 -230 -77  1  T     tai  0  0
 4:    3 -229 -76 -1  F     fis  0  0
 5:    4 -228 -76  0  G     goo  0  100
---                                  
1162: 1161  929 310 -1  S     soo  0  0
1163: 1162  930 310  0  B     bye  0  0
1164: 1163  931 310  1  G     goo  0  0
1165: 1164  932 311 -1  T     tuu  0  0
1166: 1165  933 311  0  R     roo  0  50

If I then run the following code I get an output that is close but not quite right.

#grouping data table into groups of 3
file[, grp := rep(seq_len(round(ceiling(.N/3))), each = 3,length.out=.N)]
file = file[, if(.N == 3 && V4==0) .SD, by = grp]

#generating the output formatted in the way I want
out <- file[, if(V4 == 0 && V8 > 30) c(V1[1], V1[3], V2[1], V2[3], as.list(V4), as.list(V5), as.list(V6), as.list(V8)), by=grp] 

In this step (see below) I want to add a criteria to only generate an output if column V4 = 0, column V8 of the same line is greater than 30 and if there is a value of >30 in column 8, three lines above.

file[, if(V4 == 0 && V8 > 30) ...

any ideas?

Answer 1

I would realize this in a three step approach:

df <- read.table(textConnection("
V1   V2  V3 V4 V5      V6 V7 V8
 0 -232 -77 -1  D     dog  0  0
 1 -231 -77  0  C     cat  0  40
 2 -230 -77  1  T     tai  0  0
 3 -229 -76 -1  F     fis  0  0
 4 -228 -76  0  G     goo  0  100
 1161  929 310 -1  S     soo  0  0
 1162  930 310  0  B     bye  0  0"), header=TRUE)

# condition 1
a <- df$V4 == 0 & df$V8 > 30

# condition 3 (does the row 3 rows above fulfill condition 1?)
aIdx <- which(a)
b <- (aIdx-3) %in% aIdx
a[a] <- b

# condition 2 (select also the next two rows)
i <- rep(which(a), 3) + 0:2
a[i] <- TRUE

df[a, ]
#     V1   V2  V3 V4 V5  V6 V7  V8
# 5    4 -228 -76  0  G goo  0 100
# 6 1161  929 310 -1  S soo  0   0
# 7 1162  930 310  0  B bye  0   0