If we use variables with names "i" or "j" in shift with data.table v1.14.7 we get an error.
library(data.table)
print(sessionInfo())
options(datatable.verbose = TRUE)
# using i in shift fails
X <- data.table(x = 1:4, g = as.factor(c(1,1,2,2)))
i <- 1
X[, y := shift(x, n = abs(i), type = "lag"), .(g)]
X[]
#Argument 'by' after substitute: .(g)
#Detected that j uses these columns: [x]
#Finding groups using forderv ... forder.c received 4 rows and 1 columns
#0.001s elapsed (0.000s cpu)
#Finding group sizes from the positions (can be avoided to save RAM) ... 0.000s elapsed (0.000s cpu)
#lapply optimization is on, j unchanged as 'shift(x, n = abs(i), type = "lag")'
#GForce optimized j to 'gshift(x, n = abs(i), type = "lag")'
#Making each group and running j (GForce TRUE) ... gforce initial population of grp took 0.000
#gforce assign high and low took 0.000
#Error in abs(i) : non-numeric argument to mathematical function
# using j in shift fails
X <- data.table(x = 1:4, g = as.factor(c(1,1,2,2)))
j <- 1
X[, y := shift(x, n = abs(j), type = "lag"), .(g)]
X[]
#Argument 'by' after substitute: .(g)
#Detected that j uses these columns: [x]
#Finding groups using forderv ... forder.c received 4 rows and 1 columns
#0.000s elapsed (0.001s cpu)
#Finding group sizes from the positions (can be avoided to save RAM) ... 0.000s elapsed (0.000s cpu)
#lapply optimization is on, j unchanged as 'shift(x, n = abs(j), type = "lag")'
#GForce optimized j to 'gshift(x, n = abs(j), type = "lag")'
#Making each group and running j (GForce TRUE) ... gforce initial population of grp took 0.000
#gforce assign high and low took 0.000
#Error: Check that is.data.table(DT) == TRUE. Otherwise, :=, `:=`(...) and let(...) are defined for use in j, once only and in particular ways. #See help(":=").
# using k in shift works
X <- data.table(x = 1:4, g = as.factor(c(1,1,2,2)))
k <- 1
X[, y := shift(x, n = abs(k), type = "lag"), .(g)]
X[]
#Argument 'by' after substitute: .(g)
#Detected that j uses these columns: [x]
#Finding groups using forderv ... forder.c received 4 rows and 1 columns
#0.000s elapsed (0.000s cpu)
#Finding group sizes from the positions (can be avoided to save RAM) ... 0.000s elapsed (0.000s cpu)
#lapply optimization is on, j unchanged as 'shift(x, n = abs(k), type = "lag")'
#GForce optimized j to 'gshift(x, n = abs(k), type = "lag")'
#Making each group and running j (GForce TRUE) ... gforce initial population of grp took 0.000
#gforce assign high and low took 0.000
#gforce eval took 0.000
#0.000s elapsed (0.000s cpu)
#Assigning to 4 row subset of 4 rows
#RHS_list_of_columns == false
#> x g y
# <int> <fctr> <int>
#1: 1 1 NA
#2: 2 1 1
#3: 3 2 NA
#4: 4 2 3
#>
R version 4.1.2 (2021-11-01)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: CentOS Linux 7 (Core)
Matrix products: default
BLAS: /usr/R/4.1.2/lib64/R/lib/libRblas.so
LAPACK: /usr/R/4.1.2/lib64/R/lib/libRlapack.so
locale:
[1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C
[3] LC_TIME=en_US.UTF-8 LC_COLLATE=en_US.UTF-8
[5] LC_MONETARY=en_US.UTF-8 LC_MESSAGES=en_US.UTF-8
[7] LC_PAPER=en_US.UTF-8 LC_NAME=C
[9] LC_ADDRESS=C LC_TELEPHONE=C
[11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C
attached base packages:
[1] stats graphics grDevices utils datasets methods base
other attached packages:
[1] data.table_1.14.7
loaded via a namespace (and not attached):
[1] compiler_4.1.2
# Summary
If we use variables with names "i" or "j" in shift with data.table v1.14.7 we get an error.
Last workable version: v1.14.6
# Minimal reproducible example
library(data.table) print(sessionInfo()) options(datatable.verbose = TRUE) # using i in shift fails X <- data.table(x = 1:4, g = as.factor(c(1,1,2,2))) i <- 1 X[, y := shift(x, n = abs(i), type = "lag"), .(g)] X[] #Argument 'by' after substitute: .(g) #Detected that j uses these columns: [x] #Finding groups using forderv ... forder.c received 4 rows and 1 columns #0.001s elapsed (0.000s cpu) #Finding group sizes from the positions (can be avoided to save RAM) ... 0.000s elapsed (0.000s cpu) #lapply optimization is on, j unchanged as 'shift(x, n = abs(i), type = "lag")' #GForce optimized j to 'gshift(x, n = abs(i), type = "lag")' #Making each group and running j (GForce TRUE) ... gforce initial population of grp took 0.000 #gforce assign high and low took 0.000 #Error in abs(i) : non-numeric argument to mathematical function # using j in shift fails X <- data.table(x = 1:4, g = as.factor(c(1,1,2,2))) j <- 1 X[, y := shift(x, n = abs(j), type = "lag"), .(g)] X[] #Argument 'by' after substitute: .(g) #Detected that j uses these columns: [x] #Finding groups using forderv ... forder.c received 4 rows and 1 columns #0.000s elapsed (0.001s cpu) #Finding group sizes from the positions (can be avoided to save RAM) ... 0.000s elapsed (0.000s cpu) #lapply optimization is on, j unchanged as 'shift(x, n = abs(j), type = "lag")' #GForce optimized j to 'gshift(x, n = abs(j), type = "lag")' #Making each group and running j (GForce TRUE) ... gforce initial population of grp took 0.000 #gforce assign high and low took 0.000 #Error: Check that is.data.table(DT) == TRUE. Otherwise, :=, `:=`(...) and let(...) are defined for use in j, once only and in particular ways. #See help(":="). # using k in shift works X <- data.table(x = 1:4, g = as.factor(c(1,1,2,2))) k <- 1 X[, y := shift(x, n = abs(k), type = "lag"), .(g)] X[] #Argument 'by' after substitute: .(g) #Detected that j uses these columns: [x] #Finding groups using forderv ... forder.c received 4 rows and 1 columns #0.000s elapsed (0.000s cpu) #Finding group sizes from the positions (can be avoided to save RAM) ... 0.000s elapsed (0.000s cpu) #lapply optimization is on, j unchanged as 'shift(x, n = abs(k), type = "lag")' #GForce optimized j to 'gshift(x, n = abs(k), type = "lag")' #Making each group and running j (GForce TRUE) ... gforce initial population of grp took 0.000 #gforce assign high and low took 0.000 #gforce eval took 0.000 #0.000s elapsed (0.000s cpu) #Assigning to 4 row subset of 4 rows #RHS_list_of_columns == false #> x g y # <int> <fctr> <int> #1: 1 1 NA #2: 2 1 1 #3: 3 2 NA #4: 4 2 3 #># Output of sessionInfo()