R: Nested data.table to JSON - r

Question

I want to get from a data.table like this
temp <- data.table(data = list(data.table(a = 1:2,b=1:2)), type = "A")
data
type
<data.table[2x2]>
A
to a JSON like this
{
"group":
{
"data": [
{
"a": 1,
"b": 1
},
{
"a": 2,
"b": 2
}
],
"type": "A"
}
}
The Problem is I always end up with an additional array "[" for group.
What I have tried is tidyr::nest and
temp2 <- temp[, list(group=list(.SD))]
jsonlite::toJSON(temp2,pretty = TRUE, auto_unbox = TRUE)
temp3 <- temp[, (list(group=list(as.list(.SD))))]
jsonlite::toJSON(temp3,pretty = TRUE, auto_unbox = TRUE)
Is there an "easy" solution for my problem?
Thanks
edit more complex example
temp <-
data.table(
id1 = 1:6,
id2 = c(rep("A", 2), rep("B", 2), rep("C", 2)),
data = rep(list(data.table(
a = 1:2, b = 1:2
)), 6),
type = "test"
)
nest1 <- temp[, list(list(.SD)),by=.(id1,id2)] %>% setnames("V1","group")
nest1[, type:="B"]
nest2 <- nest1[, list(list(.SD)),by=.(id2)] %>% setnames("V1","data")
nest2[, type:="C"]
nest3 <- nest2[, list(list(.SD)),by=.(id2)] %>% setnames("V1","group")
jsonlite::toJSON(nest3, pretty = TRUE)
desired output (shortend):
Group should only contain objects and no arrays
[
{
"id2": "A",
"group": {
"data": [
{
"id1": 1,
"group": {
"data": [
{
"a": 1,
"b": 1
},
{
"a": 2,
"b": 2
}
],
"type": "test"
},
"type": "B"
},
{
"id1": 2,
"group": {
"data": [
{
"a": 1,
"b": 1
},
{
"a": 2,
"b": 2
}
],
"type": "test"
},
"type": "B"
}
],
"type": "C"
}
},
{
"id2": "B",
"group": {
"data": [],
"type": "C"
}
}
]

Answer 1

We could use jq to do the unboxing as a post-processing step, since jsonlite doesn't seem to allow for this specific use case:
jsonlite::toJSON(nest3, pretty = TRUE) %>%
jqr::jq('walk(if type=="array" and length==1 then .[0] else . end)')
The jq bit is taken from jq ~ is there a better way to collapse single object arrays?