email foo bar
1 a@g.com 23 34
2 b@g.com 43 34
3 c@g.com 35 32
Now I want to create JSON docs for each email as of following form:
doc 1: { "email" : "a@g.com"}
doc 2: { "email" : "b@g.com"}
doc 3: { "email" : "c@g.com"}
My final goal is to insert these docs in MongoDB using dbInsertDocument() from RMongo.
I have been toying with toJSON() but couldn't figure out a way. How do I convert final$email in JSON docs as described above?
I think , you should first convert your data.frame to a list of lists:
## basically you split your data.frame by row
## for each row you convert it as a list
LMAo <- unname(lapply(split(dx,seq_len(nrow(dx))), function(x) as.list(x)))
## auto unboxing to no treat scalar as vector
jsonlite::toJSON(LMAo,auto_unbox = T)
[
{"email":"a@g.com","foo":23,"bar":34},
{"email":"b@g.com","foo":43,"bar":34},
{"email":"c@g.com","foo":35,"bar":32}
]
## same result using RJSONIO
cat(RJSONIO::toJSON(LMAo))
I'm not sure if this is what you need:
final$email has not column name so it is not included in the output of toJSON(final$email)
library(jsonlite)
txt <- "email foo bar
1 a@g.com 23 34
2 b@g.com 43 34
3 c@g.com 35 32"
final <- read.table(text=txt, stringsAsFactors = FALSE)
toJSON(final$email)
# [1] "[\"a@g.com\",\"b@g.com\",\"c@g.com\"]"
# final$email has to be converted to data.frame
dt <- data.frame(final$email)
colnames(dt) <- "email"
toJSON(dt)
# [{"email":"a@g.com"},{"email":"b@g.com"},{"email":"c@g.com"}]
