json - How to join nested values into one column with tidyjson? -

i want join 2 values in 1 column instead of spreading

'{"name": {"first": "bob", "last": "jones"}, "age": 32}' %>% spread_values( first.name = jstring("name", "first"),  age = jnumber("age") ) %>% unite(conc, c("first.name", "age"), sep=" ") 

but keep having following error

all select() inputs must resolve integer column positions. following not: c("first.name", "age")

my desire output have 1 new column "conc" replacing both first.name , age , concatenate every string value. such "jones 32"

which weird because if remove last line gives me proper data.frame , can access first.name , age.

any hint?

though tbl_json objects inherit tbl_df, not play nicely once done parsing , begin further manipulation in tidyjson , dplyr. reason have additional methods , attributes tagging along not necessary anymore.

as result, once finished parsing, habit use tbl_df or as_data_frame strip off tbl_json component of object. prefer tbl_df because shorter.

i using development version github: devtools::install_github('jeremystan/tidyjson')

i not reproduce error mention, seems work:

library(tidyjson) library(tidyr)  "{\"name\": {\"first\": \"bob\", \"last\": \"jones\"}, \"age\": 32}" %>%  spread_values(first.name = jstring("name","first")   , age = jnumber("age") ) %>%  tbl_df() %>%  unite(conc, c("first.name", "age"), sep = " ")  #> # tibble: 1 x 2 #>   document.id   conc #> *       <int>  <chr> #> 1           1 bob 32 

other food thought when concatenating strings use paste() or paste0() within mutate. mutate supported tidyjson, well, can used in pipeline while parsing json. paste(.,collapse=',') can helpful when aggregating multiple rows summarize, too.