Parse R Character Vector into DataFrame

Question

I'm been working with JSON response from API, and have been able to return the data in a format that looks very JSON (which is good) and not struggling to get into DataFrame. Here is the code:

message(typeof((results_response)))
# character
message(length((results_response)))
# 1 

message(results_response)
# {"central_time":{"0":"2020-04-06 15:47:00","1":"2020-04-06 15:47:15"},"Va_V":{"0":286.032,"1":286.207}}

The goal would not hard coding the keys, because there could be other responses when different keys. thank you!

@akrun yes. When i do that is causing this: `0` = "2020-04-09 15:47:00", `1` = "2020-04-09 15:47:15")list(`0` = 286.032 `1` = 286.207) — jKraut, May 16 '20 at 21:39

score 0 · Accepted Answer · answered May 16 '20 at 21:33

0

May be this helps

library(jsonlite)
library(purrr)
library(dplyr)
fromJSON(results_response) %>%
     transpose %>% 
     bind_rows
 # A tibble: 2 x 2
 #  central_time         Va_V
 #* <chr>               <dbl>
#1 2020-04-06 15:47:00  286.
#2 2020-04-06 15:47:15  286.

data

results_response <- "{\"central_time\":{\"0\":\"2020-04-06 15:47:00\",\"1\":\"2020-04-06 15:47:15\"},\"Va_V\":{\"0\":286.032,\"1\":286.207}}"

answered May 16 '20 at 21:33

akrun

874,273
37
540
662

Ran the code, got this: Warning: Error in : Argument 10 is a list, must contain atomic vectors – jKraut May 16 '20 at 22:14
@jKraut i couldn't reproduce the warning/error. May be some package version difference issue – akrun May 16 '20 at 22:15
I think it's additional fields in the JSON response coming through causing the issue. – jKraut May 16 '20 at 22:22
@jKraut ok, I run this based on the example as I showed in the post – akrun May 16 '20 at 22:23
1

In another field it had a "." in the key name. I removed that and your code worked perfectly. thanks – jKraut May 16 '20 at 22:50

MacOS · Answer 2 · 2020-05-17T09:06:54.690

The following should work

library(jsonlite)

# Your JSON charachter vector.
json <- 

df <- as.data.frame(fromJSON(json))

See jsonlite Documentation and the answer to How can I convert Json to data frame in R.

I hope this helps!

EDITED The following code did run without errors.

library(jsonlite)

json <- as.character(
  '{"central_time":{"0":"2020-04-06 15:47:00","1":"2020-04-06 15:47:15"}, "Va_V":{"0":286.032, "1":286.207}}')

as.data.frame(fromJSON(json))

Can it be that you have a duplicate key?

I also tried the following (this should be equivalent to the correct answer).

library(jsonlite)

json <- as.character(
  '{"central_time":{"0":"2020-04-06 15:47:00","1":"2020-04-06 15:47:15"}, "Va_V":{"0":286.032, "1":286.207}}')

from.json <- fromJSON(json)
from.json <- t(from.json)
df <- do.call(cbind, from.json)

I know that you have already marked the correct answer, but can you please try the code nevertheless? I'm just curious what went wrong.

Ran the code, got this: Warning: Error in : arguments imply differing number of rows: 1, 0 — jKraut, May 16 '20 at 22:17
That is weired. Because I don't. Please see the edited answer. — MacOS, May 17 '20 at 09:02

Parse R Character Vector into DataFrame

2 Answers2

data