CODEWITHSUNDEEP
jsonr
user3245256
user3245256

Reputation: 1968

Adding lines into a Json file generated in R

I want to generate a JSON file in R. For example:

forJson <- data.frame(names = c("John", "Mary", "Pat", "Rick"), values = 3:6)
library("jsonlite")
toJson <- toJSON(forJson, pretty = TRUE)
toJson

However, I am required to add some info to this file. At the end it should look like this:

[
 "CaseID": 12345,
 "graph": "Name-of-the-graph",
 "Items":[{

  {
    "names": "John",
    "values": 3
  },
  {
    "names": "Mary",
    "values": 4
  },
  {
    "names": "Pat",
    "values": 5
  },
  {
    "names": "Rick",
    "values": 6
  }]
]

I am not experienced with JSON at all. It feels like my current values for the table should be inside an object called "Items". And then I have to have 2 pieces of metainformation - CaseID and graph name - on the top. Any advice how I could add such lines or generate them to begin with using toJSON?

Thanks a lot!

Upvotes: 2

Views: 291

Answers (1)

SymbolixAU
SymbolixAU

Reputation: 26258

Put your data in a list, and then convert it to JSON

forJson <- list(CaseID = 1234, graph = "name-of-the-graph", Items = forJson)
toJSON(forJson, pretty = T)
# 
# {
#   "CaseID": [1234],
#   "graph": ["name-of-the-graph"],
#   "Items": [
#       {
#           "names": "John",
#           "values": 3
#       },
#       {
#           "names": "Mary",
#           "values": 4
#       },
#       {
#           "names": "Pat",
#           "values": 5
#       },
#       {
#           "names": "Rick",
#           "values": 6
#       }
#       ]
# }

A slight variation is to have a data.frame inside a list column of your overall data.frame

forJson <- data.frame(CaseID = 1234, graph = "name_of_graph")
forJson$Items <- list(data.frame(names = c("John", "Mary", "Pat", "Rick"), 
                                values = 3:6))

toJSON(forJson, pretty = T)

# [
#   {
#       "CaseID": 1234,
#       "graph": "name_of_graph",
#       "Items": [
#           {
#               "names": "John",
#               "values": 3
#           },
#           {
#               "names": "Mary",
#               "values": 4
#           },
#           {
#               "names": "Pat",
#               "values": 5
#           },
#           {
#               "names": "Rick",
#               "values": 6
#           }
#           ]
#   }
#   ]

The easy way to work out the R structure you need is to work backwards from the JSON by reading it in using fromJSON(), which will automatically give you the R structure

js <- '{
      "CaseID": 1234,
      "graph": "name_of_graph",
      "Items": [
          {
              "names": "John",
              "values": 3
          },
          {
              "names": "Mary",
              "values": 4
          },
          {
              "names": "Pat",
              "values": 5
          },
          {
              "names": "Rick",
              "values": 6
          }
          ]
  }'

lst <- jsonlite::fromJSON(js)
lst
$CaseID
[1] 1234

$graph
[1] "name_of_graph"

$Items
  names values
1  John      3
2  Mary      4
3   Pat      5
4  Rick      6

However, note that even though fromJSON() read in the values for CaseID and graph as single elements, the toJSON() automatically makes them into array elements (but in reality I can't see this being a problem).

toJSON(lst, pretty = T)
{
  "CaseID": [1234],
  "graph": ["name_of_graph"],
  "Items": [
    {
      "names": "John",
      "values": 3
    },
    ... etc

Upvotes: 2

Related Questions