to_ndjson {jsonify}

To ndjson

Description

Converts R objects to ndjson

Usage

to_ndjson(
  x,
  unbox = FALSE,
  digits = NULL,
  numeric_dates = TRUE,
  factors_as_string = TRUE,
  by = "row"
)

Arguments

x

object to convert to JSON

unbox

logical indicating if single-value arrays should be 'unboxed', that is, not contained inside an array.

digits

integer specifying the number of decimal places to round numerics. Default is NULL - no rounding

numeric_dates

logical indicating if dates should be treated as numerics. Defaults to TRUE for speed. If FALSE, the dates will be coerced to character in UTC time zone

factors_as_string

logical indicating if factors should be treated as strings. Defaults to TRUE.

by

either "row" or "column" indicating if data.frames and matrices should be processed row-wise or column-wise. Defaults to "row"

Details

Lists are converted to ndjson non-recursively. That is, each of the objects in the list at the top level are converted to a new-line JSON object. Any nested sub-elements are then contained within that JSON object. See examples

Examples

to_ndjson( 1:5 )
to_ndjson( letters )

mat <- matrix(1:6, ncol = 2)

to_ndjson( x = mat )
to_ndjson( x = mat, by = "col" )

df <- data.frame(
  x = 1:5
  , y = letters[1:5]
  , z = as.Date(seq(18262, 18262 + 4, by = 1 ), origin = "1970-01-01" )
  )

to_ndjson( x = df )
to_ndjson( x = df, numeric_dates = FALSE )
to_ndjson( x = df, factors_as_string = FALSE )
to_ndjson( x = df, by = "column" )
to_ndjson( x = df, by = "column", numeric_dates = FALSE )

## Lists are non-recurisve; only elements `x` and `y` are converted to ndjson
lst <- list(
  x = 1:5
  , y = list(
    a = letters[1:5]
    , b = data.frame(i = 10:15, j = 20:25)
  )
)
 
to_ndjson( x = lst )
to_ndjson( x = lst, by = "column")