Skip to contents

Create named list of options for serializing R to JSON

Source: R/json-opts.R
opts_write_json.Rd

Create named list of options for serializing R to JSON

Usage

opts_write_json(
  digits = -1L,
  digits_secs = 0L,
  digits_signif = -1L,
  pretty = FALSE,
  auto_unbox = FALSE,
  dataframe = c("rows", "columns"),
  factor = c("string", "integer"),
  name_repair = c("none", "minimal"),
  num_specials = c("null", "string"),
  str_specials = c("null", "string"),
  fast_numerics = FALSE,
  json_verbatim = FALSE,
  null = c("null", "empty_array"),
  yyjson_write_flag = 0L
)

Arguments

digits

decimal places to keep for floating point numbers. Default: -1. Positive values specify number of decimal places. Using zero will write the numeric value as an integer. Values less than zero mean that the floating point value should be written as-is (the default). This argument is ignored if digits_signif is greater than zero.

digits_secs

decimal places for fractional seconds when converting times to a string representation. Default: 0. Valid range: 0 to 6

digits_signif

significant decimal places to store in floating point numbers. Default: -1 means to output the number as-is (while respecting the digits) argument. Values above 0 will produce rounding to the given number of places and the digits argument will be ignored.

pretty

Logical value indicating if the created JSON string should have whitespace for indentation and linebreaks. Default: FALSE. Note: this option is equivalent to yyjson_write_flag = write_flag$YYJSON_WRITE_PRETTY

auto_unbox

automatically unbox all atomic vectors of length 1 such that they appear as atomic elements in JSON rather than arrays of length 1.

dataframe

how to encode data.frame objects. Options 'rows' or columns'. Default: 'rows'

factor

how to encode factor objects: must be one of 'string' or 'integer' Default: 'string'

name_repair

How should unnamed items in a partially named list be handled? 'none' means to leave their names blank in JSON (which may not be valid JSON). 'minimal' means to use the integer position index of the item as its name if it is missing. Default: 'none'

num_specials

Should special numeric values (i.e. NA, NaN, Inf) be converted to a JSON null value or converted to a string representation e.g. "NA"/"NaN" etc. Default: 'null'

str_specials

Should a special value of NA in a character vector be converted to a JSON null value, or converted to a string "NA"? Default: 'null'

fast_numerics

Does the user guarantee that there are no NA, NaN or Inf values in the numeric vectors? Default: FALSE. If TRUE then numeric and integer vectors will be written to JSON using a faster method. Note: if there are NA, NaN or Inf values, an error will be thrown. Expert users are invited to also consider the YYJSON_WRITE_ALLOW_INF_AND_NAN and YYJSON_WRITE_INF_AND_NAN_AS_NULL options for yyjson_write_flags and should consult the yyjson API documentation for further details.

json_verbatim

Write strings with class 'json' directly into the result? Default: FALSE. If json_verbatim = TRUE and a string has the class "json", then it will be written verbatim into the output.

null

How should R NULL values be written? Default: "null" means to write as a JSON "null" string. Other valid value is "empty_array".

yyjson_write_flag

integer vector corresponding to internal yyjson options. See yyjson_write_flag in this package, and read the yyjson API documentation for more information. This is considered an advanced option.

Value

Named list of options for writing JSON

See also

yyjson_write_flag()

Examples

write_json_str(head(iris, 3), opts = opts_write_json(factor = 'integer'))
#> [1] "[{\"Sepal.Length\":5.1,\"Sepal.Width\":3.5,\"Petal.Length\":1.4,\"Petal.Width\":0.2,\"Species\":1},{\"Sepal.Length\":4.9,\"Sepal.Width\":3.0,\"Petal.Length\":1.4,\"Petal.Width\":0.2,\"Species\":1},{\"Sepal.Length\":4.7,\"Sepal.Width\":3.2,\"Petal.Length\":1.3,\"Petal.Width\":0.2,\"Species\":1}]"