Struct JsonOptions

pub struct JsonOptions {

    pub schema: Option<String>,
    pub compression: Option<String>,
    pub primitives_as_string: Option<bool>,
    pub prefers_decimal: Option<bool>,
    pub allow_comments: Option<bool>,
    pub allow_unquoted_field_names: Option<bool>,
    pub allow_single_quotes: Option<bool>,
    pub allow_numeric_leading_zeros: Option<bool>,
    pub allow_backslash_escaping_any_character: Option<bool>,
    pub mode: Option<String>,
    pub column_name_of_corrupt_record: Option<String>,
    pub date_format: Option<String>,
    pub timestamp_format: Option<String>,
    pub multi_line: Option<bool>,
    pub allow_unquoted_control_chars: Option<bool>,
    pub line_sep: Option<String>,
    pub sampling_ratio: Option<f64>,
    pub drop_field_if_all_null: Option<bool>,
    pub encoding: Option<String>,
    pub locale: Option<String>,
    pub allow_non_numeric_numbers: Option<bool>,
    pub time_zone: Option<String>,
    pub timestamp_ntz_format: Option<String>,
    pub enable_datetime_parsing_fallback: Option<bool>,
    pub ignore_null_fields: Option<bool>,
    pub common: CommonFileOptions,
}

A struct that represents options for configuring JSON file parsing.

By default, this supports JSON Lines (newline-delimited JSON). For single-record-per-file JSON, set the multi_line option to true.

If the schema option is not specified, the input schema is inferred from the data.

Options

• schema: An optional schema for the JSON data, either as a StructType or a DDL string.
• compression: Compression codec to use when reading JSON files (e.g., gzip, bzip2).
• primitives_as_string: Treat primitive types (e.g., integers, booleans) as strings.
• prefers_decimal: Prefer parsing numbers as decimals rather than floating points.
• allow_comments: Allow comments in JSON files (e.g., lines starting with // or /* */).
• allow_unquoted_field_names: Allow field names without quotes.
• allow_single_quotes: Allow the use of single quotes instead of double quotes for strings.
• allow_numeric_leading_zeros: Allow numbers to have leading zeros (e.g., 007).
• allow_backslash_escaping_any_character: Allow backslashes to escape any character.
• mode: The parsing mode (e.g., PERMISSIVE, DROPMALFORMED, FAILFAST).
• column_name_of_corrupt_record: Name of the column where corrupted records are placed.
• date_format: Custom date format (e.g., yyyy-MM-dd).
• timestamp_format: Custom timestamp format (e.g., yyyy-MM-dd HH:mm:ss).
• multi_line: Read multiline JSON files (e.g., when a single JSON object spans multiple lines).
• allow_unquoted_control_chars: Allow unquoted control characters in JSON (e.g., ASCII control characters).
• line_sep: Custom line separator (default is \n).
• sampling_ratio: Fraction of the data used for schema inference (e.g., 0.1 for 10%).
• drop_field_if_all_null: Drop fields that are NULL in all rows.
• encoding: Character encoding (default is UTF-8).
• locale: Locale for parsing dates and numbers (e.g., en-US).
• allow_non_numeric_numbers: Allow special non-numeric numbers (e.g., NaN, Infinity).
• time_zone: Time zone used for parsing dates and timestamps (e.g., UTC, America/Los_Angeles).
• timestamp_ntz_format: Format for parsing timestamp without time zone (NTZ) values (e.g., yyyy-MM-dd'T'HH:mm:ss).
• enable_datetime_parsing_fallback: Enable fallback mechanism for datetime parsing if the initial parsing fails.
• ignore_null_fields: Ignore NULL fields in the JSON structure, treating them as absent.
• common - Common file options that are shared across multiple file formats.

Example

let options = JsonOptions::new()
    .schema("name STRING, salary INT")
    .multi_line(true)
    .allow_comments(true)
    .encoding("UTF-8")
    .time_zone("UTC")
    .compression("gzip");

let df = spark.read().json(["/path/to/json"], options)?;

Fields

schema: Option<String>

compression: Option<String>

primitives_as_string: Option<bool>

prefers_decimal: Option<bool>

allow_comments: Option<bool>

allow_unquoted_field_names: Option<bool>

allow_single_quotes: Option<bool>

allow_numeric_leading_zeros: Option<bool>

allow_backslash_escaping_any_character: Option<bool>

mode: Option<String>

column_name_of_corrupt_record: Option<String>

date_format: Option<String>

timestamp_format: Option<String>

multi_line: Option<bool>

allow_unquoted_control_chars: Option<bool>

line_sep: Option<String>

sampling_ratio: Option<f64>

drop_field_if_all_null: Option<bool>

encoding: Option<String>

locale: Option<String>

allow_non_numeric_numbers: Option<bool>

time_zone: Option<String>

timestamp_ntz_format: Option<String>

enable_datetime_parsing_fallback: Option<bool>

ignore_null_fields: Option<bool>

common: CommonFileOptions

Implementations

impl JsonOptions

pub fn schema(self, value: &str) -> Self

pub fn primitives_as_string(self, value: bool) -> Self

pub fn prefers_decimal(self, value: bool) -> Self

pub fn allow_comments(self, value: bool) -> Self

pub fn allow_unquoted_field_names(self, value: bool) -> Self

pub fn allow_single_quotes(self, value: bool) -> Self

pub fn allow_numeric_leading_zeros(self, value: bool) -> Self

pub fn allow_backslash_escaping_any_character(self, value: bool) -> Self

pub fn mode(self, value: &str) -> Self

pub fn column_name_of_corrupt_record(self, value: &str) -> Self

pub fn date_format(self, value: &str) -> Self

pub fn timestamp_format(self, value: &str) -> Self

pub fn multi_line(self, value: bool) -> Self

pub fn allow_unquoted_control_chars(self, value: bool) -> Self

pub fn line_sep(self, value: &str) -> Self

pub fn sampling_ratio(self, value: f64) -> Self

pub fn drop_field_if_all_null(self, value: bool) -> Self

pub fn encoding(self, value: &str) -> Self

pub fn locale(self, value: &str) -> Self

pub fn allow_non_numeric_numbers(self, value: bool) -> Self

pub fn time_zone(self, value: &str) -> Self

pub fn timestamp_ntz_format(self, value: &str) -> Self

pub fn enable_datetime_parsing_fallback(self, value: bool) -> Self

pub fn compression(self, value: &str) -> Self

pub fn ignore_null_fields(self, value: bool) -> Self

Trait Implementations

impl Clone for JsonOptions

fn clone(&self) -> JsonOptions

Returns a copy of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl ConfigOpts for JsonOptions

fn to_options(&self) -> HashMap<String, String>

impl Debug for JsonOptions

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Default for JsonOptions

fn default() -> Self

Returns the “default value” for a type.