Struct MapSpec 

pub struct MapSpec { /* private fields */ }

A map operation.

use ytsaurus_client::MapSpec;

let spec = MapSpec::new("./cat", ["//tmp/in"], ["//tmp/out"])
    .with_local_file("//tmp/cat")
    .with_memory_limit(512 * 1024 * 1024);

Implementations

impl MapSpec

pub fn new<I, O>(command: impl Into<String>, inputs: I, outputs: O) -> Self

where I: IntoIterator, I::Item: Into<String>, O: IntoIterator, O::Item: Into<String>,

A map running command over inputs, writing outputs.

pub fn with_local_file(self, path: impl Into<String>) -> Self

Adds a Cypress file the job needs — normally the worker binary.

pub fn with_local_file_named( self, path: impl Into<String>, name: impl AsRef<str>, ) -> Self

Adds a Cypress file under a different name in the job’s sandbox.

The name matters because the job runs a command: a file cached under its MD5 hash arrives as 4c8f…, and ./my_job would not find it.

pub fn with_memory_limit(self, bytes: i64) -> Self

Sets the mapper’s memory limit, in bytes.

pub fn with_formats(self, input: DataFormat, output: DataFormat) -> Self

Selects the mapper’s input and output data formats.

YSON selections apply to every table. A Skiff selection must contain one table schema per corresponding input or output table, in the same order. The default remains binary YSON.

pub fn with_skiff_formats(self, input: SkiffFormat, output: SkiffFormat) -> Self

Uses validated Skiff formats for the mapper’s input and output streams.

This compatibility convenience delegates to Self::with_formats.

pub fn with_env(self, key: impl Into<String>, value: impl Into<String>) -> Self

Sets an environment variable for the job, e.g. RUST_BACKTRACE.

pub fn with_input_table_index(self) -> Self

Asks for the input table index to be delivered with each row.

Without this, Row::table_index is always 0.

pub fn with_job_count(self, count: i64) -> Self

Requests a specific job count.

pub fn skiff_table_mismatch(&self) -> Option<String>

Describes a Skiff format that does not match this spec’s table lists.

A Skiff format needs one table schema per table, in order. Get the count wrong and the operation is submitted anyway: the cluster may refuse it, or the job may read a table the format does not describe and fail mid-stream, after it has already written output. Client::start_map checks this before sending the spec; check it here if you render the spec yourself with MapSpec::to_yson.

pub fn with_raw(self, key: impl Into<String>, value: YsonValue) -> Self

Sets any spec field this builder does not model.

pub fn to_yson(&self) -> YsonValue

Renders the spec.

Trait Implementations

impl Clone for MapSpec

fn clone(&self) -> MapSpec

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for MapSpec

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

Formats the value using the given formatter.