Struct icu_datagen::DatagenDriver

pub struct DatagenDriver { /* private fields */ }

Configuration for a data export operation.

Note that this only configures which data is exported. The input provider, usually DatagenProvider, might expose more options about the data itself.

Examples

use icu_datagen::blob_exporter::*;
use icu_datagen::prelude::*;

DatagenDriver::new()
    .with_keys([icu::list::provider::AndListV1Marker::KEY])
    .with_all_locales()
    .export(
        &DatagenProvider::new_latest_tested(),
        BlobExporter::new_with_sink(Box::new(&mut Vec::new())),
    )
    .unwrap();

Implementations

impl DatagenDriver

pub fn new() -> Self

Creates an empty DatagenDriver.

Note that keys and locales need to be set before calling export.

pub fn with_keys(self, keys: impl IntoIterator<Item = DataKey>) -> Self

Sets this driver to generate the given keys.

See icu_datagen::keys, icu_datagen::all_keys, icu_datagen::key and icu_datagen::keys_from_bin.

pub fn with_locales( self, locales: impl IntoIterator<Item = LanguageIdentifier> ) -> Self

Sets this driver to generate the given locales.

Use the langid! macro from the prelude to create an explicit list, or DatagenProvider::locales_for_coverage_levels for CLDR coverage levels.

pub fn with_all_locales(self) -> Self

Sets this driver to generate all available locales.

pub fn with_fallback_mode(self, fallback: FallbackMode) -> Self

Sets the fallback type that the data should be generated for.

If locale fallback is used at runtime, smaller data can be generated.

pub fn with_additional_collations( self, additional_collations: impl IntoIterator<Item = String> ) -> Self

This option is only relevant if using icu::collator.

By default, the collations big5han, gb2312, and those starting with search are excluded. This method can be used to reennable them.

The special string "search*" causes all search collation tables to be included.

pub fn with_recommended_segmenter_models(self) -> Self

This option is only relevant if using icu::segmenter.

Sets this driver to generate the recommended segmentation models, to the extent required by the chosen data keys.

pub fn with_segmenter_models( self, models: impl IntoIterator<Item = String> ) -> Self

This option is only relevant if using icu::segmenter.

Sets this driver to generate the given segmentation models, to the extent required by the chosen data keys.

The currently supported dictionary models are

• cjdict
• burmesedict
• khmerdict
• laodict
• thaidict

The currently supported LSTM models are

• Burmese_codepoints_exclusive_model4_heavy
• Khmer_codepoints_exclusive_model4_heavy
• Lao_codepoints_exclusive_model4_heavy
• Thai_codepoints_exclusive_model4_heavy

If a model is not included, the resulting line or word segmenter will apply rule-based segmentation when encountering text in a script that requires the model, which will be incorrect.

If multiple models for the same language and segmentation type (dictionary/LSTM) are listed, the first one will be used.

pub fn export( self, provider: &impl ExportableProvider, sink: impl DataExporter ) -> Result<(), DataError>

Exports data from the given provider to the given exporter.

See DatagenProvider, make_exportable_provider!, BlobExporter, FileSystemExporter, and BakedExporter.

Trait Implementations

impl Clone for DatagenDriver

fn clone(&self) -> DatagenDriver

Returns a copy of the value.

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

Performs copy-assignment from source.

impl Debug for DatagenDriver

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

Formats the value using the given formatter.