Module dictionary 

Available on crate feature dict_builder only.

Code for creating a separate content dictionary.

Effective dictionaries are up to 1% the size of the complete training body, and are trained on many examples of the original data.

Implemented following the paper “Effective construction of Relative Lempel-Ziv Dictionaries”, by Kewen Liao, Matthias Petri, Alistair Moffat, and Anthony Wirth

Structs

FastCoverOptions

Tuning knobs for pure-Rust FastCOVER training.

FastCoverParams

FastCoverTuned

FinalizeOptions

Constants

DEFAULT_D_CANDIDATES

DEFAULT_F_CANDIDATES

DEFAULT_K_CANDIDATES

Functions

create_fastcover_dict_from_source

Train and finalize a FastCOVER dictionary in pure Rust.

create_fastcover_raw_dict_from_source

Train a raw FastCOVER dictionary from a source stream.

create_raw_dict_from_dir

Creates a “raw content” dictionary, training off of every file in this directory and all sub-directories.

create_raw_dict_from_source

Read from source to create a “raw content” dictionary of dict_size. The completed dictionary is written to output.

finalize_raw_dict

Finalize raw dictionary content into a full zstd dictionary binary (magic + dict_id + entropy tables + offset history + content).

train_fastcover_raw_from_slice

Train a raw FastCOVER dictionary directly from an in-memory sample.