lojban-rs

A Rust library suite for Lojban, the logical language. It covers morphology, word lists, zbalermorna conversion, and the CLL machine grammar.

This vendored copy is used in a std LSP environment.

Crates

• lojban-rs: facade crate.
• lojban-core: shared letters, selma'o, word classes, spans, diagnostics, and token IDs.
• lojban-morphology: word classification, tokenization, source token tables, and lujvo decomposition.
• lojban-words: embedded dictionary, rafsi, definitions, glosses, prefix search, and normalized lookup.
• lojban-zbalermorna: Latin/zbalermorna conversion with mapped spans.
• lojban-grammar: CLL chapter 21 parser, magic-word preprocessing, parser diagnostics, and recovered CST output.

Example

use lojban_rs::grammar::{parse, Rule};
use lojban_rs::morphology::{classify, WordClass};
use lojban_rs::zbalermorna::to_zbalermorna;

let tree = parse("mi klama le zarci")?;
let selbri = tree.find(Rule::Selbri).unwrap();
assert_eq!(selbri.words().next().unwrap().text, "klama");

assert_eq!(classify("lobypli"), Ok(WordClass::Lujvo));

let script = to_zbalermorna("coi rodo")?;
# Ok::<(), Box<dyn std::error::Error>>(())

Try the tree printer:

$ cargo run -p lojban-rs --example parse -- "mi klama le zarci"

Parse failures render as source-annotated reports via ariadne.

Sources and fidelity

The language data comes from two sources:

• The Complete Lojban Language, 1.1 - the EBNF grammar of chapter 21, the morphology rules of chapters 3-4, and the dictionary shipped with the book's sources.
• The zbalermorna repository - codepoint assignments are recorded from the fonts themselves, which are the ground truth for the script.

The language data lives in each crate's data/ directory as checked-in TSV files, so builds do not need local source checkouts or extraction scripts.

The grammar follows the book's LALR-oriented EBNF. PEG-specific adaptations are documented in lojban-grammar/src/parser.rs. This is not the community camxes PEG grammar.

Minimum supported Rust version

1.81.

License

The code is licensed under either of

• Apache License, Version 2.0 (LICENSE-APACHE)
• MIT license (LICENSE-MIT)

at your option.

The embedded language data (data/*.tsv in the crates) is derived from The Complete Lojban Language, (c) the Logical Language Group, licensed CC-BY-SA 4.0. The zbalermorna sources are public domain.

Unless you explicitly state otherwise, any contribution intentionally submitted for inclusion in the work by you, as defined in the Apache-2.0 license, shall be dual licensed as above, without any additional terms or conditions.