bnf

A library for parsing Backus–Naur form context-free grammars.

Installation

Add to your Cargo.toml:

[dependencies]
bnf = "0.6"

What does a parsable BNF grammar look like?

The following grammar from the Wikipedia page on Backus-Naur form exemplifies a compatible grammar. (*Note: parser allows for an optional ';' to indicate the end of a production)

 <postal-address> ::= <name-part> <street-address> <zip-part>

      <name-part> ::= <personal-part> <last-name> <opt-suffix-part> <EOL>
                    | <personal-part> <name-part>

  <personal-part> ::= <initial> "." | <first-name>

 <street-address> ::= <house-num> <street-name> <opt-apt-num> <EOL>

       <zip-part> ::= <town-name> "," <state-code> <ZIP-code> <EOL>

<opt-suffix-part> ::= "Sr." | "Jr." | <roman-numeral> | ""
    <opt-apt-num> ::= <apt-num> | ""

Extended syntax (groups and optionals)

When parsing grammar text (e.g. [str::parse] or [Grammar::parse_from]), the parser accepts two shortcuts:

Parenthesized groups ( ... )

Group alternatives so they act as one unit in a sequence.

Without parentheses, | binds loosely. This rule:

<s> ::= <a> | <b> <c>

means "<a> or <b> <c>". So "a" matches, and "b c" matches, but "a c" does not.

With parentheses, you get "(a or b) then c":

<s> ::= (<a> | <b>) <c>

So only "a c" and "b c" match.

Optionals [ ... ]

Zero or one of the grouped alternatives (like ? in regex).

<word> ::= <letter> [<digit>]

means: a letter, optionally followed by a digit. Both "x" and "x1" match.

Equivalent long form without extended syntax:

<word>      ::= <letter> <opt-digit>
<opt-digit> ::= <digit> | ""

Normalization

Groups and optionals are normalized into a grammar that uses only plain nonterminals and terminals: each group or optional is turned into a fresh internal nonterminal (e.g. __anon0, __anon1). The public [Term] type has only [Term::Terminal] and [Term::Nonterminal]; parsing and generation use this normalized form.

Round-trip: Formatting a grammar (e.g. format!("{}", grammar)) does not preserve ( ) or [ ]; the result uses __anon* names. Re-parsing yields an equivalent grammar.

Empty groups or optionals — () or [] with nothing inside — are invalid; at least one alternative is required.

Output

Take the following grammar for DNA sequences to be input to this library's parse function.

<dna> ::= <base> | <base> <dna>
<base> ::= "A" | "C" | "G" | "T"

The output is a Grammar object representing a tree that looks like this:

Grammar
├── <dna> ::=
│   ├── <base>
│   └── <base> <dna>
└── <base> ::=
    ├── "A"
    ├── "C"
    ├── "G"
    └── "T"

Once the Grammar object is populated, to generate a random sentence from it call the object's generate function. grammar.generate(). For the above grammar you could expect something like TGGC or AG.

If the generate function can't find a production for a nonterminal it tries to evaluate it will print the identifer as a nonterminal, i.e. <identifier>.

The generate function will return an error if it detects an infinite loop caused by a production such as <PATTERN> ::= <PATTERN>.

Parse Example

use bnf::Grammar;

let input =
    "<postal-address> ::= <name-part> <street-address> <zip-part>

        <name-part> ::= <personal-part> <last-name> <opt-suffix-part> <EOL>
                        | <personal-part> <name-part>

    <personal-part> ::= <initial> '.' | <first-name>

    <street-address> ::= <house-num> <street-name> <opt-apt-num> <EOL>

        <zip-part> ::= <town-name> ',' <state-code> <ZIP-code> <EOL>

    <opt-suffix-part> ::= 'Sr.' | 'Jr.' | <roman-numeral> | ''
        <opt-apt-num> ::= <apt-num> | ''";

let grammar: Result<Grammar, _> = input.parse();
match grammar {
    Ok(g) => println!("{:#?}", g),
    Err(e) => println!("Failed to make grammar from String: {}", e),
}

Generate Example

use bnf::Grammar;

let input =
    "<dna> ::= <base> | <base> <dna>
    <base> ::= 'A' | 'C' | 'G' | 'T'";
let grammar: Grammar = input.parse().unwrap();
let sentence = grammar.generate();
match sentence {
    Ok(s) => println!("random sentence: {}", s),
    Err(e) => println!("something went wrong: {}!", e)
}

Parse Sentence via Grammar

use bnf::Grammar;

let input =
    "<dna> ::= <base> | <base> <dna>
    <base> ::= 'A' | 'C' | 'G' | 'T'";
let grammar: Grammar = input.parse().unwrap();

// Create a parser from the grammar (validates all nonterminals are defined)
let parser = grammar.build_parser().unwrap();

let sentence = "GATTACA";

let mut parse_trees = parser.parse_input(sentence);
match parse_trees.next() {
    Some(parse_tree) => println!("{}", parse_tree),
    _ => println!("Grammar could not parse sentence"),
}

By default, parse_input implicitly starts from the first rule. To match another rule, parse_input_starting_with can be used:

use bnf::{Grammar, Term};

let input =
    "<dna> ::= <base> | <base> <dna>
    <base> ::= 'A' | 'C' | 'G' | 'T'";
let grammar: Grammar = input.parse().unwrap();

// Create a parser from the grammar (validates all nonterminals are defined)
let parser = grammar.build_parser().unwrap();

let sentence = "G";
let target_production = Term::Nonterminal("base".to_string());

let mut parse_trees = parser.parse_input_starting_with(sentence, &target_production);
match parse_trees.next() {
    Some(parse_tree) => println!("{}", parse_tree),
    _ => println!("Grammar could not parse sentence"),
}