b

Grako takes a grammar in a variation of EBNF as input, and outputs a memoizing PEG/Packrat parser in Python.

EBNF，是一个计算机用语，意思是扩展的巴科斯范式。

扩展的 巴科斯范式

用来描述计算机语言语法的符号集。现在，几乎每一位新编程语言书籍的作者都使用巴科斯范式来定义编程语言的语法规则。

扩展巴科斯-瑙尔范式(EBNF)是表达作为描述 计算机编程语言和形式语言的正规方式的上下文无关文法的元语法符号表示法。它是基本巴科斯范式(BNF)元语法符号表示法的一种扩展。

它最初由尼古拉斯·沃斯开发，最常用的 EBNF 变体由标准，特别是 ISO-14977 所定义。

In computer science, a parsing expression grammar, or PEG, is a type of analytic formal grammar, i.e. it describes a formal language in terms of a set of rules for recognizing strings in the language. The formalism was introduced by Bryan Ford in 2004^[1] and is closely related to the family of top-down parsing languages introduced in the early 1970s. Syntactically, PEGs also look similar to context-free grammars (CFGs), but they have a different interpretation: the choice operator selects the first match in PEG, while it is ambiguous in CFG. This is closer to how string recognition tends to be done in practice, e.g. by a recursive descent parser.

Unlike CFGs, PEGs cannot be ambiguous; if a string parses, it has exactly one valid parse tree. It is conjectured that there exist context-free languages that cannot be recognized by a PEG, but this is not yet proven.^[1] PEGs are well-suited to parsing computer languages (and artificial human languages such as Lojban), but not natural languages where the performance of PEG algorithms is comparable to general CFG algorithms such as the Earley algorithm

#!/usr/bin/env python
import json
import grako # $ pip install grako

grammar_ebnf = """
    bracketed = '{' @:( { bracketed }+ | any ) '}' ;
    any = /[^{}]+?/ ;
"""
model = grako.genmodel("Bracketed", grammar_ebnf)
ast = model.parse("{ { a } { b } { { { c } } } }", "bracketed")
print(json.dumps(ast, indent=4))

Output

[
    "a", 
    "b", 
    [
        [
            "c"
        ]
    ]
]