在 ANTLR 中，是否有一种快捷符号来表示某些规则集的所有排列的交替? [英] In ANTLR, is there a shortcut notation for expressing alternation of all the permutations of some set of rules?

问题描述

在 ANTLR 中，我想定义这样的规则:

In ANTLR I want to define a rule like this:

规则:( a b c | a c b | b a c | b c a | c a b | c b a );

rule : ( a b c | a c b | b a c | b c a | c a b | c b a );

但在我的情况下，我有 10 条规则而不是 3 条，我想对这些规则进行置换，因此它变得非常不切实际.有没有什么方法可以在 ANTLR 中表达这一点而不必编写所有排列?

But in my case I have 10 rules instead of three, that I want to permute so it gets very impractical. Is there any way of expressing this in ANTLR without having to write all the permutations?

推荐答案

我只会匹配任何 a、b 或 c 一次或更多:

I would just match any a, b or c once or more:

rule
 : ( a | b | c )+
 ;

然后，在解析之后，遍历解析树并检查a、b和c是否都完全匹配一次.

and then, after parsing, traversing the parse tree and checking if a, b and c all matched exactly once.

但是是的，在语法本身中可以通过在需要的地方使用谓词来实现.

But Yes, it is possible in the grammar itself by using predicates where needed.

演示:

grammar Permutation;

parse
  :  permutation[5] {System.out.println("parsed: " + $permutation.text);} EOF
  ;

permutation[final int n]
@init{
  java.util.Set set = new java.util.HashSet();
  int counter = n;
}
  :  (
       {counter > 0}?=> token   // keep matching a `token` as long as `counter > 0`
       {                        //
         set.add($token.text);  // add the contents of `token` to `set`
         counter--;             // decrease `counter`
       }                        //
     )+ 
     {set.size() == n}?         // if `set.size() != n`, an exception is thrown
  ;

token
  :  A
  |  B
  |  C
  |  D
  |  E
  ;

A : 'A';
B : 'B';
C : 'C';
D : 'D';
E : 'E';

Space : ' ' {skip();};

上面的演示语法使用了两种不同类型的谓词:1)一个门控语义谓词ⁱ以确保permutation 规则匹配不超过参数 final int n 标记，以及 2) 一个 验证语义谓词 ⁱ 以确保 set 恰好包含 final int n 元素，以确保它是 5 个标记的正确排列.

The demo grammar above uses 2 different types of predicates: 1) a gated semantic predicate ⁱ to make sure that the permutation rule matches no more than the parameter final int n tokens, and 2) a validating semantic predicate ⁱ to ensure that the set holds exactly the final int n elements to ensure that it's a proper permutation of the 5 tokens.

可以在此处找到有关语义谓词的更多信息:什么是语义"ANTLR 中的谓词'?

More info about semantic predicates can be found here: What is a 'semantic predicate' in ANTLR?

您可以使用以下课程测试语法:

You can test the grammar with the following class:

import org.antlr.runtime.*;

public class Main {
  public static void main(String[] args) throws Exception {
    PermutationLexer lexer = new PermutationLexer(new ANTLRStringStream(args[0]));
    PermutationParser parser = new PermutationParser(new CommonTokenStream(lexer));
    parser.parse();
  }
}

java -cp antlr-3.3.jar org.antlr.Tool Permutation.g 
javac -cp antlr-3.3.jar *.java

java -cp .:antlr-3.3.jar Main "A B C D E"
parsed: ABCDE

java -cp .:antlr-3.3.jar Main "B D C E A"
parsed: BDCEA

java -cp .:antlr-3.3.jar Main "A B C D B"
line 1:9 rule permutation failed predicate: {set.size() == n}?
parsed: null

这篇关于在 ANTLR 中，是否有一种快捷符号来表示某些规则集的所有排列的交替?的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！