javascript - 在 JavaScript 中自动释义句子

Question

在 JavaScript 中，是否可以使用随机生成的正则表达式匹配自动替换句子中的正则表达式？我正在尝试使用这种方法使用正则表达式列表自动解释一个句子，如下所示:

replaceWithRandomFromRegexes("你不是疯子!你是神童!", ["(genius|prodigy)", "(freak|loony|crackpot|crank|crazy)", "(You 're |You are |Thou art )", "(arent|ain't|are not)"])

此处，输入字符串中每个正则表达式的每个匹配项都应替换为随机生成的正则表达式匹配项。

function replaceWithRandomFromRegexes(theString, theRegexes){
    //For each regex in theRegexes, replace the first match of the regex in the string with a randomly generated match of that regex.
}

Answer 1

这似乎比你想象的要简单得多。怎么样:

function randomReplace(subject, groups, wordsOnly) {
    var meta = /([.?*+^$[\]\\(){}|-])/g, all = {};
    groups.forEach(function(group) {
        group.forEach(function(word) { all[word] = group })
    });
    var r = Object.keys(all).
        sort(function(x, y) { return y.length - x.length }).
        map(function(x) { return x.replace(meta, "\\$&") }).
        join("|");
    if(wordsOnly)
        r = "\\b(" + r + ")\\b";
    return subject.replace(new RegExp(r, "g"), function($0) {
        return all[$0][Math.floor(Math.random() * all[$0].length)]
    });
}

例子:

s = randomReplace(
    "You aren't a crackpot! You're a prodigy!",
    [
        ["genius", "prodigy"], 
        ["freak", "loony", "crackpot", "crank", "crazy"], 
        ["You're ", "You are ", "Thou art "], 
        ["aren't", "ain't", "are not"]
    ]
);
console.log(s) // You ain't a crank! Thou art a genius!

如评论中讨论的那样，扩展功能可能是这样的:

function expand(s) {
    var d = [];

    function product(a, b) {
        var p = [];
        a.map(function(x) { b.map(function(y) { p.push(x + y) })});
        return p;
    }

    function reduce(s) {
        var m;
        if(s.indexOf("|") >= 0)
            return [].concat.apply([], s.split("|").map(reduce));
        if(m = s.match(/~(\d+)(.*)/))
            return product(reduce(d[m[1]]), reduce(m[2]));
        return [s];
    }

    function add($0, $1) { d.push($1); return '~' + (d.length - 1) }

    s = s.replace(/([^()|]+)/g, add);
    for(var r = /\(([^()]*)\)/g; s.match(r);)
        s = s.replace(r, add);

    return reduce(s);
}

例子:

z = "(He|She|It|(B|R)ob(by|)) (real|tru|sure)ly is"
console.log(expand(z)) 

结果:

[
 "He really is",
 "He truly is",
 "He surely is",
 "She really is",
 "She truly is",
 "She surely is",
 "It really is",
 "It truly is",
 "It surely is",
 "Bobby really is",
 "Bobby truly is",
 "Bobby surely is",
 "Bob really is",
 "Bob truly is",
 "Bob surely is",
 "Robby really is",
 "Robby truly is",
 "Robby surely is",
 "Rob really is",
 "Rob truly is",
 "Rob surely is"
]