根据另一个数组中的条目拆分数组

Question

我正在编写一个 JavaScript 代码，它将 RegExp 分解成它的基本组件，并对它的作用给出一个简短的解释。

我的总体想法是将输入字符串（作为 RegExp）拆分为另一个数组中的条目。

我当前的代码：

function interpret(regex){
    var r = regex + "";
    r = r.split("/");
    body = r[1];
    flags = r[2];
    var classes     = [".","\w","\d","\s","\W","\D","\S","[","]"];
    var classdefs   = ["any non-newline character","any word (digit or letter)","any digit (characters 0-9)","any whitespace character","any non-word (non-digit and non-letter)","any non-digit (not characters 0-9)","open matchset","close matchset"];
    var quantifiers = ["*","+","?",
                        /{(\d+)}/g,         // a{n} 
                        /{(\d+),}/g,        // a{n,}
                        /{(\d+),(\d+)}/g,   // a{n,m}
                        /[+*?]\?/g          // a<quant>? - lazy quantification
                      ];
    var quantDefs   = ["repeated 0 or more times","repeated 1 or more times","repeated once or not at all","repeated exactly  time","repeated  or more times","repeated between  and  times"];
    var escaped     = ["\t","\n","\r","\.","\*","\","\^","\?","\|"];
    var escapedDefs = ["a tab","a linefeed","a carriage return","a period","an asterisk","a backslash","a carot","a question mark","a vertical bar"];

    // code to split r based on entries in classes, quantifiers, and escaped.
}

理想情况下，这个函数（我们称之为 splitR）将 return 输出如下内容：

> splitR("hello",["he","l"]);
["he", "l", "l", "o"]
> splitR("hello",["he"]);
["he", "llo"]
> splitR("hello",["he","o"]);
["he", "ll", "o"];
> splitR("5 is the square root of 25",[/\d+/g,/\w{3,}/g,"of"]);
["5", " is ", "the", " ", "square", " ", "root", " ", "of", " ", "25"]

明确定义，splitR 函数应该在 interpret 函数的上下文中采用 RegExp 并将其拆分为基本组件；例如\d+[0-9]\w*? 应该拆分成 ["\d", "+", "[", "0-9", "]", "\w", "*", "?"]。这些组件在其他数组中单独定义，使用各种正则表达式（例如 /{(\d+)}/g 来查找 a{n}）和字符串（例如 "."）。

真的，我对splitR的定义感到困惑。感谢您的帮助！

Answer 1

这会将正则表达式拆分成多个部分，并用各部分的描述填充第二个数组。它会跳过意想不到的字符，但没有真正的正则表达式语法检查，即如果你开始一个范围而不结束它，脚本不会抱怨。我冒昧地添加了一些你的列表中遗漏的东西，比如分组括号、开始和结束锚点……

function interpret(regex)
{
    var body = regex.source;
    var flags = (regex.global ? "g" : "") + (regex.ignoreCase ? "i" : "") + (regex.multiline ? "m" : "");
    var classes     = [/^\w\-\w/,/^\./,/^\w/,/^\d/,/^\s/,/^\W/,/^\D/,/^\S/,/^\[/,/^\]/,/^\(/,/^\)/,/^\^/,/^$/,/^\|/];
    var classDefs   = ["character range","any non-newline character","any word (digit or letter)","any digit (characters 0-9)","any whitespace character","any non-word (non-digit and non-letter)","any non-digit (not characters 0-9)","any non-whitespace character","open matchset","close matchset","open group","close group","start anchor or negation","end anchor","alternative"];
    var quantifiers = [/^[+*?]\?/,/^\*/,/^\+/,/^\?/,/^{(\d+)}/,/^{(\d+),}/,/^{(\d+),(\d+)}/];
    var quantDefs   = ["lazy quantification","repeated 0 or more times","repeated 1 or more times","repeated once or not at all","repeated exactly  time","repeated  or more times","repeated between  and  times"];
    var escaped     = [/^\t/,/^\n/,/^\r/,/^\\./,/^\\*/,/^\\+/,/^\\-/,/^\\/,/^\\^/,/^\$/,/^\\?/,/^\\|/,/^\\[/,/^\\]/,/^\\(/,/^\\)/,/^\\{/,/^\\}/];
    var escapedDefs = ["a tab","a linefeed","a carriage return","a period","an asterisk","a plus","a minus","a backslash","a caret","a dollar sign","a question mark","a vertical bar","a square bracket","a square bracket","a bracket","a bracket","a curly bracket","a curly bracket"];
    var literal     = [/^[^\.\\[\]\(\)\^$\|\*\+\-\?\{\}]+/];
    var literalDefs = ["literal text"];
    var regs = classes.concat(quantifiers,escaped,literal);
    var defs = classDefs.concat(quantDefs,escapedDefs,literalDefs);
    var reglist = [];
    var deflist = [];

    while (body.length)
    {
        var found = false;
        var chunk = null;

        for (var i = 0; i < regs.length; i++)
        {
            chunk = body.match(regs[i]);

            if (chunk)
            {
                reglist.push(chunk[0]);
                deflist.push(defs[i]);
                body = body.substr(chunk[0].length);
                found = true;
                break;
            }
        }

        if (!found)
        {
            body = body.substr(1); // skip unexpected character
        }
    }

    console.log(regex.source);
    console.log(reglist);
    console.log(deflist);
    alert("see console for output");
}

var x = new RegExp("^[a-z0-9]\^.\.\w\d\s\W\D\S+(te|\|st)*\*\n+\+\-\}(\W?\?\s{1,3})\\*?a{3}b{4,}c{}\r\t\$$", "ig");
interpret(x);

根据另一个数组中的条目拆分数组

Splitting an array based on entries in another array

javascript

regex

arrays

split