根据另一个数组中的条目拆分数组

Question

我正在研究一个JavaScript代码，该代码将RegExp分解为其基本组件，并对其做了什么小小的说明。

我的一般想法是将输入字符串（作为RegExp）拆分为另一个数组的条目。

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 $1 time","repeated $1 or more times","repeated between $1 and $2 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. 
} 

理想的情况下，该功能（允许调用它splitR）将返回输出是这样的：

> 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"] 

明确的规定，在splitR函数应该在上下文interpret函数，获取RegExp并将其分解到其基本组件;例如\d+[0-9]\w*?应拆分为["\d", "+", "[", "0-9", "]", "\w", "*", "?"]。这些组件使用各种RegExps（例如/{(\d+)}/g查找a{n}）和字符串（例如"."）在其他阵列中单独定义。

真的，我很难定义为splitR。任何帮助表示赞赏！

Answer 1

这将部分拆分正则表达式，并用部分描述填充第二个数组。它会跳过意想不到的字符，但是没有真正的正则表达式语法检查，即如果您启动了一个范围并且没有结束它，那么脚本就不会发出抱怨。我加入了一些事情从你的列表中缺少的，像分组括号的自由，开始和结束锚...

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 $1 time","repeated $1 or more times","repeated between $1 and $2 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); \t // 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);