Question

鉴于URL：

var urlString = "http://somehost:9090/cars;color=red;make=Tesla?page=1&perPage=10"

我想要一些javascript（节点）库，我可以使用它来获取汽车路径段的矩阵参数（颜色和品牌），例如：

var url = URL.parse(urlString)
url.pathSegments["cars"].params

会产生

{
  "color": "red",
  "make": "Tesla"
}

另外，理想情况下，这样的库应该考虑路径段参数的正确解码，这与查询参数的解码不同。

以下文章详细介绍了这些参数（以及一些有关网址的其他有用信息）：

https://www.talisman.org/~erlkonig/misc/lunatech%5Ewhat-every-webdev-must-know-about-url-encoding/

我做了大量的谷歌搜索，但是空洞的，但希望我只是失明！

Answer 1

我找到了URI.js。但是，如果您不想使用该库，我认为此功能可以满足您的需求（对decodeURIComponent不太确定）：

var urlString = "http://somehost:9090/cars;color=red;make=Tesla?page=1&perPage=10"

var getParams = function (urlString) {
    return decodeURIComponent(urlString) // decode the URL (?)
        .match(/\/((?!.+\/).+)\?/)
            // the regex looks for a slash that is NOT
            // followed by at least one character and eventually another slash
            // given var urlString = "http://somehost:9090/cars;color=red;make=Tesla?page=1&perPage=10"
            // we don't want        -------^              ^                         ^
            // we want this slash                   ------|                         |
            // all the way until this question mark --------------------------------|
            // regex explanation:
            /*
                \/                  first slash
                (                   open capturing group
                    (?!             lookbehind for NOT
                        .+\/        any character followed by a slash (/)
                    )
                    .+              capture one or more characters (greedy) past
                )                   the close of the capturing group and until
                \?                  a question mark
            */
        [1] // match will return two groups, which will look like:
            // ["/cars;color=red;make=Tesla?", "cars;color=red;make=Tesla"]
            // we want the second one (otherwise we'd have to .slice(1,-1) the string)
        .split(";")
            // split it at the semicolons
            // if you know you're always going to have "name" followed by a semicolon,
            // you might consider using .slice(1) on this part, so you can get rid of 
            // the if statement below (still keep the p[c[0]] = c[1] part though )
        .reduce(function (p, c) {
            // split it at the equals sign for a key/value in indices 0 and 1
            c = c.split("=");
            // if the length is greater than one, aka we have a key AND a value
            // e.g., c == ["color", "red"]
            if (c.length > 1) {
                // give the previous object a key of c[0] equal to c[1]
                // i.e., p["color"] = "red"
                p[c[0]] = c[1];
            }
            return p; // return p, so that we can keep adding keys to the object
        }, {}); // we pass an object, which will act as p on the first call
}

console.log(getParams(urlString));  // { color: "red", make: "Tesla" }

除了正则表达式，您还可以使用我在上面的评论中发布的内容：

urlString.split("?")[0].split("/").pop().split(";").reduce( /* etc */)

现在我想要特斯拉......

Answer 2

我最近写了一个Node.js中间件来解析矩阵参数。我已经指定了它遵循的规则以及它生成的输出格式。

例如，这就是你的app.js的样子：

let app = require ('express') (),
  matrixParser = require ('matrix-parser');

app.use (matrixParser ());

app.get ('/cars*', (req, res) => {
  //notice the asterisk after '/cars'
  console.log (JSON.stringify (req.matrix, null, 2));
  res.send ('Thanks=)');
});

app.listen (9090);

，你的URI看起来像：

http://localhost:9090/cars;color=red;make=Tesla?page=1&perPage=10

然后，您可以使用curl测试矩阵解析器功能：

curl "http://localhost:9090/cars;color=red;make=Tesla?page=1&perPage=10"

然后将req.matrix设置为以下对象：

[
  {
    "segment": "cars",
    "matrix": {
      "color": "red",
      "make": "Tesla"
    }
  }
]

查询字符串（page，per_page）保持不变（您可以通过简单地写req.query来看到这一点）

此时写答案可能为时已晚，但未来可能仍然会有所帮助。

这是回购： https://github.com/duaraghav8/matrix-parser

npm install matrix-parser

编辑：很抱歉没有提前用代码提供更精细的答案，这是我对SO的第一个贡献，我需要一些时间才能掌握它。

是否存在用于解析URL路径段（矩阵）参数的javascript库？

2 个答案: