是否有一种通过jQuery(或不使用)检索查询字符串值的无插件方法?
如果是,怎么办?如果没有,是否有插件可以这样做?
是否有一种通过jQuery(或不使用)检索查询字符串值的无插件方法?
如果是,怎么办?如果没有,是否有插件可以这样做?
当前回答
这是Andy E链接的“句柄数组样式查询字符串”版本的扩展版本。修复了一个错误(?key=1&key[]=2&key[]=3;1丢失并替换为[2,3]),进行了一些小的性能改进(重新解码值,重新计算“[”位置等),并添加了一些改进(功能化,支持?key=1&key=2,支持;分隔符)。我将变量留得很短,但添加了大量注释以使其可读(哦,我在本地函数中重用了v,如果这令人困惑,很抱歉;)。
它将处理以下查询字符串。。。
?test=Hello&pers=neek&pers[]=jeff&pers[][]=jim&pers[extra]=john&test3&nocache=13989148914891264
…把它做成一个看起来像。。。
{
"test": "Hello",
"person": {
"0": "neek",
"1": "jeff",
"2": "jim",
"length": 3,
"extra": "john"
},
"test3": "",
"nocache": "1398914891264"
}
如上所述,此版本处理一些“格式错误”数组,即-person=neek&person[]=jeff&person[]=jim或person=neek/person=jeff/person=jim,因为密钥是可识别的和有效的(至少在dotNet的NameValueCollection.Add中):
如果目标NameValueCollection中已存在指定的键例如,指定的值将添加到现有的逗号分隔的格式为“value1,value2,value3”的值列表。
似乎陪审团对重复的键有点不满意,因为没有规范。在这种情况下,多个键被存储为一个(假)数组。但请注意,我不会将基于逗号的值处理为数组。
代码:
getQueryStringKey = function(key) {
return getQueryStringAsObject()[key];
};
getQueryStringAsObject = function() {
var b, cv, e, k, ma, sk, v, r = {},
d = function (v) { return decodeURIComponent(v).replace(/\+/g, " "); }, //# d(ecode) the v(alue)
q = window.location.search.substring(1), //# suggested: q = decodeURIComponent(window.location.search.substring(1)),
s = /([^&;=]+)=?([^&;]*)/g //# original regex that does not allow for ; as a delimiter: /([^&=]+)=?([^&]*)/g
;
//# ma(make array) out of the v(alue)
ma = function(v) {
//# If the passed v(alue) hasn't been setup as an object
if (typeof v != "object") {
//# Grab the cv(current value) then setup the v(alue) as an object
cv = v;
v = {};
v.length = 0;
//# If there was a cv(current value), .push it into the new v(alue)'s array
//# NOTE: This may or may not be 100% logical to do... but it's better than loosing the original value
if (cv) { Array.prototype.push.call(v, cv); }
}
return v;
};
//# While we still have key-value e(ntries) from the q(uerystring) via the s(earch regex)...
while (e = s.exec(q)) { //# while((e = s.exec(q)) !== null) {
//# Collect the open b(racket) location (if any) then set the d(ecoded) v(alue) from the above split key-value e(ntry)
b = e[1].indexOf("[");
v = d(e[2]);
//# As long as this is NOT a hash[]-style key-value e(ntry)
if (b < 0) { //# b == "-1"
//# d(ecode) the simple k(ey)
k = d(e[1]);
//# If the k(ey) already exists
if (r[k]) {
//# ma(make array) out of the k(ey) then .push the v(alue) into the k(ey)'s array in the r(eturn value)
r[k] = ma(r[k]);
Array.prototype.push.call(r[k], v);
}
//# Else this is a new k(ey), so just add the k(ey)/v(alue) into the r(eturn value)
else {
r[k] = v;
}
}
//# Else we've got ourselves a hash[]-style key-value e(ntry)
else {
//# Collect the d(ecoded) k(ey) and the d(ecoded) sk(sub-key) based on the b(racket) locations
k = d(e[1].slice(0, b));
sk = d(e[1].slice(b + 1, e[1].indexOf("]", b)));
//# ma(make array) out of the k(ey)
r[k] = ma(r[k]);
//# If we have a sk(sub-key), plug the v(alue) into it
if (sk) { r[k][sk] = v; }
//# Else .push the v(alue) into the k(ey)'s array
else { Array.prototype.push.call(r[k], v); }
}
}
//# Return the r(eturn value)
return r;
};
其他回答
这是Andy E链接的“句柄数组样式查询字符串”版本的扩展版本。修复了一个错误(?key=1&key[]=2&key[]=3;1丢失并替换为[2,3]),进行了一些小的性能改进(重新解码值,重新计算“[”位置等),并添加了一些改进(功能化,支持?key=1&key=2,支持;分隔符)。我将变量留得很短,但添加了大量注释以使其可读(哦,我在本地函数中重用了v,如果这令人困惑,很抱歉;)。
它将处理以下查询字符串。。。
?test=Hello&pers=neek&pers[]=jeff&pers[][]=jim&pers[extra]=john&test3&nocache=13989148914891264
…把它做成一个看起来像。。。
{
"test": "Hello",
"person": {
"0": "neek",
"1": "jeff",
"2": "jim",
"length": 3,
"extra": "john"
},
"test3": "",
"nocache": "1398914891264"
}
如上所述,此版本处理一些“格式错误”数组,即-person=neek&person[]=jeff&person[]=jim或person=neek/person=jeff/person=jim,因为密钥是可识别的和有效的(至少在dotNet的NameValueCollection.Add中):
如果目标NameValueCollection中已存在指定的键例如,指定的值将添加到现有的逗号分隔的格式为“value1,value2,value3”的值列表。
似乎陪审团对重复的键有点不满意,因为没有规范。在这种情况下,多个键被存储为一个(假)数组。但请注意,我不会将基于逗号的值处理为数组。
代码:
getQueryStringKey = function(key) {
return getQueryStringAsObject()[key];
};
getQueryStringAsObject = function() {
var b, cv, e, k, ma, sk, v, r = {},
d = function (v) { return decodeURIComponent(v).replace(/\+/g, " "); }, //# d(ecode) the v(alue)
q = window.location.search.substring(1), //# suggested: q = decodeURIComponent(window.location.search.substring(1)),
s = /([^&;=]+)=?([^&;]*)/g //# original regex that does not allow for ; as a delimiter: /([^&=]+)=?([^&]*)/g
;
//# ma(make array) out of the v(alue)
ma = function(v) {
//# If the passed v(alue) hasn't been setup as an object
if (typeof v != "object") {
//# Grab the cv(current value) then setup the v(alue) as an object
cv = v;
v = {};
v.length = 0;
//# If there was a cv(current value), .push it into the new v(alue)'s array
//# NOTE: This may or may not be 100% logical to do... but it's better than loosing the original value
if (cv) { Array.prototype.push.call(v, cv); }
}
return v;
};
//# While we still have key-value e(ntries) from the q(uerystring) via the s(earch regex)...
while (e = s.exec(q)) { //# while((e = s.exec(q)) !== null) {
//# Collect the open b(racket) location (if any) then set the d(ecoded) v(alue) from the above split key-value e(ntry)
b = e[1].indexOf("[");
v = d(e[2]);
//# As long as this is NOT a hash[]-style key-value e(ntry)
if (b < 0) { //# b == "-1"
//# d(ecode) the simple k(ey)
k = d(e[1]);
//# If the k(ey) already exists
if (r[k]) {
//# ma(make array) out of the k(ey) then .push the v(alue) into the k(ey)'s array in the r(eturn value)
r[k] = ma(r[k]);
Array.prototype.push.call(r[k], v);
}
//# Else this is a new k(ey), so just add the k(ey)/v(alue) into the r(eturn value)
else {
r[k] = v;
}
}
//# Else we've got ourselves a hash[]-style key-value e(ntry)
else {
//# Collect the d(ecoded) k(ey) and the d(ecoded) sk(sub-key) based on the b(racket) locations
k = d(e[1].slice(0, b));
sk = d(e[1].slice(b + 1, e[1].indexOf("]", b)));
//# ma(make array) out of the k(ey)
r[k] = ma(r[k]);
//# If we have a sk(sub-key), plug the v(alue) into it
if (sk) { r[k][sk] = v; }
//# Else .push the v(alue) into the k(ey)'s array
else { Array.prototype.push.call(r[k], v); }
}
}
//# Return the r(eturn value)
return r;
};
我喜欢Ryan Phelan的解决方案。但我看不出扩展jQuery有什么意义?没有使用jQuery功能。
另一方面,我喜欢Google Chrome中的内置函数:window.location.getParameter。
那么为什么不使用这个呢?好吧,其他浏览器没有。因此,如果不存在,让我们创建此函数:
if (!window.location.getParameter ) {
window.location.getParameter = function(key) {
function parseParams() {
var params = {},
e,
a = /\+/g, // Regex for replacing addition symbol with a space
r = /([^&=]+)=?([^&]*)/g,
d = function (s) { return decodeURIComponent(s.replace(a, " ")); },
q = window.location.search.substring(1);
while (e = r.exec(q))
params[d(e[1])] = d(e[2]);
return params;
}
if (!this.queryStringParams)
this.queryStringParams = parseParams();
return this.queryStringParams[key];
};
}
该函数或多或少与Ryan Phelan有所不同,但包装方式不同:名称清晰,不依赖其他javascript库。有关此功能的更多信息,请访问我的博客。
只是另一个建议。插件Purl允许检索URL的所有部分,包括锚点、主机等。
它可以与jQuery一起使用,也可以不使用jQuery。
用法很简单,很酷:
var url = $.url('http://example.com/folder/dir/index.html?item=value'); // jQuery version
var url = purl('http://example.com/folder/dir/index.html?item=value'); // plain JS version
url.attr('protocol'); // returns 'http'
url.attr('path'); // returns '/folder/dir/index.html'
然而,截至2014年11月11日,Purl不再被维护,作者建议改用URI.js。jQuery插件的不同之处在于它关注元素-对于字符串使用,只需直接使用URI,无论是否使用jQuery。类似的代码看起来是这样的,更完整的文档如下:
var url = new URI('http://example.com/folder/dir/index.html?item=value'); // plain JS version
url.protocol(); // returns 'http'
url.path(); // returns '/folder/dir/index.html'
获取查询的一行代码:
var value = location.search.match(new RegExp(key + "=(.*?)($|\&)", "i"))[1];
我使用这里列出的技术开发了一个小型库,以创建一个易于使用的解决方案;可以在这里找到:
https://github.com/Nijikokun/query-js
用法
正在获取特定参数/密钥:
query.get('param');
使用生成器获取整个对象:
var storage = query.build();
console.log(storage.param);
还有很多。。。查看github链接以获取更多示例。
特征
解码和参数缓存支持哈希查询字符串#hello?第3页支持传递自定义查询支持数组/对象参数user[]=“jim”&user[]=”bob“支持空管理&&支持不带值的声明参数name&hello=“world”支持重复参数param=1¶m=2干净、紧凑、可读的源4kbAMD,需要,节点支持