javascript查找协议，域，加上来自src标记的regexp的第一个斜杠，替换为空字符串

Question

I tried to construct a regex for this task but I'm afraid I am still failing to have an intuitive understanding of regexp.

The problem is the regex matches until the last slash in a string. I want it to stop at the first match of the string.

My pathetic attempt at regex:

/^http(s?):\/\/.+\/{1}/

Test subject:

http://foo.com/bar/test/foo.jpeg

The goal is to obtain bar/test/foo.jpeg , so that I may then split the string, pop the last element and then join the remainder, resulting in having the path to the JavaScript file.

Example

var str = 'http://foo.com/bar/test/foo.jpeg';
str.replace(regexp,'');

Answer 1

While the other answer shows how to match a part of a string, I think a replace solution is more appropriate for the current task.

The issue you have is that .+ matches one or more characters other than a newline greedily , that is, all the string is grabbed first in one go, and then the regex engine starts backtracking (moving backwards along the input string looking for a / to accommodate in the match). Thus, you get the match from http until the last / .

To restrict the match from http to the first / use a negated character class [^/]+ instead of .+ .

^https?:\/\/[^\/]+\/
            ^^^^^^

See the regex demo

Note that you do not need to place s into a capturing group to make it optional, unescaped ? is a quantifier that makes the preceding character match one or zero times. Also, {1} is a redundant quantifier since this is default behavior, c will only match 1 c , (?:something) will only match one something .

 var re = /^https?:\\/\\/[^\\/]+\\//; var str = 'http://foo.com/bar/test/foo.jpeg'; var result = str.replace(re, ''); document.getElementById("r").innerHTML = result; 

 <div id="r"/> 

Note that you will need to assign the replace result to some variable, since in JS, strings are immutable.

Regex explanation :

• ^ - start of string
• https? - either http or https substring
• :\\/\\/ - a literal sequence of ://
• [^\\/]+ - 1 or more characters other than a /
• \\/ - a literal / symbol

Answer 2

Use capturing group based regex.

> var s = "http://foo.com/bar/test/foo.jpeg"
> s.match(/^https?:\/\/[^\/]+((?:\/[^\/]*)*)/)[1]
'/bar/test/foo.jpeg'