javascript正则表达式不起作用

Question

I am trying to validate a textbox to ensure that a URL is written inside (minus the "http://" part). Here is my code:

var isUrl;
var regex = new RegExp("^(?!http)(www\.)?(([a-z])+\.[a-z]{2,}(\.[a-z]{2,})?)");
var siteUrl = e.target.value;
if (siteUrl.match(regex)) {
   isUrl = true;
}
else {
   isUrl = false;
}

So my regular expression is ^(?!http)(www\\.)?(([az])+\\.[az]{2,}(\\.[az]{2,})?) . I was under the impression that this would do the following:

1. NOT match anything beginning with http .. which it does correctly
2. Allow an optional www. at the start
3. Accept 1 or more characters from az to be typed in
4. Accept a dot after those characters
5. Accept 2 or more characters following the dot, and
6. Allow an optional dot followed by two or more characters again.

In practice, the textbox accepts strings like "ashdiguasdf" and "aksjdnfa:://',~" which it should not do. Can anyone explain why?

Answer 1

The main problem is that you're using the \\ character in a quoted string, which javascript will interpret as the start of a "control character" (such as \\n for a newline, etc).

One option is to escape it by replacing \\ with \\\\ .

But the easiest solution is to use the following format...

var regex = new RegExp(/^(?!http)(www\.)?(([a-z])+\.[a-z]{2,}(\.[a-z]{2,})?)/);

This also allows you to make it case insensitive (if you wish) by using the i character at the end, like this...

var regex = new RegExp(/^(?!http)(www\.)?(([a-z])+\.[a-z]{2,}(\.[a-z]{2,})?)/i);

---

As an extra bit, you're using more capture groups than really necessary. Your expression could also be written like this with the same result...

^(?!http)(?:www\.)?[a-z]+(?:\.[a-z]{2,}){1,2}