How can I make this URL validation regular expression less greedy?

So I have the following regular expressi开发者_运维知识库on:

https?://(www\.)?flickr\.com/photos/(.+)/?

To match against the following URL:

http://www.flickr.com/photos/username/

How can I stop the final forward slash (/) from being included in the username sub-pattern (.+)?

I have tried:

https?://(www\.)?flickr\.com/photos/(.+?)/?

But then it only matches the first letter of the username.

---

https?://(?:www\.)?flickr\.com/photos/([^/]+)/?

I added ?: to the first group so it's not capturing, then used [^/] instead of the dot in the last match. This assures you that everything between "photos/" and the very next "/" is captured.

If you need to capture the first www just use this:

https?://(www\.)?flickr\.com/photos/([^/]+)/?

---

You need to make sure it doesn't match the forward slash:

https?://(?:www\.)?flickr\.com/photos/([^/]+)/?

You could also make the regex lazy (which is what I guess you were doing with the (.+?) syntax), but the above will work just fine

---

Change (.+) to ([^/]+). This will match until it encounters a /, so you might want to throw some other stuff in the class too.

---

There are generally two ways to do this:

Append a question mark, to make the matching non-greedy. .* will match as much as possible, .*? will match as little as possible.

Exclude the character you want to match next. If you want to stop on /, use [^/]*.

---

If you know there will be a trailing slash, take out the final ?.