Get only URI segments from URL

Question

I am trying to get the URI segments using regular expression.

Example URI:

http://abc.com/hello/hi/bye?humm/ok=hi&ya=yaya/wow/waaah
               ^^^^^ ^^ ^^^                    ^^^ ^^^^^

I am trying:

/(?<=\/)[\w-]+(?=(\/|$|\r|\?))/g

But it's not working properly. The query string is not getting excluded (wow/waaah).

So, when I tried the following, everything got excluded:

/(?<!?.+)(?<=\/)[\w-]+(?=(\/|$|\r|\?))/g

What's wrong with this?

Answer 1

You forgot to escape the second ? in the second regex. It should read:

/(?<!\?.+)(?<=\/)[\w-]+(?=(\/|$|\r|\?))/g

Note: You could improve the regex by using character classes like so:

/(?<!\?.+)(?<=\/)[\w-]+(?=[/\r\n?]|$)/g

EDIT:

For a lowest common denominator solution to cater for all the different flavours of regex, you need a two step process:

• Remove the trailing ? and all following chars (if it exists):
  

  
  ^[^/]+//[^/]+([^?]+)

  Keep the string returned in capture group 1.

• Extract the URI segments by looping through:

  
  /([\w-]+)

  The segments are returned in capture group 1.