Regex to extract both video id or playlist id from youtube url

Question

I would like to know how to extract youtube video id or playlist id depending upon the url using a single regex expression. The regex should also ensure that the domain is youtube.com Here are some of the results I need:

Extract Playlist ID For

    https://www.youtube.com/playlist?list=PLuC2HflhhpLGQ4RgqA76_Gv52fGA0909r
    www.youtube.com/playlist?list=PLuC2HflhhpLGQ4RgqA76_Gv52fGA0909r
    http://www.youtube.com/playlist?list=PLuC2HflhhpLGQ4RgqA76_Gv52fGA0909r
    https://www.youtube.com/embed/videoseries?list=PLuC2HflhhpLGQ4RgqA76_Gv52fGA0909r  

Extract Video ID For

https://www.youtube.com/watch?v=fqMfRi2gJok&index=1&list=PLuC2HflhhpLGQ4RgqA76_Gv52fGA0909r
https://www.youtube.com/watch?v=fqMfRi2gJok
http://youtu.be/cCnrX1w5luM 
http://youtube.com/embed/cCnrX1w5luM
http://youtube.com/v/cCnrX1w5luM
https://www.youtube.com/v/cCnrX1w5luM
www.youtube.com/v/cCnrX1w5luM
youtube.com/v/cCnrX1w5luM

These are just example urls only. I need to extract respective ID's for all possible youtube link structures.

In short extract video id and if it is absent obtain playlist id.

Answer 1

Your problem is explicitly has two patterns

The first:

^.*?(?:v|list)=(.*?)(?:&|$)

For any urls which have explicit attribute, or you can say they have = symbol in url.

Explanation

^.*?(?:v|list)=: Any string till word v= or list= which here we prefer v over list,

(.*?)(?:&|$): Any string which ended by & symbol or ending line symbol $ which here we prefer & over $.

The second:

^(?:(?!=).)*\/(.*)$

For any url which don't have attribute or there is no = symbol in url.

Explanation

^(?:(?!=).)*\/: Any string which don't have = symbol (here handle by the negative lookahead (?!=)) till / symbol,

(.*)$: Any string till the end of line.

Combine them into one regex we get

^(?:https?:\/\/)?(?:www\.)?youtu\.?be(?:\.com)?.*?(?:v|list)=(.*?)(?:&|$)|^(?:https?:\/\/)?(?:www\.)?youtu\.?be(?:\.com)?(?:(?!=).)*\/(.*)$

here,

(?:https?:\/\/)?(?:www\.)?youtu\.?be(?:\.com)? is added to handle various form of www.youtube.com's url

and this should help you get what you want

see: DEMO

IMPORTANT NOTE: This question, questioner want to extract id from www.youtube.com which he prefer "video id" over "playlist id".

Answer 2

https://regex101.com/r/mI3qY9/4

This regex assumes you are giving it a legitimate Youtube link. This grabs all the v and lists together:

/(?:(?:\?|&)(?:v|list)=|embed\/|v\/|youtu\.be\/)((?!videoseries)[a-zA-Z0-9_]*)/g

Breakdown:

/
(?:                         //non-capturing group
  (?:\?|&)(?:v|list)=       //? or & following a v or list
  |                         //or
  embed\/                   //embed/
  |                         //or
  v\/                       //v/            
  |                         //or
  youtu\.be\/               //youtu.be/
)
(
  (?!videoseries)           //will not capture "videoseries"
  [a-zA-Z0-9_]*             //capture any alphabet digits or underscore that follows afterwards
)          
/g                          //global

But you may not be able to tell which is v and which is list, so,

This only grabs the v:

/(?:(?:\?|&)v=|embed\/|v\/|youtu\.be\/)((?!videoseries)[a-zA-Z0-9_]*)/g

This only grabs the list:

/(?:(?:\?|&)list=)((?!videoseries)[a-zA-Z0-9_]*)/g

This only grabs YouTube vs:

/(?:youtube\.com.*(?:\?|&)(?:v)=|youtube\.com.*embed\/|youtube\.com.*v\/|youtu\.be\/)((?!videoseries)[a-zA-Z0-9_]*)/g

Only YouTube lists:

/(?:youtube\.com.*(?:\?|&)(?:list)=)((?!videoseries)[a-zA-Z0-9_]*)/g

This is basically the same but adding youtube\.com.* too to the regex. It won't grab e.g. http://example.com/v/abc

https://regex101.com/r/mI3qY9/5

Explanation:

youtube\.com.*          //Matches youtube.com and any multiple characters followed