lorey / social-media-profiles-regexs Goto Github PK

https?://[w.]{4}?(youtube.com/user/[A-z_0-9-.]{2,100})

JS regex equivalent for all of these be very helpful.

Example: https://www.linkedin.com/company/%C5%AB-hub-made-by-belarusians/

Currently these types of facebook urls are not supported:

https://es-la.facebook.com/Andres.Agulla.ESPN/
https://fr-fr.facebook.com/Andres.Agulla.ESPN/

(?:https?:)?//(?:www.)?(?:facebook|fb).com/(?P(?![A-z]+.php)(?!marketplace|gaming|watch|me|messages|help|search|groups)[A-z0-9_-.]+)/?

my string looks like this

[
   {
      "node":{
         "id":"100084753152635",
         "url":"https:\/\/www.facebook.com\/profile.php?id=100084753152635",
         "name":"Hannah"
      }
   },
   {
      "node":{
         "id":"100049247496610",
         "__isProfile":"User",
         "url":"https:\/\/www.facebook.com\/sayar.tole.31",
         "name":"\u1010\u102d\u102f\u1038 \u101d\u1031 \u101c\u1004\u103a\u1038"
      }
   }
]

Can you add Pinterest regex?

LinkedIn Recruiter urls do not have an in or pub on the path but recruiter.

Example : http(s)?:\/\/([\w]+\.)?linkedin\.com\/(in|pub|recruiter)\/[A-z0-9_-]+\/?

I noticed that the regexes used do not provide username back as part of the regex. An example on how to do this is:

(?:(?:http|https)://)?(?:www.|m.)?facebook.com/(?!home.php)(?:(?:\w)#!/)?(?:pages/)?(?:[?\w-]/)?(?:profile.php?id=(?=\d.*))?([\w.-]+)$

This allows you to filter out all the things you do not need and only return username as a result.
Wouldn't it be a good idea to take this as the standard for this repo?

Example for instagram that only returns username:
https://regex101.com/r/DBVLCq/1

(Might also be a good idea to always include a regex101.com address so people can add tests and improve regexes

Most social networks use https or redirect entries on port 80 to 443 ... unknown that does not use https 😨

In cases (obvious) where the service uses https but the user informs http ... the link may even be correct because it points to a valid endpoint once the application server redirects it, however in my opinion (not based on a normalization) the regex should fail (deny)

The GitHub for example (the regex) in my view is incorrect because the site itself does not transmit data if not by https and the www subdomain is not used

// how are you now
http(s)?:\/\/(www\.)?github\.com/[A-z 0-9 _ -]+\/?

// perhaps a more correct approach
https:\/\/github\.com/[a-zA-Z0-9_-]+\/?

As I was testing some patterns, I noticed that you used 'A-z' for alphanumeric characters. When you do that, it will match any character between 65-122 ASCII, that includes lowercase, uppercase, and a few special characters, including underscores. To fix that and only include characters a-z(97-122) and A-Z(65-90), you need to change 'A-z' to 'a-zA-Z'

Example: https://dk.linkedin.com/company/&-friends

Please review this proposal, because current is false

/^http(s)?:\/\/(www\.)?linkedin\.com\/in\/([a-z0-9_-])*\/?$/i

Flag i is active

Example
https://regexr.com/44amu

I've been developing a library for server-side & client-side link identification & validation. I've been using the non-capturing groups (?:s)? and (?:www\.)?. The regex parser will use it to match the text, but ignores it later when returning the results and can make it easier to identify the account/user/media ID.

Example: ^http(?:s)?:\/\/(?:www\.)?facebook\.com\/([a-zA-Z0-9_]+)$

I see that there hasn't been any additions within the last 2 years. Here are the ones that you've identified that match some of the initial regex filters I've written. (NOTE: These also work extremely well when using the jQuery validation and adding the regex below to the data-rule-pattern attributes.)

Instagram
^http(?:s)?:\/\/(?:www\.)?instagram\.com\/([a-zA-Z0-9_]+)$

Google Plus
^http(?:s)?:\/\/plus\.google\.com\/(.+)$

Pinterest
^http(?:s)?:\/\/(?:www\.)?pinterest\.com\/([a-zA-Z0-9_]+)$

Vimeo Video
^http(?:s)?:\/\/(?:www\.)?vimeo\.com\/(?:channels\/[0-9]+\/)?([0-9]+)$ ^http(?:s)?:\/\/(?:www\.)?vine\.co\/v\/([a-zA-Z0-9_\/]+)$

VimeoChannel
^http(?:s)?:\/\/(?:www\.)?vimeo\.com\/channels\/([0-9]+)$

Skype
^skype:([a-zA-Z0-9_]+)\?call$

WordPress
^http(?:s)?:\/\/(?:[a-zA-Z0-9-]+)?.wordpress\.com\/(?!feed)(.+)?$

YouTube Channel
^http(?:s)?:\/\/(?:www\.)?youtube\.com\/channel\/(?:\w+\/)?([a-zA-Z0-9_-]+)$

YouTube Embed
^http(?:s)?:\/\/(?:www\.)?youtube.com\/embed\/([a-zA-Z0-9_-]+)$

YouTube User
^http(?:s)?:\/\/(?:www\.)?youtube\.com\/user\/(?:\w+\/)?([a-zA-Z0-9_-]+)$

YouTube Video
^http(?:s)?:\/\/(?:www\.)?youtube\.com\/(?!user)(?!channel)(?:\w+\/)?([a-zA-Z0-9_-]+)(?:\/)?$

YouTube Video
^http(?:s)?:\/\/youtu.be/([a-zA-Z0-9_-]+)$

StackOverflow
^http(?:s)?:\/\/(?:www\.)?stackoverflow\.com\/(.+)$

For example https://www.facebook.com and //www.facebook.com would work, whereas www.facebook.com wouldn't work.

Is this an error?

It would be beneficial to have regex's not require the https:// for the off chance they're not specified in some document being scraped

There is a bug in the linkedin recruiter URL regex, it lacks the "profile" part

This is how the link should be
https://www.linkedin.com/recruiter/profile/476162262,HHNH,name

Also, the URL regex for the "talent" endpoint is missing
https://www.linkedin.com/talent/profile/AEEAABxhqNYBJa6QzJWImzsC_q6ugSZg2H6s7pA

Please let me know if you'd like a pull request or you'll fix them by yourself :)

Edit: /sales is also missing.
I'll create a regex and a PR later :)

You should show in the readme.md of your GitHub repo under which license your code can be used.

Regex should include protocol and host Name part as Option not mandatory
Example:
(http(s)?://[\w]+.)?linkedin.com/in/(A-z 0-9 _ -)/?

Let's add a test case for #23 to ensure people see examples of accented slugs in the wild 😁