CODEWITHSUNDEEP
javascriptregex
Code Guy
Code Guy

Reputation: 3198

Extract the middle names and last name from the string using regex of Javascript

I need to extract the middle and last names

Elev: 7EBB49 (Dan Greg Järgenstedt <[email protected]>)
Expected: Greg Järgenstedt

Elev: 6EBB49 (Dan Järgenstedt <[email protected]>)
Expected: Järgenstedt

Elev: 6EBB49 (Järgenstedt <[email protected]>)
Expected: Järgenstedt

Elev: 6EBB49 (<[email protected]>)
Expected:

Tried with

function getSNames(input) {
    const names = input.match(/(?<!\[)(?<=\s)\w+(?=\s)/g);
    return names ? names.join(' ') : '';
}

Upvotes: 3

Views: 243

Answers (2)

Peter V. M&#248;rch
Peter V. M&#248;rch

Reputation: 15907

This way madness lies.

I don't think you can "extract the middle and last names" in general. If all you have is "Name", then you're stuck with it. Whatever rule you come up with, I'll show you a person where it doesn't work. E.g.

  • My name is "Peter Valdemar Mørch". first name=Peter, middle name=Valdemar, last name=Mørch
  • I have a friend called Jens Erik Redacted. first name=Jens Erik, no middle name, last name=Redacted. It insults him a little every time a website says "Hi Jens", because he isn't "Jens", he's "Jens Erik".
  • I have another friend called Kristian von Hornsleth. first name=Kristian, no middle name, lastname=von Hornsleth
  • In Chinese, the last name comes before the first name. E.g. (I don't know him) "王秀英" is spelled "Wang Xiuying" in English. firstname=Xiuying, no middlename, lastname=Wang (notice how they're swapped)

What you're doing won't work in the general case, especially not internationally.

Upvotes: 1

Wiktor Stribiżew
Wiktor Stribiżew

Reputation: 626747

You can use

const names = input.match(/(?<!\(\p{L}+\s+|\p{L})\p{L}+(?:\s+\p{L}+)*(?=\s*<)/gu)

See the regex demo. The u flag enables the Unicode category classes.

Pattern details

  • (?<!\(\p{L}+\s+|\p{L}) - immediately to the left, there cannot be a ( followed with 1+ letters and then one or more whitespaces, or just a letter (this works as a Unicode word boundary)
  • \p{L}+ - one or more letters
  • (?:\s+\p{L}+)* - zero or more occurrences of 1+ whitespaces and then 1+ letters
  • (?=\s*<) - immediately to the right, there must be 0+ whitespaces and then <.

Upvotes: 2

Related Questions