[Ltru] Regex

[Mark Davis <mark at macchiato.com>]

I just redid my regex for the new version. The only tweak I made was that it only allows one extlang (as per the description, item 4), which means that I had to retain zh-min-nan as irregular.

(?i)
  (?:
      (?: ( [a-z]{2,8} | [a-z]{2,3} [-_] [a-z]{3} )
      (?: [-_] ( [a-z]{4} ) )? 
      (?: [-_] ( [a-z]{2} | [0-9]{3} ) )? 
      (?: [-_] ( (?: [a-z 0-9]{5,8} | [0-9] [a-z 0-9]{3} ) (?: [-_] (?: [a-z 0-9]{5,8} | [0-9] [a-z 0-9]{3} ) )* ) )? 
      (?: [-_] ( [a-w y-z] (?: [-_] [a-z 0-9]{2,8} )+ (?: [-_] [a-w y-z] (?: [-_] [a-z 0-9]{2,8} )+ )* ) )? 
      (?: [-_] ( x (?: [-_] [a-z 0-9]{1,8} )+ ) )? ) 
    | ( x (?: [-_] [a-z 0-9]{1,8} )+ ) 
    | ( en [-_] GB [-_] oed
      | i [-_] (?: ami | bnn | default | enochian | hak | klingon | lux | mingo | navajo | pwn | tao | tay | tsu )
      | no [-_] (?: bok | nyn )
      | sgn [-_] (?: BE [-_] (?: fr | nl) | CH [-_] de )
      | zh [-_] min [-_] nan ) )

As before,

• the (?i) is for case insensitive matching,
  
• the [-_] is for implemenatations (like Unicode) that allow alternate separators (can be replaced by [-] otherwise), and
  
• the (?: is Perl/Java syntax for non-capturing groups. [That is, the regular (..) capture the main components of the regex, for extraction later.]
  

Mark