Down Time Project

I strained some lower back muscles on Thursday and since then I've been taking muscle relaxants, ibuprofen, sleeping a lot and generally taking it easy. I took advantage of the down time to start writing the alternate regular expression parser I recently blogged about. The syntax has changed a little but the spirit of the original idea are intact. The following are some samples of what it can generate.

Standard Regex:

^(([^<>()[\]\\.,;:@"]+(\.[^<>()[\]\\.,;:@"]+)*)|(".+"))

Verbose Regex:

define restricted ('<>()[]\\.,;:@"')

anchor begin

group(

   group(

      oneOrMore(notAny(restricted)) +

         zeroOrMore('.' + oneOrMore(notAny(restricted))

      )

   )

or group('"' + oneOrMore(any) + '"')

)

Standard Regex:

^[a-zA-Z0-9](([_\.\-]?[a-zA-Z0-9]+)*)@([a-zA-Z0-9]+)

(([\.\-]?[a-zA-Z0-9]+)*)\.([a-zA-Z]{2,})$

Verbose Regex:

define ALPHANUM (range('a','z','A','Z','0','9'))

define ALPHA (range('a','z','A','Z'))

anchor begin

ALPHANUM + 

group(zeroOrMore(zeroOrOne('_.-') + oneOrMore(ALPHANUM)))

+ '@' +

group(oneOrMore(ALPHANUM)) + 

group(zeroOrMore(zeroOrOne(any('.-')) + oneOrMore(ALPHANUM))) + 

'.' + 

group(repeat(ALPHA,2, ))

end anchor

The expression (Standard and Verbose) in each box are identical. The result of running the verbose syntax parser is the standard expression.

I haven’t added the named groups feature I talked about yet. I wanted to get the basic parser done first. I also need to expand my set of unit tests and flush out the capabilities. I’ve seen certain expressions that I cannot even read so it’s quite difficult to know how to parse them. If nothing else this project is honing my regular expression skills.