Re: regexec() trouble

[Scott Klement <midrange-l@xxxxxxxxxxxxxxxx>]

Ooops... I have to correct one thing in my last post. You also have to 
treat { and } differently in a BASIC regular expression. They need to bre 
preceded by backslashes. So, if you want to use REG_BASIC, your pattern 
has to look like this:

^[A-Z0-9._%-][A-Z0-9._%-]*@[A-Z0-9.-][A-Z0-9.-]*\.[A-Z]\{2,4\}$

I had forgotten about the backslashes in my previous post.

With an extended RE, you don't need to add the backslashes, and your 
original pattern should work just fine.  Clearly, that pattern was 
intended to be an extended expression.

For more info on RE's, see the following page:
http://en.wikipedia.org/wiki/Regular_expression

The "Syntax" section has some info on the differences between basic and 
extended regular expressions.


On Wed, 28 Jun 2006, Scott Klement wrote:

> [SNIP]
> > // From: http://www.regular-expressions.info/email.html
> > RegPattern = '^[A-Z0-9._%-]+@[A-Z0-9.-]+\.[A-Z]{2,4}$';
> > rc = regcomp(reg: %trimr(RegPattern): REG_BASIC+REG_ICASE+REG_NOSUB);
> > // Why won't this work?!?
> > Chr256 = 'test@xxxxxxxxxxx';
> > rc = regexec(reg: %trim(Chr256): 0: match: 0);
> [SNIP]
>
> The problem is the plus symbol.  On the iSeries (I don't know about 
> elsewhere?) the + symbol is only supported in extended regular expressions, 
> it's not supported in basic ones.
>
> What the + symbol does is match "one or more" of the preceding character. 
> This is in contrast to the * (asterisk) symbol which matches "zero or more" 
> of the preceding symbol.
>
> So if you had a pattern of [A-Z]+ (in an environment that supports the plus 
> symbol) it'll match one or more of the letters from A-Z. Or, slightly more 
> complicated (in your example) you have [A-Z0-9._%-]+, which will match one or 
> more of any letter, any number, a dot, a percent symbol, or a dash.
>
> If you change it to [A-Z0-9._%-]* it'll match zero or more -- which means 
> that it could omit those characters entirely.  Whereas with a +, there has to 
> be at least one.
>
> Since the iSeries doesn't support the + operator, you could achieve the same 
> results by changing your pattern to look like this:
>
> ^[A-Z0-9._%-][A-Z0-9._%-]*@[A-Z0-9.-][A-Z0-9.-]*\.[A-Z]{2,4}$
>
> All I did was repeat the bracketed expressions, and change the + to a *. Now 
> it has to match one character.  Followed by zero or more characters. That'd 
> have the same effect as the original expression you posted, and would work on 
> the iSeries as a basic expression.
>
> The other alternative is to compile your expression as an EXTENDED regular 
> expression instead of a BASIC one.  Extended expressions DO support the plus 
> symbol, and therefore your original expression should work fine.  To make it 
> extended, change REG_BASIC to REG_EXTENDED on the call to regcomp().
>
> I hope my explanation makes sense, I was struggling to find the right words!
>
> ---
> Scott Klement  http://www.scottklement.com