Re: Regex advice

Steve Bertrand wrote:
>>> if 
>>> (/.*simscan~\[\d+\]~([\w|\s]+).*?~(\d+\.\d+)s~.*?~(.*?)~(.*?)~(.*)/) {
>>
>> The .* pattern at the beginning is useless as it causes unnecessary 
>> back-tracking, just remove it.  
> 
> Ok.
> 
>> The character class [\w|\s] includes both the '|' and '_' characters 
>> but it doesn't appear that that field would normally contain either a 
>> '|' or a '_' character.
> 
> It wouldn't (normally). I was with a misconception on how this worked. I 
> thought it meant 'capture this hunk if it contains word or whitespace in 
> any amount'.

The \w character class matches any "word" character where "word" is 
defined as any alphabetic character or any numerical digit or the '_' 
character.  The '|' which is used for alternation outside of a character 
class just matches a literal '|' character inside a character class. 
The \s character class matches any whitespace character.


> I have reading to do.
> 
>>>                 $scantime = $2 if $2;
>>
>> (\d+\.\d+) will always match at least three characters so it can never 
>> be false so the test is superfluous.
> 
> ...hrm... more reading to do.
> 
>> John
> 
> This is more about receiving feedback and not about an answer, so as I 
> go find out why things work as they have been described, would you 
> project the statement as if you had written it yourself, just for 
> comparison sake?

I don't think I understand your last question.



John
-- 
Perl isn't a toolbox, but a small machine shop where you
can special-order certain sorts of tools at low cost and
in short order.                            -- Larry Wall