Hi,

I'd like some thoughts ~> input on extended regular expressions used 
with grep, specifically GNU grep -e / egrep.

What are the pros / cons to creating extended regular expressions like 
the following:

    ^\w{3}

vs:

    ^(Jan|Feb|Mar|Apr|May|Jun|Jul|Aug|Sep|Oct|Nov|Dec)

Or:

    [ :[:digit:]]{11}

vs:

    ( 1| 2| 3| 4| 5| 6| 7| 8| 
9|10|11|12|13|14|15|16|17|18|19|20|21|22|23|24|25|26|27|28|29|30|31) 
(0|1|2)[[:digit:]]:(0|1|2|3|4|5)[[:digit:]]:(0|1|2|3|4|5)[[:digit:]]

I'm currently eliding the 61st (60) second, the 32nd day, and dealing 
with February having fewer days for simplicity.

For matching patterns like the following in log files?

    Mar  2 03:23:38

I'm working on organically training logcheck to match known good log 
entries.  So I'm *DEEP* in the bowels of extended regular expressions 
(GNU egrep) that runs over all logs hourly.  As such, I'm interested in 
making sure that my REs are both efficient and accurate or at least not 
WILDLY badly structured.  The pedantic part of me wants to avoid 
wildcard type matches (\w), even if they are bounded (\w{3}), unless it 
truly is for unpredictable text.

I'd appreciate any feedback and recommendations from people who have 
been using and / or optimizing (extended) regular expressions for longer 
than I have been using them.

Thank you for your time and input.



-- 
Grant. . . .
unix || die