awk programming question

[Warren Block]

On Fri, 24 Jan 2014, dteske at FreeBSD.org wrote:

>>> I'm not entirely sure how I feel about that -- in terms of
>>> readability, I'm not sure if the following is more readable:
>>>
>>> 	/(a[^z]*z)/
>>
>> Wait till you see the other PCRE stuff.  There are lots of things 
>> that really make it much more powerful.  perlre(1) covers it all, 
>> sketchily and not really in order.
>>
>> I've said elsewhere, and will repeat again: "Mastering Regular 
>> Expressions" by Jeffrey E. F. Friedl (the owl book) is amazing.  My 
>> first edition (1997) does not have some of the newer Perl stuff, but 
>> it's now up to a third edition:
>>
>> http://shop.oreilly.com/product/9780596528126.do
>
> Does it cover "back references" ? Definitely one of the more powerful 
> but esoteric regular expressions (e.g., you want to match a string 
> that starts with a quote and has a matching terminating quote, but not 
> match a string that has a quote without matching termination).

It does cover backreferences, in the sense of egrep:
\<[A-Za-z]+) +\1\>

That will find a word followed by one or more spaces and the same word 
again.

> If-so, I'm totally buying that book.

Buy it.  I've never seen another book that treats such a complex subject 
with such clarity and readability.