Re: repeated matches in grep

[Alain Magloire]

> 
> I have seemingly bug-like behaviour from gnu grep version 2.3 and 2.4 for 
> linux.
> 
> I'm trying to find all the instances of a word that contains 6 of the same
> letter by grepping through /usr/dict/words.
> 
> this works:
> 
> grep '\(.\).*\1.*\1.*\1.*\1.*\1' /usr/dict/words
> 
> but this does not:
> 
> grep '\(.\)\(.*\1\)\{5\}' /usr/dict/words
> 
> Shouldn't they be equivalent?
> 
> In perl, the regular expression (.)(.*\1){5} which seems to be equivalent
> gives the correct results.
> 
> I guess this may not be a failure of grep, but rather the regular
> expression library... In any case. 

Without too much thought to it, I would say that the engine
is being greedy and matches the longest possible match with
this expression (.*\1)  before trying {5}