Regex match iterative grouping

[SamCG]

Okay, the code:

perl -e "@re='abcdefg'=~/(..)/g;print join ',', @re;"
#prints ab,cd,ef

but, what if I don't want the position to set to after the last match, but to one character after the beginning of the last match to produce ab,bc,cd,de,ef,fg? I've tried a few strange variants of adding code into the expression to alter the 'abcdefg' midway, but these don't seem to produce the results I want.

perl -e "$_='abcdefg';@re=/(..)(?{reverse;chop;reverse;})/g;print join
+ ',', @re;"
ab,cd,
perl -e "$_='abcdefg';@re=/(..)(?{$_=substr($_,1)})/g;print join ',', 
+@re;"
bc,ef,

No particular purpose here other than just to wield regexes better, so if there are practical reasons this is ludicrous feel free to tell me.

-----------------
_{s''limp';@p=split '!','n!h!p!';s,m,s,;$s=y;$c=slice @p1;so brutally;d;$n=reverse;$c=$s**$#p;print(''.$c^chop($n))while($c/=$#p)>=1;}

[Ido]

Quiet interesting, but:


'abcdefg'=~/(?=(..))/g;


Does the job...
Please see: Repeated patterns matching zero length substring

[sfink]

Adjust the quotes for your shell:

perl -le '$_ = "abcd"; while (/(..)/g) { print $1; --pos }'

Or closer to your exact question:

perl -le '$_ = "abcdefg"; while (/(...)/g) { print $1; pos($_) -= leng
+th($1); pos($_)++ }'

You might be able to do that in a list-context match with an embedded expression, but that would be too scary for me.

[SamCG]

Thank you, both solutions work, though I will have to work on understanding Ido's a bit better.

Note that as far as "adjusting my quotes to my shell", that technique doesn't work on my <Windows XP> shell. Inverting them (double-quotes surrounding all executable code, with either single quotes in the middle or escaped double-quotes) does work.