remove chars with regex

frank1 has asked for the wisdom of the Perl Monks concerning the following question:

Replies are listed 'Best First'.
Re: remove chars with regex by ikegami (Patriarch) on Feb 26, 2025 at 14:15 UTC
The error come from using `=` when you meant to use `=~`. But the pattern is also wrong. There will never be 5 non-space characters before the first character of the string, for starters. Fixed: `my ( $output ) = $char =~ /(?<=^\S{5})(\S+)(?=\S{5}\z)/;` [download] There's really no point in using look-arounds or `\S`, though. `my ( $output ) = $char =~ /^.{5}(.*).{5}\z/s;` [download] Faster and cleaner: `my $output = substr( $char, 5, -5 );` [download]	[reply] [d/l] [select]
Re^2: remove chars with regex by ForgotPasswordAgain (Vicar) on Feb 27, 2025 at 03:07 UTC
I think that would return the middle part but not result in "(5 chars) ... (5 chars)". More like `$ perl -Mstrict -wE'my $char = "0x388c818ca8b9251b398a736a67ccb19297"; substr($char, 5, -5, " ... "); say $char' 0x388 ... 19297` [download] (assuming the string is long enough)	[reply] [d/l]
Re^3: remove chars with regex by ikegami (Patriarch) on Feb 27, 2025 at 11:18 UTC
To remove the middle instead of keeping the middle, you could use `substr( $char, 5, -5, "" ); # In-place` [download] or `my $output = substr( $char, 0, 5 ) . substr( $char, -5 );` [download]	[reply] [d/l] [select]
Re: remove chars with regex by Marshall (Canon) on Feb 26, 2025 at 09:29 UTC
`use strict; use warnings; my $char = "0x388c818ca8b9251b393131c08a736a67ccb19297"; $char =~ s/^([0-9a-fx]{6}).+([0-9a-f]{6})$/$1 \.\.\. $2/; print $char; # 0x388c ... b19297` [download]	[reply] [d/l]
Re^2: remove chars with regex by frank1 (Monk) on Feb 26, 2025 at 09:36 UTC
Thanks @Marshall	[reply]
Re: remove chars with regex by choroba (Cardinal) on Feb 26, 2025 at 09:41 UTC
To get the output you want, you don't need any look-arounds: `my ($start, $end) = $char =~ /^(\S{6}).(\S{6})$/;` [download] Note that it's six characters, not five. Also, don't forget to check the length of the string, as shorter strings won't match at all. `map{substr$_->[0],$_->[1]\|\|0,1}[\\|\|{},3],[[]],[ref qr-1,-,-1],[{}],[sub{}^*ARGV,3]`	[reply] [d/l] [select]
Re: remove chars with regex by johngg (Canon) on Feb 26, 2025 at 11:05 UTC
An alternative to using a regex is substr applied at either end of the string. `johngg@aleatico:~$ perl -Mstrict -Mwarnings -E 'say q{}; my $char = q{0x388c818ca8b9251b393131c08a736a67ccb19297}; say join q{ ... }, substr( $char, 0, 6 ), substr( $char, length( $char ) - 6, 6 );' 0x388c ... b19297` [download] I hope this is helpful. Update: Re. AnonyMonk's comment regarding negative offsets, I had completely forgotten that, how dumb am I :-( Update 2: Here's the amended code. `johngg@aleatico:~$ perl -Mstrict -Mwarnings -E 'say q{}; my $char = q{0x388c818ca8b9251b393131c08a736a67ccb19297}; say join q{ ... }, substr( $char, 0, 6 ), substr( $char, -6 );' 0x388c ... b19297` [download] Cheers, JohnGG	[reply] [d/l] [select]
Re^2: remove chars with regex by Anonymous Monk on Feb 26, 2025 at 15:32 UTC
FWIW, the second `substr ...` (for the last 6 characters) could be `substr( $char, -6 )`. Negative offsets are from the end, and if you omit the length you get all of the string after the start position.	[reply] [d/l] [select]
Re: remove chars with regex by hippo (Archbishop) on Feb 27, 2025 at 08:14 UTC
Since there hasn't yet been a solution posted using lookarounds, here is one: `use strict; use warnings; use Test::More tests => 1; my $have = '0x388c818ca8b9251b393131c08a736a67ccb19297'; my $want = '0x388c ... b19297'; $have =~ s/(?<=^.{6}).*(?=.{6}$)/ ... /; is $have, $want;` [download] Note that I probably would not do this in production and would instead use a substr-based approach. 🦛	[reply] [d/l]