Backwhacking with \ as delimiter

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

Replies are listed 'Best First'.
Re: Backwhacking with \ as delimiter by BrotherAde (Pilgrim) on Aug 26, 2001 at 16:13 UTC
I usually use any "weird" character which does not occur in the regex. My favourite is the tilde, not only for regexes, but also for `print qq~I said "foo"~;` constructs. BrotherAde	[reply] [d/l]
Re: Backwhacking with \ as delimiter by John M. Dlugosz (Monsignor) on Aug 27, 2001 at 00:33 UTC
Try interpolating it in from another string?	[reply]
Re: Backwhacking with \ as delimiter by BrentDax (Hermit) on Aug 27, 2001 at 08:41 UTC
Normally I'd say "Don't use \ as a delimiter!" but this is a JAPH... :^) Perhaps you could do something clever with tr/// and eval: `($re='s\\foo~.bar\\foobar\\') =~ tr/\\~/~\\/; # ~ for \ eval $re;` [download] =cut --Brent Dax There is no sig.	[reply] [d/l]
Re: Backwhacking with \ as delimiter by flocto (Pilgrim) on Aug 26, 2001 at 16:04 UTC
I'd never use backslash ('\') as "regex-delimiter". All special characters are escaped with it so you'll just create confusion. Use the normal slash ('/') and you won't have any problems. I've seen '#' being used, too, but personally I don't like the look of it. But that might be just me..	[reply]
Re: Re: Backwhacking with \ as delimiter by rob_au (Abbot) on Aug 26, 2001 at 16:24 UTC
You could always still use the \ as a delimiter within $_, pass it through another substitution (`s///`) to replace it and then another to escape characters where necessary. Eval, rinse and repeat! Yes, it makes the process more convoluted, but then again with obfuscation, this can be a good thing! :) Ooohhh, Rob no beer function well without!	[reply] [d/l]