(Daddio) Re: Deleting from front of string

[Daddio]

Other than substr, there is always the regex method:

$_ = "just another perl hacker";
s/^.{2}(.*)$/$1/;
print;

D a d d i o

[hopes]

What happens with this string?

$_=qq!\nJ\nu\ns\t another Perl hacker!;


Your regexp fails, because \n is not matched by /./ Nobody have said that the two first characters matches with /./
Try this


s/(\w|\W){2}//;

Regards
Hopes

[pjf]

Very well spotted, Hopes. Dot not matching a newline is such a common mistake it's easy to forget. I'm glad that you noticed it.

<pedant>

Actually, it's possible to get dot to match everything by using the /s switch. This also has the benefit that ^ and $ will match the beginning and end of the string, and not at line breaks, which for our purposes here is desirable. See the perlre man page for details.

Hence $string =~ s/^..//s does what we want, even if we are dealing with newlines. If you don't mind losing clarity, you can even drop the ^ in this situation.

</pedant>

Having said all that, the substr method described above would be my recommendation. (It's also likely to be the fastest). ;)

Cheers,
Paul

[blakem]

I think you gave /s a few features that it doesn't posses. /s changes what '.' will match, just like you said, but if you want the '^' and '$' anchors to match around an embedded newline, you'll need to toss in the /m switch as well...

Update: After re-reading the node I followed up to, I think we're both correct.... consider this a clarification of the node above it, rather than a correction. ;-)

-Blake

[cLive ;-)]

why use $1?

s/^.{2}//s;

The s modifier at the end takes care of \n chars as well.

But substr() is the best way anyway... :)

.02

cLive ;-)