Perl，模式匹配和元字符

Question

I am trying to match two things which both are full of metacharacters that needs to be used as 'Literal' in my match pattern. \\Q is suppose to quote all metacharacter in a string until \\E ...but it doesn't work.

Whats up with that?

this is the line that gives me trouble : if (/\\Q$prev\\E/ !~ /\\Q$ww[0]\\E/) {

Answer 1

Absent the use of =~ or !~ ,

/.../

is short for

$_ =~ m/.../

so

/\Q$prev\E/ !~ /\Q$ww[0]\E/

is short for

($_ =~ /\Q$prev\E/) !~ /\Q$ww[0]\E/

which is equivalent to one of the following depending on whether the left regex match succeeds or not:

"" !~ /\Q$ww[0]\E/
"1" !~ /\Q$ww[0]\E/

You simply want:

$prev !~ /\Q$ww[0]\E/   # $ww[0] doesn't contains $prev

If you actually want

$prev !~ /^\Q$ww[0]\E\z/   # $ww[0] isn't equal to $prev

then you can simplify that to

$prev ne $ww[0]   # $ww[0] isn't equal to $prev

By the way, always use use strict; use warnings; use strict; use warnings; . It may have identified a problem here (but not necessarily, depending on the value of $_ ).

Answer 2

It looks like you want to compare a string in $prev to one in $ww[0] . If this is the case, a regex match should look like this:

$result = $prev !~ /\Q$ww[0]\E/

$result will return 1 if $prev is not the same as whatever is in www[0] , ignoring metacharacters.

However if that is all you wanted to do, you might as well use ne :

if ($prev ne $ww[0]){ 
   #do this if $prev and $ww[0] are not the same
} 

---

Also, as @toolic has mentioned, add the following line to the top of your script:

use warnings;

This will give you some feedback on possible problems in your scripts.