replace many words with a new one

问题:

I have the following string :

Cat dog fox catepillar bear foxy

I need to replace "cat" and "fox" words from this sentence to word "animal"

I do the following:

$Str1="cat";
$Str2="fox";
$NewStr="animal";

open(F1, "<$inputFile") or die "Error: $!";
open(F2, ">$outputFile") or die "Error: $!";

while ($line = <F1>) {
     $line =~ s/$Str1|$Str2/NewStr/g;
     print F2 "$line";

}

But the problem that word's "catepillar" and "foxy" parts("cat" and fox) also are replaced. How to replace only words "cat" and "fox"?

回答1:

There's a couple more problems here.

# First, always use strict and warnings
# This will save you tons 
use warnings;
use strict;

# declare your vars with 'my' for lexical scope
my $inputFile = "somefile";
my $outputFile = "someotherfile";
my $Str1="cat";
my $Str2="fox";
my $NewStr="animal";

# use 3-arg lexically scoped open
open(my $F1, "<", $inputFile) or die "Error: $!";
open(my $F2, ">", $outputFile) or die "Error: $!";

while (my $line = <$F1>) {
     # surround with word boundary '\b'
     # NewStr is missing the "$"
     $line =~ s/\b(?:$Str1|$Str2)\b/$NewStr/g;
     # UPDATED
     # this should work to remove the parens
     $line =~ s/(\($NewStr\))/$NewStr/g;
     print $F2 "$line";

}

# close your file handles
close $F1;
close $F2;

回答2:

$line =~ s/\b(?:$Str1|$Str2)\b/$NewStr/g;

What the changes mean:

\b zero width assertion for a word boundary

(?: start a group but don't use it for capturing, just grouping

回答3:

Use a word boundary assertion to build your regex. You can find information on it at http://perldoc.perl.org/perlre.html.