#!/usr/bin/perl -w use strict; use warnings; use autodie; binmode STDOUT, ':utf8'; chdir "storepages"; for my $name (glob "*.html") { (my $id = $name) =~ s/\.html$//; open my $fh, "<", $name; binmode $fh, ':utf8'; my @lines = <$fh>; my ($line) = grep { /"tagid":/ } @lines; my $tags = ''; if ($line) { $tags = join ' ', map { s/^\s+|\s+$//g; s/ /-/g; lc $_ } $line =~ /"name":"(.*?)"/g; $tags =~ s/\\u(\S{4})/chr(hex($1))/ge; } print qq{$id,"$tags"\n}; }