From 5e6c8aeb17de4f88bc6366c2fe045a1b82acb2bc Mon Sep 17 00:00:00 2001
From: Brian O'Donnell <brian@hzsogood.net>
Date: Wed, 22 Apr 2015 10:56:11 -0400
Subject: [PATCH] Changed string parsing to handle unicode characters properly

---
 pod_feeder.pl | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/pod_feeder.pl b/pod_feeder.pl
index 4d3f573..597179a 100755
--- a/pod_feeder.pl
+++ b/pod_feeder.pl
@@ -243,13 +243,13 @@ sub get_feed_items {
                         $link_part =~ s/(\?.*)$//;
 
                         # split up string
-                        my @parts = split( /[^A-Za-z0-9]/, $link_part );
+                        my @parts = split( /^(\p{Letter}|\p{Number})]/, $link_part );
 
                         push( @hashtags, @parts );
                 }                                 
 
                 # try to guess tags from the title
-                if( $params{'extract_tags_from_title'} ){   
+                if( $params{'extract_tags_from_title'} ){
                         # split up string
                         my @parts = split( /\s+/, $item->{'title'} );
 
@@ -321,8 +321,8 @@ sub hashtagify {
         my @list = @$list_ref;
 
         foreach my $item ( @list ){
-                # remove non-alphanumerics and spaces
-                $item =~ s/[^A-Za-z0-9]//g;   
+                # remove non-alphanumerics
+                $item =~ s/[^(\p{Letter}|\p{Number})]//g;   
 
                 # drop stop words
                 next if length( $item ) < 3;