From 8f6bf384c28276b6f05f9679e9a4e690bdc79803 Mon Sep 17 00:00:00 2001
From: Jonas Smedegaard <dr@jones.dk>
Date: Sun, 26 Sep 2010 18:59:53 +0200
Subject: Fix UTF-8 IO handling.

---
 localmarkdown2sms | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

(limited to 'localmarkdown2sms')

diff --git a/localmarkdown2sms b/localmarkdown2sms
index 7d03df2..86b0f6d 100755
--- a/localmarkdown2sms
+++ b/localmarkdown2sms
@@ -22,6 +22,7 @@
 
 use strict;
 use warnings;
+use utf8; # this document itself is UTF-8 encoded.
 
 use Env qw[$debug $info $warn $dummy $nosleep $urldecode];
 use Log::Log4perl qw(:easy);
@@ -68,8 +69,8 @@ if ($urldecode) {
 	@ARGV = uri_unescape(@ARGV);
 }
 my ($phone) = shift @ARGV;
-my $inputstring = join(' ', @ARGV);
-my ($key) = lc (shift @ARGV);
+my $inputstring = decode('UTF-8', join(' ', @ARGV));
+my ($key) = lc(decode('UTF-8', shift @ARGV));
 
 # strip international prefix
 # (prefix is optional some places and illegal at other places - forgot where)
@@ -107,8 +108,8 @@ foreach my $file (read_dir( $path )) {
     my ($warn_nonkey_delay, $warn_nonkey_content);
     next unless ($file =~ /\.mdwn$/);
     foreach my $line (read_file( File::Spec->catfile($path, $file))) {
+	$line = &transliterate(decode('UTF-8', $line));
 	chomp $line;
-	$line = &transliterate($line);
 	my $content;
 	# headline
 	if ($line =~ /^(#+)\s*(.*?)\s*$/) {
@@ -207,11 +208,11 @@ sub tidymsg {
 sub transliterate {
 	my $string = shift @_;
 
-	my $res = encode('UTF-8', decode('GSM0338', encode('GSM0338', $string, sub {
+	my $res = decode('GSM0338', encode('GSM0338', $string, sub {
 		my $ascii = unidecode(chr $_[0]);
 		_utf8_off($ascii);
 		$ascii;
-		})));
+		}));
 
 	return $res || "";
 }
-- 
cgit v1.2.3