From 7b02e3ff704d028fea8653ff63139c674c3226b3 Mon Sep 17 00:00:00 2001
From: Kalle Olavi Niemitalo <kon@iki.fi>
Date: Wed, 31 Oct 2007 22:39:19 +0200
Subject: [PATCH] Convert *.strings to UTF-16 at build time.

<http://developer.apple.com/documentation/MacOSX/Conceptual/BPInternational/Articles/StringsFiles.html>
"Note: Strings files encoded using UTF-8 are not guaranteed to work.
When in doubt, encode the file using UTF-16."

However, Git does not display readable diffs for UTF-16 files.
So keep *.strings files as UTF-8 in the source tree, but convert
them to UTF-16 during the build.

Not all the strings files in the source tree are in UTF-8 yet,
and not all of the UTF-8 ones begin with a byte-order mark.
I intend to convert files to UTF-8 with BOM as I modify them.
---
 MacTF.xcodeproj/project.pbxproj | 17 +++++++++++++++++
 to-UTF-16.pl                    | 37 +++++++++++++++++++++++++++++++++++++
 2 files changed, 54 insertions(+)
 create mode 100755 to-UTF-16.pl

diff --git a/MacTF.xcodeproj/project.pbxproj b/MacTF.xcodeproj/project.pbxproj
index bf6e0c5..900038e 100644
--- a/MacTF.xcodeproj/project.pbxproj
+++ b/MacTF.xcodeproj/project.pbxproj
@@ -73,6 +73,20 @@
 		FFE483AC0CCDC38400DC3B45 /* TFUSBControllerCommunicationBlock.m in Sources */ = {isa = PBXBuildFile; fileRef = FFE483AA0CCDC38400DC3B45 /* TFUSBControllerCommunicationBlock.m */; };
 /* End PBXBuildFile section */
 
+/* Begin PBXBuildRule section */
+		FF6AEA8B0CD90329004AA180 /* PBXBuildRule */ = {
+			isa = PBXBuildRule;
+			compilerSpec = com.apple.compilers.proxy.script;
+			filePatterns = "*.strings";
+			fileType = pattern.proxy;
+			isEditable = 1;
+			outputFiles = (
+				"${DERIVED_FILES_DIR}/${INPUT_FILE_REGION_PATH_COMPONENT}${INPUT_FILE_NAME}",
+			);
+			script = "./to-UTF-16.pl \"${INPUT_FILE_REGION_PATH_COMPONENT}${INPUT_FILE_NAME}\" \"${DERIVED_FILES_DIR}/${INPUT_FILE_REGION_PATH_COMPONENT}${INPUT_FILE_NAME}\"";
+		};
+/* End PBXBuildRule section */
+
 /* Begin PBXContainerItemProxy section */
 		FFD5623A0CBF60430099621D /* PBXContainerItemProxy */ = {
 			isa = PBXContainerItemProxy;
@@ -178,6 +192,7 @@
 		C1E0460C0946D7E600B808CA /* mp3.tif */ = {isa = PBXFileReference; lastKnownFileType = image.tiff; path = mp3.tif; sourceTree = "<group>"; };
 		C1E0460E0946D7F300B808CA /* txt.tif */ = {isa = PBXFileReference; lastKnownFileType = image.tiff; path = txt.tif; sourceTree = "<group>"; };
 		FF10EEF20CC123890027156F /* COPYING */ = {isa = PBXFileReference; fileEncoding = 7; lastKnownFileType = text; path = COPYING; sourceTree = "<group>"; };
+		FF6AEABB0CD919E5004AA180 /* to-UTF-16.pl */ = {isa = PBXFileReference; explicitFileType = text.script.perl; fileEncoding = 4; path = "to-UTF-16.pl"; sourceTree = "<group>"; };
 		FFD562350CBF60420099621D /* iLifeControls.xcodeproj */ = {isa = PBXFileReference; lastKnownFileType = "wrapper.pb-project"; name = iLifeControls.xcodeproj; path = iLifeControls/iLifeControls.xcodeproj; sourceTree = "<group>"; };
 		FFE483A90CCDC38400DC3B45 /* TFUSBControllerCommunicationBlock.h */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.c.h; path = TFUSBControllerCommunicationBlock.h; sourceTree = "<group>"; };
 		FFE483AA0CCDC38400DC3B45 /* TFUSBControllerCommunicationBlock.m */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.c.objc; path = TFUSBControllerCommunicationBlock.m; sourceTree = "<group>"; };
@@ -255,6 +270,7 @@
 		29B97314FDCFA39411CA2CEA /* MacTF */ = {
 			isa = PBXGroup;
 			children = (
+				FF6AEABB0CD919E5004AA180 /* to-UTF-16.pl */,
 				FF10EEF20CC123890027156F /* COPYING */,
 				C16FADEB06C75BF100EE0B5E /* ToDo */,
 				C16FADEF06C75C0200EE0B5E /* ChangeLog */,
@@ -399,6 +415,7 @@
 				8D11072E0486CEB800E47090 /* Frameworks */,
 			);
 			buildRules = (
+				FF6AEA8B0CD90329004AA180 /* PBXBuildRule */,
 			);
 			dependencies = (
 				FFD562400CBF60790099621D /* PBXTargetDependency */,
diff --git a/to-UTF-16.pl b/to-UTF-16.pl
new file mode 100755
index 0000000..042e93a
--- /dev/null
+++ b/to-UTF-16.pl
@@ -0,0 +1,37 @@
+#! /usr/bin/perl
+
+use strict;
+use warnings;
+
+my($in_name, $out_name) = @ARGV;
+
+open my $in_h, "<", $in_name or die "$in_name: $!";
+binmode $in_h or die "$in_name: $!";
+my $head;
+defined(read $in_h, $head, 4) or die "$in_name: $!";
+close $in_h or die "$in_name: $!";
+
+my $charset;
+if ($head =~ /^\x00\x00\xFF\xFE/) {
+	print "$in_name: Found big-endian UTF-32 BOM\n";
+	$charset = "UTF-32";
+} elsif ($head =~ /^\xFE\xFF\x00\x00/) {
+	print "$in_name: Found little-endian UTF-32 BOM\n";
+	$charset = "UTF-32";
+} elsif ($head =~ /^\xFE\xFF/) {
+	print "$in_name: Found big-endian UTF-16 BOM\n";
+	$charset = "UTF-16";
+} elsif ($head =~ /^\xFF\xFE/) {
+	print "$in_name: Found little-endian UTF-16 BOM\n";
+	$charset = "UTF-16";
+} elsif ($head =~ /^\xEF\xBB\xBF/) {
+	print "$in_name: Found UTF-8 BOM\n";
+	$charset = "UTF-8";
+} else {
+	print "$in_name: No BOM found -- assuming UTF-8\n";
+	$charset = "UTF-8";
+}
+
+open STDOUT, ">", $out_name or die "$out_name: $!";
+exec "iconv", "-f", $charset, "-t", "UTF-16", $in_name
+	or die "exec iconv: $!";
-- 
2.11.4.GIT