From 1afcc914c5c7e45655ebe37646ed499de8d8d2fc Mon Sep 17 00:00:00 2001
From: Vladimir 'phcoder' Serbinenko <phcoder@gmail.com>
Date: Wed, 17 Nov 2010 16:13:16 +0100
Subject: [PATCH] 	Make better UTF compliant.

	* grub-core/normal/charset.c (grub_utf8_to_utf16): Handle 6- and 7-byte
	sequences as incorrect.
	(grub_is_valid_utf8): Likewise.
	(grub_utf8_to_ucs4): Likewise.
	(grub_ucs4_to_utf8): Handle codepoints outside of BMP.
	(grub_ucs4_to_utf8_alloc): Likewise.
	* include/grub/charset.h (grub_utf16_to_utf8): Likewise.
---
 ChangeLog                  | 12 ++++++++++
 grub-core/normal/charset.c | 45 +++++++++++---------------------------
 include/grub/charset.h     |  9 +++++++-
 3 files changed, 33 insertions(+), 33 deletions(-)

diff --git a/ChangeLog b/ChangeLog
index 3dc2e06da..e5863f6b4 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,15 @@
+2010-11-16  Vladimir Serbinenko  <phcoder@gmail.com>
+
+	Make better UTF compliant.
+
+	* grub-core/normal/charset.c (grub_utf8_to_utf16): Handle 6- and 7-byte
+	sequences as incorrect.
+	(grub_is_valid_utf8): Likewise.
+	(grub_utf8_to_ucs4): Likewise.
+	(grub_ucs4_to_utf8): Handle codepoints outside of BMP.
+	(grub_ucs4_to_utf8_alloc): Likewise.
+	* include/grub/charset.h (grub_utf16_to_utf8): Likewise.
+
 2010-11-16  Vladimir Serbinenko  <phcoder@gmail.com>
 
 	Make legacy_source behave like source.
diff --git a/grub-core/normal/charset.c b/grub-core/normal/charset.c
index b7f775c4f..85ead53c4 100644
--- a/grub-core/normal/charset.c
+++ b/grub-core/normal/charset.c
@@ -113,16 +113,6 @@ grub_utf8_to_utf16 (grub_uint16_t *dest, grub_size_t destsize,
 	      count = 3;
 	      code = c & GRUB_UINT8_3_TRAILINGBITS;
 	    }
-	  else if ((c & GRUB_UINT8_6_LEADINGBITS) == GRUB_UINT8_5_LEADINGBITS)
-	    {
-	      count = 4;
-	      code = c & GRUB_UINT8_2_TRAILINGBITS;
-	    }
-	  else if ((c & GRUB_UINT8_7_LEADINGBITS) == GRUB_UINT8_6_LEADINGBITS)
-	    {
-	      count = 5;
-	      code = c & GRUB_UINT8_1_TRAILINGBIT;
-	    }
 	  else
 	    return -1;
 	}
@@ -177,7 +167,7 @@ grub_ucs4_to_utf8 (grub_uint32_t *src, grub_size_t size,
 	  /* No surrogates in UCS-4... */
 	  *dest++ = '?';
 	}
-      else
+      else if (code < 0x10000)
 	{
 	  if (dest + 2 >= destend)
 	    break;
@@ -185,6 +175,15 @@ grub_ucs4_to_utf8 (grub_uint32_t *src, grub_size_t size,
 	  *dest++ = ((code >> 6) & 0x3F) | 0x80;
 	  *dest++ = (code & 0x3F) | 0x80;
 	}
+      else
+	{
+	  if (dest + 3 >= destend)
+	    break;
+	  *dest++ = (code >> 18) | 0xF0;
+	  *dest++ = ((code >> 12) & 0x3F) | 0x80;
+	  *dest++ = ((code >> 6) & 0x3F) | 0x80;
+	  *dest++ = (code & 0x3F) | 0x80;
+	}
     }
   *dest = 0;
 }
@@ -212,8 +211,10 @@ grub_ucs4_to_utf8_alloc (grub_uint32_t *src, grub_size_t size)
 	       || (code >= 0xD800 && code <= 0xDBFF))
 	/* No surrogates in UCS-4... */
 	cnt++;
-      else
+      else if (code < 0x10000)
 	cnt += 3;
+      else
+	cnt += 4;
     }
   cnt++;
 
@@ -273,16 +274,6 @@ grub_is_valid_utf8 (const grub_uint8_t *src, grub_size_t srcsize)
 	      count = 3;
 	      code = c & 0x07;
 	    }
-	  else if ((c & 0xfc) == 0xf8)
-	    {
-	      count = 4;
-	      code = c & 0x03;
-	    }
-	  else if ((c & 0xfe) == 0xfc)
-	    {
-	      count = 5;
-	      code = c & 0x01;
-	    }
 	  else
 	    return 0;
 	}
@@ -375,16 +366,6 @@ grub_utf8_to_ucs4 (grub_uint32_t *dest, grub_size_t destsize,
 	      count = 3;
 	      code = c & 0x07;
 	    }
-	  else if ((c & 0xfc) == 0xf8)
-	    {
-	      count = 4;
-	      code = c & 0x03;
-	    }
-	  else if ((c & 0xfe) == 0xfc)
-	    {
-	      count = 5;
-	      code = c & 0x01;
-	    }
 	  else
 	    {
 	      /* invalid */
diff --git a/include/grub/charset.h b/include/grub/charset.h
index 1d79d5d2c..c8247f78a 100644
--- a/include/grub/charset.h
+++ b/include/grub/charset.h
@@ -97,12 +97,19 @@ grub_utf16_to_utf8 (grub_uint8_t *dest, grub_uint16_t *src,
 	      /* Error... */
 	      *dest++ = '?';
 	    }
-	  else
+	  else if (code < 0x10000)
 	    {
 	      *dest++ = (code >> 12) | 0xE0;
 	      *dest++ = ((code >> 6) & 0x3F) | 0x80;
 	      *dest++ = (code & 0x3F) | 0x80;
 	    }
+	  else
+	    {
+	      *dest++ = (code >> 18) | 0xF0;
+	      *dest++ = ((code >> 12) & 0x3F) | 0x80;
+	      *dest++ = ((code >> 6) & 0x3F) | 0x80;
+	      *dest++ = (code & 0x3F) | 0x80;
+	    }
 	}
     }