From ec6f4f36ec679a9434c71915f22c45b165cd227a Mon Sep 17 00:00:00 2001
From: Linus Torvalds <torvalds@linux-foundation.org>
Date: Tue, 10 Jul 2012 17:36:30 -0700
Subject: [PATCH] Use utf8 helper functions for keyboard input

ttgetc() used some homebrew utf8 to unicode translation, limited to just
the normal latin1 characters.  Use the utf8 helper functions to get it
right for the more complex cases.

NOTE! We don't actually handle characters > 0xff right anyway.  And we
still end up doing Latin1 in the buffers on input.  One small step at a
time.

Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 posix.c | 68 +++++++++++++++------------------------------------------
 1 file changed, 18 insertions(+), 50 deletions(-)

diff --git a/posix.c b/posix.c
index 445724e..e6ebc83 100644
--- a/posix.c
+++ b/posix.c
@@ -150,10 +150,10 @@ void ttflush(void)
  */
 int ttgetc(void)
 {
-	static unsigned char buffer[32];
+	static char buffer[32];
 	static int pending;
-	unsigned char c, second;
-	int count;
+	unicode_t c;
+	int count, bytes = 1;
 
 	count = pending;
 	if (!count) {
@@ -163,7 +163,7 @@ int ttgetc(void)
 		pending = count;
 	}
 
-	c = buffer[0];
+	c = (unsigned char) buffer[0];
 	if (c >= 32 && c < 128)
 		goto done;
 
@@ -181,56 +181,24 @@ int ttgetc(void)
 		ntermios.c_cc[VTIME] = 0;
 		tcsetattr(0, TCSANOW, &ntermios);
 
-		if (n <= 0)
+		if (n > 0)
+			pending += n;
+	}
+	if (pending > 1) {
+		unsigned char second = buffer[1];
+
+		/* Turn ESC+'[' into CSI */
+		if (c == 27 && second == '[') {
+			bytes = 2;
+			c = 128+27;
 			goto done;
-		pending += n;
+		}
 	}
-	second = buffer[1];
-
-	/* Turn ESC+'[' into CSI */
-	if (c == 27 && second == '[') {
-		pending -= 2;
-		memmove(buffer, buffer+2, pending);
-		return 128+27;
-	}
-
-	/* Normal 7-bit? */
-	if (!(c & 0x80))
-		goto done;
-
-	/*
-	 * Unexpected UTF-8 continuation character? Maybe
-	 * we're in non-UTF mode, or maybe it's a control
-	 * character.. Regardless, just pass it on.
-	 */
-	if (!(c & 0x40))
-		goto done;
-
-	/*
-	 * Multi-byte sequences.. Right now we only
-	 * want to get characters that can be represented
-	 * in a single byte, so we're not interested in
-	 * anything else..
-	 */
-	if (c & 0x3c)
-		goto done;
-
-	if ((second & 0xc0) != 0x80)
-		goto done;
-
-	/*
-	 * Ok, it's a two-byte UTF-8 character that can be represented
-	 * as a single-byte Latin1 character!
-	 */
-	c = (c << 6) | (second & 0x3f);
-	pending -= 2;
-	memmove(buffer, buffer+2, pending);
-
-	return c;
+	bytes = utf8_to_unicode(buffer, 0, pending, &c);
 
 done:
-	pending--;
-	memmove(buffer, buffer+1, pending);
+	pending -= bytes;
+	memmove(buffer, buffer+bytes, pending);
 	return c;
 }