*** empty log message ***

1 files changed, 103 insertions, 43 deletions

diff --git a/src/utf8.c b/src/utf8.c
index 09066db..9d04d19 100644
--- a/src/utf8.c
+++ b/src/utf8.c
@@ -10,6 +10,9 @@ static char rcsid[] = "$Id$";
 
 /*
  * $Log$
+ * Revision 1.3  2008/02/23 11:48:37  james
+ * *** empty log message ***
+ *
  * Revision 1.2  2008/02/22 23:39:27  james
  * *** empty log message ***
  *
@@ -21,61 +24,118 @@ static char rcsid[] = "$Id$";
 #include "project.h"
 
 
-void utf8_flush(Context *c)
+void
+utf8_flush (Context * c)
 {
-UTF8 *u=c->u;
-int i;
+  UTF8 *u = c->u;
+  int i;
 
-for (i=0;i<u->utf_ptr;++i) 
-	vt102_parse(c,u->utf_buf[i]);
+  switch (u->utf_ptr)
+    {
+    case 1:
+      log_f (c->l, "%s:%d invalid utf-8 sequence: \\%03o",
+             __FILE__, __LINE__, u->utf_buf[0]);
+      break;
+    case 2:
+      log_f (c->l, "%s:%d invalid utf-8 sequence: \\%03o \\%03o",
+             __FILE__, __LINE__, u->utf_buf[0], u->utf_buf[1]);
+      break;
+    case 3:
+      log_f (c->l, "%s:%d invalid utf-8 sequence: \\%03o \\%03o \\%03o",
+             __FILE__, __LINE__, u->utf_buf[0], u->utf_buf[1], u->utf_buf[2]);
+      break;
+    case 4:
+      log_f (c->l,
+             "%s:%d invalid utf-8 sequence: \\%03o \\%03o \\%03o \\%03o",
+             __FILE__, __LINE__,
+             u->utf_buf[0], u->utf_buf[1], u->utf_buf[2], u->utf_buf[3]);
+      break;
+    }
 
-u->utf_ptr=0;
-u->in_utf8=0;
+  for (i = 0; i < u->utf_ptr; ++i)
+    vt102_parse_char (c, u->utf_buf[i]);
+
+  u->utf_ptr = 0;
+  u->in_utf8 = 0;
 }
 
-int utf8_parse(Context *c,int ch)
+void
+utf8_parse (Context * c, int ch)
 {
-UTF8 *u=&c->u;
-  
-  if (!u->in_utf8) {
-	  /*FIXME: for the moment we bodge utf8 support*/
-	  if (ch==0xb9) { /*CSI, not a valid utf8 start char*/
-		vt102_parse(c,ch);
-	  } else if ((ch & 0xe0) == 0xc0) { /*Start of two byte unicode sequence*/
-		u->in_utf8=2;
-		u->utf_ptr=0;
-		u->utf_buf[u->utf_ptr++]=ch;
-	  } else if ((ch & 0xf0) ==0xe0) { /*Start of three byte unicode sequence*/
-		u->in_utf8=3;
-		u->utf_ptr=0;
-		u->utf_buf[u->utf_ptr++]=ch;
-	  } else if ((ch & 0xf8) ==0xf0) {
-		u->in_utf8=4;
-		u->utf_ptr=0;
-		u->utf_buf[u->utf_ptr++]=ch;
-	  } else {
-		vt102_parse(c,ch);
-	  } 
-  } else {
-	if ((ch & 0xc0) != 0x80) {
-		utf8_flush(c);
-		vt102_parse(c,ch);
-	} else {
-		u->utf_buf[u->utf_ptr++]=ch;
-		u->in_utf8--;
-	}
-  }
+  UTF8 *u = c->u;
+
+  if (ch == SYM_CHAR_RESET)
+    {
+      u->in_utf8 = 0;
+      vt102_parse_char (c, ch);
+      return;
+    }
+
+  if (!u->in_utf8)
+    {
+      /*FIXME: for the moment we bodge utf8 support */
+      if (ch == 0xb9)
+        {                       /*CSI, not a valid utf8 start char */
+          vt102_parse_char (c, ch);
+        }
+      else if ((ch & 0xe0) == 0xc0)
+        {                       /*Start of two byte unicode sequence */
+          u->in_utf8 = 1;
+          u->utf_ptr = 0;
+          u->utf_buf[u->utf_ptr++] = ch;
+          u->ch = (ch & 0x1f) << 6;
+          u->sh = 0;
+        }
+      else if ((ch & 0xf0) == 0xe0)
+        {                       /*Start of three byte unicode sequence */
+          u->in_utf8 = 2;
+          u->utf_ptr = 0;
+          u->utf_buf[u->utf_ptr++] = ch;
+          u->ch = (ch & 0x0f) << 12;
+          u->sh = 6;
+        }
+      else if ((ch & 0xf8) == 0xf0)
+        {
+          u->in_utf8 = 3;
+          u->utf_ptr = 0;
+          u->utf_buf[u->utf_ptr++] = ch;
+          u->ch = (ch & 0x07) << 18;
+          u->sh = 12;
+        }
+      else
+        {
+          vt102_parse_char (c, ch);
+        }
+    }
+  else
+    {
+      if ((ch & 0xc0) != 0x80)
+        {
+          utf8_flush (c);
+          vt102_parse_char (c, ch);
+        }
+      else
+        {
+          u->utf_buf[u->utf_ptr++] = ch;
+          u->ch |= (ch & 0x3f) << u->sh;
+          u->sh -= 6;
+          u->in_utf8--;
+
+          if (!u->in_utf8)
+            vt102_parse_char (c, ch);
+        }
+    }
 }
 
 
 
-UTF8 *utf8_new(void)
+UTF8 *
+utf8_new (void)
 {
-UTF8 *ret;
+  UTF8 *ret;
 
-ret=(UTF8 *) malloc(sizeof(UTF8));
+  ret = (UTF8 *) malloc (sizeof (UTF8));
 
-ret->in_utf8=0;
+  ret->in_utf8 = 0;
 
 }
-