obtained gcc-4.6.4.tar.bz2 from upstream website;upstream

verified gcc-4.6.4.tar.bz2.sig;
imported gcc-4.6.4 source tree from verified upstream tarball.

downloading a git-generated archive based on the 'upstream' tag
should provide you with a source tree that is binary identical
to the one extracted from the above tarball.

if you have obtained the source via the command 'git clone',
however, do note that line-endings of files in your working
directory might differ from line-endings of the respective
files in the upstream repository.

1 files changed, 102 insertions, 0 deletions

diff --git a/libjava/gnu/gcj/convert/natOutput_EUCJIS.cc b/libjava/gnu/gcj/convert/natOutput_EUCJIS.cc
new file mode 100644
index 000000000..533fb9f1b
--- /dev/null
+++ b/libjava/gnu/gcj/convert/natOutput_EUCJIS.cc
@@ -0,0 +1,102 @@
+/* Copyright (C) 1999  Free Software Foundation
+
+   This file is part of libgcj.
+
+This software is copyrighted work licensed under the terms of the
+Libgcj License.  Please consult the file "LIBGCJ_LICENSE" for
+details.  */
+
+#include <config.h>
+#include <gcj/cni.h>
+#include <gnu/gcj/convert/Output_EUCJIS.h>
+
+/* A trie structure to map unicode values to JIS codes.
+ * code == -1: the character is undefined.
+ * code >= 0 && code < 128:  JIS-Roman - mostly Ascii.
+ * code >= 128 && code < 256:  Half-width Katakana.
+ * code >= 256 && code < 0x8000:  JIS X 0208:1997.
+ * code >= 0x8000 && code < 0xFFFF:  JIX X 0212-1990.
+ */
+
+extern unsigned short Unicode_to_JIS[];
+
+int
+trie_lookup (unsigned short *trie, unsigned short key)
+{
+  unsigned short branch = trie[(key >> 12) & 0xf];
+  if (branch == 0)
+    return -1;
+  branch = trie[branch + ((key >> 8) & 0xf)];
+  if (branch == 0)
+    return -1;
+  branch = trie[branch + ((key >> 4) & 0xf)];
+  if (branch == 0)
+    return -1;
+  return trie[branch + (key & 0xf)];
+}
+
+static jint
+convert_TO_EUCJIS (gnu::gcj::convert::Output_EUCJIS *encoder,
+			  jchar *ptr, jint inlength)
+{
+  int orig_inlength = inlength;
+  jint outbuf_length = encoder->buf->length;
+  for (;;)
+    {
+      if (encoder->count >= outbuf_length)
+	break;
+      if (encoder->pending1 >= 0)
+	{
+	  elements(encoder->buf)[encoder->count++] = encoder->pending1;
+	  encoder->pending1 = encoder->pending2;
+	  encoder->pending2 = -1;
+	  continue;
+	}
+      if (inlength == 0)
+	break;
+      jchar ch = *ptr++;
+      inlength--;
+      unsigned short val = trie_lookup(Unicode_to_JIS, ch);
+      if (val < 0x80)
+	{
+	  if (val == 0xffff)
+	    val = '?';
+	}
+      else if (val <= 0xFF)
+	{
+	  encoder->pending1 = val;
+	  encoder->pending2 = -1;
+	  val = 0x8e;
+	}
+      else if (val < 0x8000)
+	{
+	  val |= 0x8080;
+	  encoder->pending1 = val & 0xff;
+	  val = val >> 8;
+	  encoder->pending2 = -1;
+	}
+      else
+	{
+	  val |= 0x8080;
+	  encoder->pending1 = val >> 8;
+	  encoder->pending2 = val & 0xff;
+	  val = 0x8f;
+	}
+      elements(encoder->buf)[encoder->count++] = val;
+    }
+  return orig_inlength - inlength;
+}
+
+jint
+gnu::gcj::convert::Output_EUCJIS::write (jcharArray inbuffer,
+					 jint inpos, jint inlength)
+{
+  return convert_TO_EUCJIS(this, &elements(inbuffer)[inpos], inlength);
+}
+
+jint
+gnu::gcj::convert::Output_EUCJIS::write (jstring str, jint inpos,
+					 jint inlength, jcharArray)
+{
+  return convert_TO_EUCJIS(this, _Jv_GetStringChars(str)+inpos, inlength);
+}