Update JAVA encoding for UCS-4.

Author: bhaible

ChangeLog

@@ -1,3 +1,9 @@
+2002-05-26  Bruno Haible  <bruno@clisp.org>
+
+	* lib/java.h (java_mbtowc): Accept 12-byte sequences for non-BMP
+	characters.
+	(java_wctomb): Produce 12-byte sequences for non-BMP characters.
+
 2002-05-29  Bruno Haible  <bruno@clisp.org>
 
 	Fix installation of iconv program when linked with a libintl that was

lib/java.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (C) 1999-2000 Free Software Foundation, Inc.
+ * Copyright (C) 1999-2002 Free Software Foundation, Inc.
  * This file is part of the GNU LIBICONV Library.
  *
  * The GNU LIBICONV Library is free software; you can redistribute it
@@ -20,14 +20,16 @@
 
 /*
  * JAVA
- * This is ISO 8859-1 with \uXXXX escape sequences, denoting Unicode characters.
+ * This is ISO 8859-1 with \uXXXX escape sequences, denoting Unicode BMP
+ * characters. Consecutive pairs of \uXXXX escape sequences in the surrogate
+ * range, as in UTF-16, denote Unicode characters outside the BMP.
  */
 
 static int
 java_mbtowc (conv_t conv, ucs4_t *pwc, const unsigned char *s, int n)
 {
   unsigned char c;
-  ucs4_t wc;
+  ucs4_t wc, wc2;
   int i;
 
   c = s[0];
@@ -54,8 +56,39 @@ java_mbtowc (conv_t conv, ucs4_t *pwc, const unsigned char *s, int n)
       goto simply_backslash;
     wc |= (ucs4_t) c << (4 * (5-i));
   }
-  *pwc = wc;
-  return 6;
+  if (!(wc >= 0xd800 && wc < 0xe000)) {
+    *pwc = wc;
+    return 6;
+  }
+  if (wc >= 0xdc00)
+    goto simply_backslash;
+  if (n < 7)
+    return RET_TOOFEW(0);
+  if (s[6] != '\\')
+    goto simply_backslash;
+  if (n < 8)
+    return RET_TOOFEW(0);
+  if (s[7] != 'u')
+    goto simply_backslash;
+  wc2 = 0;
+  for (i = 8; i < 12; i++) {
+    if (n <= i)
+      return RET_TOOFEW(0);
+    c = s[i];
+    if (c >= '0' && c <= '9')
+      c -= '0';
+    else if (c >= 'A' && c <= 'Z')
+      c -= 'A'-10;
+    else if (c >= 'a' && c <= 'z')
+      c -= 'a'-10;
+    else
+      goto simply_backslash;
+    wc2 |= (ucs4_t) c << (4 * (11-i));
+  }
+  if (!(wc2 >= 0xdc00 && wc2 < 0xe000))
+    goto simply_backslash;
+  *pwc = 0x10000 + ((wc - 0xd800) << 10) + (wc2 - 0xdc00);
+  return 12;
 simply_backslash:
   *pwc = '\\';
   return 1;
@@ -67,7 +100,7 @@ java_wctomb (conv_t conv, unsigned char *r, ucs4_t wc, int n)
   if (wc < 0x80) {
     *r = wc;
     return 1;
-  } else {
+  } else if (wc < 0x10000) {
     if (n >= 6) {
       unsigned int i;
       r[0] = '\\';
@@ -79,5 +112,26 @@ java_wctomb (conv_t conv, unsigned char *r, ucs4_t wc, int n)
       return 6;
     } else
       return RET_TOOSMALL;
+  } else if (wc < 0x110000) {
+    if (n >= 12) {
+      ucs4_t wc1 = 0xd800 + ((wc - 0x10000) >> 10);
+      ucs4_t wc2 = 0xdc00 + ((wc - 0x10000) & 0x3ff);
+      unsigned int i;
+      r[0] = '\\';
+      r[1] = 'u';
+      i = (wc1 >> 12) & 0x0f; r[2] = (i < 10 ? '0'+i : 'a'-10+i);
+      i = (wc1 >> 8) & 0x0f;  r[3] = (i < 10 ? '0'+i : 'a'-10+i);
+      i = (wc1 >> 4) & 0x0f;  r[4] = (i < 10 ? '0'+i : 'a'-10+i);
+      i = wc1 & 0x0f;         r[5] = (i < 10 ? '0'+i : 'a'-10+i);
+      r[6] = '\\';
+      r[7] = 'u';
+      i = (wc2 >> 12) & 0x0f; r[8] = (i < 10 ? '0'+i : 'a'-10+i);
+      i = (wc2 >> 8) & 0x0f;  r[9] = (i < 10 ? '0'+i : 'a'-10+i);
+      i = (wc2 >> 4) & 0x0f; r[10] = (i < 10 ? '0'+i : 'a'-10+i);
+      i = wc2 & 0x0f;        r[11] = (i < 10 ? '0'+i : 'a'-10+i);
+      return 12;
+    } else
+      return RET_TOOSMALL;
   }
+  return RET_ILUNI;
 }