Source for gnu.gcj.convert.IOConverter

   1: /* Copyright (C) 2000, 2001, 2005  Free Software Foundation
   2: 
   3:    This file is part of libgcj.
   4: 
   5: This software is copyrighted work licensed under the terms of the
   6: Libgcj License.  Please consult the file "LIBGCJ_LICENSE" for
   7: details.  */
   8: 
   9: /* This is a base class that handles aliasing issues for
  10:    UnicodeToBytes to BytesToUnicode.  */
  11: 
  12: package gnu.gcj.convert;
  13: 
  14: import java.util.Hashtable;
  15: 
  16: public abstract class IOConverter
  17: {
  18:   // Map encoding aliases to our canonical form.
  19:   static private Hashtable hash = new Hashtable ();
  20: 
  21:   // True if we have to do byte-order conversions on iconv()
  22:   // arguments.
  23:   static protected boolean iconv_byte_swap;
  24: 
  25:   static
  26:   {
  27:     // Manually maintained aliases.  Note that the value must be our
  28:     // canonical name.
  29:     hash.put ("iso-latin-1", "8859_1");
  30:     hash.put ("iso8859_1", "8859_1");
  31:     hash.put ("utf-16le", "UnicodeLittle");
  32:     hash.put ("utf-16be", "UnicodeBig");
  33:     // At least one build script out there uses 'utf8'.
  34:     hash.put ("utf8", "UTF8");
  35:     // On Solaris the default encoding, as returned by nl_langinfo(),
  36:     // is `646' (aka ASCII), but the Solaris iconv_open() doesn't
  37:     // understand that.  We work around the problem by adding an
  38:     // explicit alias for Solaris users.
  39:     hash.put ("646", "ASCII");
  40: 
  41:     // See PR 24552, PR 14358.
  42:     hash.put ("euc_jp", "EUCJIS");
  43:     hash.put ("eucjp", "EUCJIS");
  44: 
  45:     // All aliases after this point are automatically generated by the
  46:     // `encodings.pl' script.  Run it to make any corrections.
  47:     hash.put ("ansi_x3.4-1968", "ASCII");
  48:     hash.put ("ansi_x3.4-1986", "ASCII");
  49:     hash.put ("ascii", "ASCII");
  50:     hash.put ("cp367", "ASCII");
  51:     hash.put ("cp819", "8859_1");
  52:     hash.put ("csascii", "ASCII");
  53:     hash.put ("cseucpkdfmtjapanese", "EUCJIS");
  54:     hash.put ("csisolatin1", "8859_1");
  55:     hash.put ("csshiftjis", "SJIS");
  56:     hash.put ("euc-jp", "EUCJIS");
  57:     hash.put ("extended_unix_code_packed_format_for_japanese", "EUCJIS");
  58:     hash.put ("ibm367", "ASCII");
  59:     hash.put ("ibm819", "8859_1");
  60:     hash.put ("iso-8859-1", "8859_1");
  61:     hash.put ("iso-ir-100", "8859_1");
  62:     hash.put ("iso-ir-6", "ASCII");
  63:     hash.put ("iso646-us", "ASCII");
  64:     hash.put ("iso_646.irv:1991", "ASCII");
  65:     hash.put ("iso_8859-1", "8859_1");
  66:     hash.put ("iso_8859-1:1987", "8859_1");
  67:     hash.put ("l1", "8859_1");
  68:     hash.put ("latin1", "8859_1");
  69:     hash.put ("ms_kanji", "SJIS");
  70:     hash.put ("shift_jis", "SJIS");
  71:     hash.put ("us", "ASCII");
  72:     hash.put ("us-ascii", "ASCII");
  73:     hash.put ("utf-8", "UTF8");
  74:     hash.put ("utf16-be", "UnicodeBig");
  75:     hash.put ("utf16-le", "UnicodeLittle");
  76:     // End script-generated section.
  77: 
  78:     iconv_byte_swap = iconv_init ();
  79:   }
  80: 
  81:   private static native boolean iconv_init ();
  82: 
  83:   // Turn an alias into the canonical form.
  84:   protected static final String canonicalize (String name)
  85:   {
  86:     String c = (String) hash.get (name.toLowerCase ());
  87:     return c == null ? name : c;
  88:   }
  89: }