EBCDIC to ASCII CodePrinter

EBCDICToASCII_CodePrinter.java

import java.nio.charset.Charset;
import java.nio.charset.StandardCharsets;
import java.text.Normalizer;

public class EBCDICToASCII_CodePrinter
{

  public static void main(final String[] args)
    throws Exception
  {
    final String[] EBCDIC_CODE_PAGES = { "IBM037", "IBM01148", "IBM1047" };
    for (final String element: EBCDIC_CODE_PAGES)
    {
      final Charset charset = Charset.forName(element);
      System.out.println(String
        .format("private static final char[] %s_TO_ASCII_MAP = {",
                charset.name()));
      for (int i = 0; i < 256; i++)
      {
        // Convert EBCDIC character to a Java string
        final byte ebcdicChar = (byte) i;
        final String chstr = new String(new byte[] { ebcdicChar }, charset);
        final char normchstr = normalizeChar(ebcdicChar, charset);

        // Get character (code point) properties
        final int ch = chstr.codePointAt(0);
        final int numUTF8Bytes = chstr.getBytes(StandardCharsets.UTF_8).length;
        final String unicodeCharacterName = Character.getName(ch);
        final boolean isISOControl = Character
          .isISOControl(chstr.codePointAt(0));
        final boolean isASCIIPrintable = ch > 31 && ch < 127;

        // Sanity checks
        final boolean alternateASCIIPrintable = numUTF8Bytes == 1
                                                && !isISOControl;

        if (isASCIIPrintable && !alternateASCIIPrintable)
        {
          throw new Exception(String
            .format("%3d %d %b - We have not detected printable ASCII characters correctly",
                    ch,
                    numUTF8Bytes,
                    isISOControl));
        }

        // Print properties
        System.out.println(String
          .format("  '%s', // %s", normchstr, unicodeCharacterName));
      }
      System.out.println("};\n");
    }
  }

  private static char normalizeChar(final byte ebcdicChar,
                                    final Charset charset)
  {
    // Decode the EBCDIC character (which is a byte), based on the
    // EBCDIC charset
    final String decodedString = new String(new byte[] { ebcdicChar }, charset);
    // Normalize by removing accents and diacriticals
    final char normalizedChar = Normalizer
      .normalize(decodedString, Normalizer.Form.NFKD).charAt(0);
    // Check for characters within the ASCII printable range
    if (normalizedChar > 31 && normalizedChar < 127)
    {
      return normalizedChar;
    }
    else
    {
      return ' ';
    }
  }

}