4116222: Errors in Arabic code-conversion tables, part II

Summary: updated the IBM420 datatable
Reviewed-by: alanb

make/tools/CharsetMapping/IBM420.c2b

-0x15	U+0085
+#
+#  The diff of 01A434B0.TXMAP110 and 34B001A4.RXMAP110
+#
+#  Added: 0x15    U+0085
+#
+0x15    U+0085
+0x42    U+FE7C
+0x46    U+FE80
+0x47    U+FE81
+0x49    U+FE83
+0x4B    U+066C
+0x4B    U+FF0E
+0x4C    U+FF1C
+0x4D    U+FF08
+0x4E    U+FF0B
+0x4F    U+FF5C
+0x50    U+FF06
+0x52    U+FE85
+0x52    U+FE86
+0x55    U+FE89
+0x55    U+FE8A
+0x55    U+FE8B
+0x55    U+FE8C
+0x56    U+0625
+0x56    U+FE87
+0x56    U+FE8D
+0x57    U+FE88
+0x58    U+FE8F
+0x58    U+FE90
+0x59    U+FE92
+0x5A    U+FF01
+0x5B    U+FF04
+0x5C    U+066D
+0x5C    U+FF0A
+0x5D    U+FF09
+0x5E    U+FF1B
+0x60    U+FF0D
+0x61    U+FF0F
+0x62    U+FE93
+0x62    U+FE94
+0x63    U+FE95
+0x63    U+FE96
+0x64    U+FE98
+0x65    U+FE99
+0x65    U+FE9A
+0x66    U+FE9C
+0x67    U+FE9D
+0x67    U+FE9E
+0x68    U+FEA0
+0x69    U+FEA1
+0x69    U+FEA2
+0x6B    U+066B
+0x6B    U+FF0C
+0x6C    U+066A
+0x6C    U+FF05
+0x6D    U+FF3F
+0x6E    U+FF1E
+0x6F    U+FF1F
+0x70    U+FEA4
+0x71    U+FEA5
+0x71    U+FEA6
+0x72    U+FEA8
+0x73    U+FEA9
+0x73    U+FEAA
+0x74    U+FEAB
+0x74    U+FEAC
+0x75    U+FEAD
+0x75    U+FEAE
+0x76    U+FEAF
+0x76    U+FEB0
+0x77    U+FEB1
+0x77    U+FEB2
+0x78    U+FEB4
+0x7A    U+FF1A
+0x7B    U+FF03
+0x7C    U+FF20
+0x7D    U+FF07
+0x7E    U+FF1D
+0x7F    U+FF02
+0x80    U+FEB5
+0x80    U+FEB6
+0x81    U+FF41
+0x82    U+FF42
+0x83    U+FF43
+0x84    U+FF44
+0x85    U+FF45
+0x86    U+FF46
+0x87    U+FF47
+0x88    U+FF48
+0x89    U+FF49
+0x8A    U+FEB8
+0x8B    U+FEB9
+0x8B    U+FEBA
+0x8C    U+FEBC
+0x8D    U+FEBD
+0x8D    U+FEBE
+0x8E    U+FEC0
+0x8F    U+FEC1
+0x8F    U+FEC2
+0x8F    U+FEC3
+0x8F    U+FEC4
+0x90    U+FEC5
+0x90    U+FEC6
+0x90    U+FEC7
+0x90    U+FEC8
+0x91    U+FF4A
+0x92    U+FF4B
+0x93    U+FF4C
+0x94    U+FF4D
+0x95    U+FF4E
+0x96    U+FF4F
+0x97    U+FF50
+0x98    U+FF51
+0x99    U+FF52
+0x9A    U+FEC9
+0x9E    U+FECD
+0xA2    U+FF53
+0xA3    U+FF54
+0xA4    U+FF55
+0xA5    U+FF56
+0xA6    U+FF57
+0xA7    U+FF58
+0xA8    U+FF59
+0xA9    U+FF5A
+0xAB    U+FED1
+0xAB    U+FED2
+0xAC    U+FED4
+0xAD    U+FED5
+0xAD    U+FED6
+0xAE    U+FED8
+0xAF    U+FED9
+0xAF    U+FEDA
+0xB0    U+FEDC
+0xB1    U+FEDD
+0xB1    U+FEDE
+0xB8    U+FEF9
+0xB9    U+FEFA
+0xBA    U+FEE0
+0xBB    U+FEE1
+0xBB    U+FEE2
+0xBC    U+FEE4
+0xBD    U+FEE5
+0xBD    U+FEE6
+0xBE    U+FEE8
+0xBF    U+FEE9
+0xBF    U+FEEA
+0xC1    U+FF21
+0xC2    U+FF22
+0xC3    U+FF23
+0xC4    U+FF24
+0xC5    U+FF25
+0xC6    U+FF26
+0xC7    U+FF27
+0xC8    U+FF28
+0xC9    U+FF29
+0xCF    U+FEED
+0xCF    U+FEEE
+0xD1    U+FF2A
+0xD2    U+FF2B
+0xD3    U+FF2C
+0xD4    U+FF2D
+0xD5    U+FF2E
+0xD6    U+FF2F
+0xD7    U+FF30
+0xD8    U+FF31
+0xD9    U+FF32
+0xDA    U+FEEF
+0xDC    U+FEF1
+0xDE    U+FEF4
+0xE2    U+FF33
+0xE3    U+FF34
+0xE4    U+FF35
+0xE5    U+FF36
+0xE6    U+FF37
+0xE7    U+FF38
+0xE8    U+FF39
+0xE9    U+FF3A
+0xF0    U+FF10
+0xF1    U+FF11
+0xF2    U+FF12
+0xF3    U+FF13
+0xF4    U+FF14
+0xF5    U+FF15
+0xF6    U+FF16
+0xF7    U+FF17
+0xF8    U+FF18
+0xF9    U+FF19

make/tools/CharsetMapping/IBM420.map

-#Generated from IBM420.java
-0x00	U+0000
-0x01	U+0001
-0x02	U+0002
-0x03	U+0003
-0x04	U+009c
-0x05	U+0009
-0x06	U+0086
-0x07	U+007f
-0x08	U+0097
-0x09	U+008d
-0x0a	U+008e
-0x0b	U+000b
-0x0c	U+000c
-0x0d	U+000d
-0x0e	U+000e
-0x0f	U+000f
-0x10	U+0010
-0x11	U+0011
-0x12	U+0012
-0x13	U+0013
-0x14	U+009d
-0x15	U+000a
-0x16	U+0008
-0x17	U+0087
-0x18	U+0018
-0x19	U+0019
-0x1a	U+0092
-0x1b	U+008f
-0x1c	U+001c
-0x1d	U+001d
-0x1e	U+001e
-0x1f	U+001f
-0x20	U+0080
-0x21	U+0081
-0x22	U+0082
-0x23	U+0083
-0x24	U+0084
-0x25	U+000a
-0x26	U+0017
-0x27	U+001b
-0x28	U+0088
-0x29	U+0089
-0x2a	U+008a
-0x2b	U+008b
-0x2c	U+008c
-0x2d	U+0005
-0x2e	U+0006
-0x2f	U+0007
-0x30	U+0090
-0x31	U+0091
-0x32	U+0016
-0x33	U+0093
-0x34	U+0094
-0x35	U+0095
-0x36	U+0096
-0x37	U+0004
-0x38	U+0098
-0x39	U+0099
-0x3a	U+009a
-0x3b	U+009b
-0x3c	U+0014
-0x3d	U+0015
-0x3e	U+009e
-0x3f	U+001a
-0x40	U+0020
-0x41	U+00a0
-0x42	U+fe7c
-0x43	U+fe7d
-0x44	U+0640
-0x45	U+f8fc
-0x46	U+fe80
-0x47	U+fe81
-0x48	U+fe82
-0x49	U+fe83
-0x4a	U+00a2
-0x4b	U+002e
-0x4c	U+003c
-0x4d	U+0028
-0x4e	U+002b
-0x4f	U+007c
-0x50	U+0026
-0x51	U+fe84
-0x52	U+fe85
-0x53	U+fffd
-0x54	U+fffd
-0x55	U+fe8b
-0x56	U+fe8d
-0x57	U+fe8e
-0x58	U+fe8f
-0x59	U+fe91
-0x5a	U+0021
-0x5b	U+0024
-0x5c	U+002a
-0x5d	U+0029
-0x5e	U+003b
-0x5f	U+00ac
-0x60	U+002d
-0x61	U+002f
-0x62	U+fe93
-0x63	U+fe95
-0x64	U+fe97
-0x65	U+fe99
-0x66	U+fe9b
-0x67	U+fe9d
-0x68	U+fe9f
-0x69	U+fea1
-0x6a	U+00a6
-0x6b	U+002c
-0x6c	U+0025
-0x6d	U+005f
-0x6e	U+003e
-0x6f	U+003f
-0x70	U+fea3
-0x71	U+fea5
-0x72	U+fea7
-0x73	U+fea9
-0x74	U+feab
-0x75	U+fead
-0x76	U+feaf
-0x77	U+f8f6
-0x78	U+feb3
-0x79	U+060c
-0x7a	U+003a
-0x7b	U+0023
-0x7c	U+0040
-0x7d	U+0027
-0x7e	U+003d
-0x7f	U+0022
-0x80	U+f8f5
-0x81	U+0061
-0x82	U+0062
-0x83	U+0063
-0x84	U+0064
-0x85	U+0065
-0x86	U+0066
-0x87	U+0067
-0x88	U+0068
-0x89	U+0069
-0x8a	U+feb7
-0x8b	U+f8f4
-0x8c	U+febb
-0x8d	U+f8f7
-0x8e	U+febf
-0x8f	U+fec3
-0x90	U+fec7
-0x91	U+006a
-0x92	U+006b
-0x93	U+006c
-0x94	U+006d
-0x95	U+006e
-0x96	U+006f
-0x97	U+0070
-0x98	U+0071
-0x99	U+0072
-0x9a	U+fec9
-0x9b	U+feca
-0x9c	U+fecb
-0x9d	U+fecc
-0x9e	U+fecd
-0x9f	U+fece
-0xa0	U+fecf
-0xa1	U+00f7
-0xa2	U+0073
-0xa3	U+0074
-0xa4	U+0075
-0xa5	U+0076
-0xa6	U+0077
-0xa7	U+0078
-0xa8	U+0079
-0xa9	U+007a
-0xaa	U+fed0
-0xab	U+fed1
-0xac	U+fed3
-0xad	U+fed5
-0xae	U+fed7
-0xaf	U+fed9
-0xb0	U+fedb
-0xb1	U+fedd
-0xb2	U+fef5
-0xb3	U+fef6
-0xb4	U+fef7
-0xb5	U+fef8
-0xb6	U+fffd
-0xb7	U+fffd
-0xb8	U+fefb
-0xb9	U+fefc
-0xba	U+fedf
-0xbb	U+fee1
-0xbc	U+fee3
-0xbd	U+fee5
-0xbe	U+fee7
-0xbf	U+fee9
-0xc0	U+061b
-0xc1	U+0041
-0xc2	U+0042
-0xc3	U+0043
-0xc4	U+0044
-0xc5	U+0045
-0xc6	U+0046
-0xc7	U+0047
-0xc8	U+0048
-0xc9	U+0049
-0xca	U+00ad
-0xcb	U+feeb
-0xcc	U+fffd
-0xcd	U+feec
-0xce	U+fffd
-0xcf	U+feed
-0xd0	U+061f
-0xd1	U+004a
-0xd2	U+004b
-0xd3	U+004c
-0xd4	U+004d
-0xd5	U+004e
-0xd6	U+004f
-0xd7	U+0050
-0xd8	U+0051
-0xd9	U+0052
-0xda	U+feef
-0xdb	U+fef0
-0xdc	U+fef1
-0xdd	U+fef2
-0xde	U+fef3
-0xdf	U+0660
-0xe0	U+00d7
-0xe1	U+2007
-0xe2	U+0053
-0xe3	U+0054
-0xe4	U+0055
-0xe5	U+0056
-0xe6	U+0057
-0xe7	U+0058
-0xe8	U+0059
-0xe9	U+005a
-0xea	U+0661
-0xeb	U+0662
-0xec	U+fffd
-0xed	U+0663
-0xee	U+0664
-0xef	U+0665
-0xf0	U+0030
-0xf1	U+0031
-0xf2	U+0032
-0xf3	U+0033
-0xf4	U+0034
-0xf5	U+0035
-0xf6	U+0036
-0xf7	U+0037
-0xf8	U+0038
-0xf9	U+0039
-0xfa	U+fffd
-0xfb	U+0666
-0xfc	U+0667
-0xfd	U+0668
-0xfe	U+0669
-0xff	U+009f
+#
+#  Frm IBMCDC datatable 01A434B0.TXMAP110
+#
+#  Changed
+#  0x15    U+0085  ->  0x15    U+000a
+#
+0x00    U+0000
+0x01    U+0001
+0x02    U+0002
+0x03    U+0003
+0x04    U+009C
+0x05    U+0009
+0x06    U+0086
+0x07    U+007F
+0x08    U+0097
+0x09    U+008D
+0x0A    U+008E
+0x0B    U+000B
+0x0C    U+000C
+0x0D    U+000D
+0x0E    U+000E
+0x0F    U+000F
+0x10    U+0010
+0x11    U+0011
+0x12    U+0012
+0x13    U+0013
+0x14    U+009D
+0x15    U+000A
+0x16    U+0008
+0x17    U+0087
+0x18    U+0018
+0x19    U+0019
+0x1A    U+0092
+0x1B    U+008F
+0x1C    U+001C
+0x1D    U+001D
+0x1E    U+001E
+0x1F    U+001F
+0x20    U+0080
+0x21    U+0081
+0x22    U+0082
+0x23    U+0083
+0x24    U+0084
+0x25    U+000A
+0x26    U+0017
+0x27    U+001B
+0x28    U+0088
+0x29    U+0089
+0x2A    U+008A
+0x2B    U+008B
+0x2C    U+008C
+0x2D    U+0005
+0x2E    U+0006
+0x2F    U+0007
+0x30    U+0090
+0x31    U+0091
+0x32    U+0016
+0x33    U+0093
+0x34    U+0094
+0x35    U+0095
+0x36    U+0096
+0x37    U+0004
+0x38    U+0098
+0x39    U+0099
+0x3A    U+009A
+0x3B    U+009B
+0x3C    U+0014
+0x3D    U+0015
+0x3E    U+009E
+0x3F    U+001A
+0x40    U+0020
+0x41    U+00A0
+0x42    U+0651
+0x43    U+FE7D
+0x44    U+0640
+0x45    U+200B
+0x46    U+0621
+0x47    U+0622
+0x48    U+FE82
+0x49    U+0623
+0x4A    U+00A2
+0x4B    U+002E
+0x4C    U+003C
+0x4D    U+0028
+0x4E    U+002B
+0x4F    U+007C
+0x50    U+0026
+0x51    U+FE84
+0x52    U+0624
+0x55    U+0626
+0x56    U+0627
+0x57    U+FE8E
+0x58    U+0628
+0x59    U+FE91
+0x5A    U+0021
+0x5B    U+0024
+0x5C    U+002A
+0x5D    U+0029
+0x5E    U+003B
+0x5F    U+00AC
+0x60    U+002D
+0x61    U+002F
+0x62    U+0629
+0x63    U+062A
+0x64    U+FE97
+0x65    U+062B
+0x66    U+FE9B
+0x67    U+062C
+0x68    U+FE9F
+0x69    U+062D
+0x6A    U+00A6
+0x6B    U+002C
+0x6C    U+0025
+0x6D    U+005F
+0x6E    U+003E
+0x6F    U+003F
+0x70    U+FEA3
+0x71    U+062E
+0x72    U+FEA7
+0x73    U+062F
+0x74    U+0630
+0x75    U+0631
+0x76    U+0632
+0x77    U+0633
+0x78    U+FEB3
+0x79    U+060C
+0x7A    U+003A
+0x7B    U+0023
+0x7C    U+0040
+0x7D    U+0027
+0x7E    U+003D
+0x7F    U+0022
+0x80    U+0634
+0x81    U+0061
+0x82    U+0062
+0x83    U+0063
+0x84    U+0064
+0x85    U+0065
+0x86    U+0066
+0x87    U+0067
+0x88    U+0068
+0x89    U+0069
+0x8A    U+FEB7
+0x8B    U+0635
+0x8C    U+FEBB
+0x8D    U+0636
+0x8E    U+FEBF
+0x8F    U+0637
+0x90    U+0638
+0x91    U+006A
+0x92    U+006B
+0x93    U+006C
+0x94    U+006D
+0x95    U+006E
+0x96    U+006F
+0x97    U+0070
+0x98    U+0071
+0x99    U+0072
+0x9A    U+0639
+0x9B    U+FECA
+0x9C    U+FECB
+0x9D    U+FECC
+0x9E    U+063A
+0x9F    U+FECE
+0xA0    U+FECF
+0xA1    U+00F7
+0xA2    U+0073
+0xA3    U+0074
+0xA4    U+0075
+0xA5    U+0076
+0xA6    U+0077
+0xA7    U+0078
+0xA8    U+0079
+0xA9    U+007A
+0xAA    U+FED0
+0xAB    U+0641
+0xAC    U+FED3
+0xAD    U+0642
+0xAE    U+FED7
+0xAF    U+0643
+0xB0    U+FEDB
+0xB1    U+0644
+0xB2    U+FEF5
+0xB3    U+FEF6
+0xB4    U+FEF7
+0xB5    U+FEF8
+0xB8    U+FEFB
+0xB9    U+FEFC
+0xBA    U+FEDF
+0xBB    U+0645
+0xBC    U+FEE3
+0xBD    U+0646
+0xBE    U+FEE7
+0xBF    U+0647
+0xC0    U+061B
+0xC1    U+0041
+0xC2    U+0042
+0xC3    U+0043
+0xC4    U+0044
+0xC5    U+0045
+0xC6    U+0046
+0xC7    U+0047
+0xC8    U+0048
+0xC9    U+0049
+0xCA    U+00AD
+0xCB    U+FEEB
+0xCD    U+FEEC
+0xCF    U+0648
+0xD0    U+061F
+0xD1    U+004A
+0xD2    U+004B
+0xD3    U+004C
+0xD4    U+004D
+0xD5    U+004E
+0xD6    U+004F
+0xD7    U+0050
+0xD8    U+0051
+0xD9    U+0052
+0xDA    U+0649
+0xDB    U+FEF0
+0xDC    U+064A
+0xDD    U+FEF2
+0xDE    U+FEF3
+0xDF    U+0660
+0xE0    U+00D7
+0xE2    U+0053
+0xE3    U+0054
+0xE4    U+0055
+0xE5    U+0056
+0xE6    U+0057
+0xE7    U+0058
+0xE8    U+0059
+0xE9    U+005A
+0xEA    U+0661
+0xEB    U+0662
+0xED    U+0663
+0xEE    U+0664
+0xEF    U+0665
+0xF0    U+0030
+0xF1    U+0031
+0xF2    U+0032
+0xF3    U+0033
+0xF4    U+0034
+0xF5    U+0035
+0xF6    U+0036
+0xF7    U+0037
+0xF8    U+0038
+0xF9    U+0039
+0xFB    U+0666
+0xFC    U+0667
+0xFD    U+0668
+0xFE    U+0669
+0xFF    U+009F

make/tools/CharsetMapping/IBM420.nr

-0x25	U+000a
+0x25    U+000a

make/tools/src/build/tools/charsetmapping/GenerateSBCS.java

@@ -26,6 +26,7 @@
 package build.tools.charsetmapping;
 
 import java.io.*;
+import java.util.Arrays;
 import java.util.ArrayList;
 import java.util.Scanner;
 import java.util.Formatter;
@@ -54,33 +55,19 @@ public class GenerateSBCS {
             String pkgName  = fields[4];
             System.out.printf("%s,%s,%s,%b,%s%n", clzName, csName, hisName, isASCII, pkgName);
 
-            StringBuilder b2c = new StringBuilder();
-            int c2bLen = genB2C(
-                new FileInputStream(new File(args[0], clzName+".map")), b2c);
-
-            String b2cNR = null;
-            File nrF = new File(args[0], clzName+".nr");
-            if (nrF.exists()) {
-                b2cNR = genNR(new FileInputStream(nrF));
-            }
-
-            String c2bNR = null;
-            File c2bF = new File(args[0], clzName+".c2b");
-            if (c2bF.exists()) {
-                c2bNR = genC2BNR(new FileInputStream(c2bF));
-            }
-
-            genSBCSClass(args[0], args[1], "SingleByte-X.java",
-                         clzName, csName, hisName, pkgName, isASCII,
-                         b2c.toString(), b2cNR, c2bNR, c2bLen);
+            genClass(args[0], args[1], "SingleByte-X.java",
+                     clzName, csName, hisName, pkgName, isASCII);
         }
     }
 
     private static void toString(char[] sb, int off, int end,
-                                 Formatter out, String closure) {
+                                 Formatter out, String closure,
+                                 boolean comment) {
         while (off < end) {
             out.format("        \"");
             for (int j = 0; j < 8; j++) {
+                if (off == end)
+                    break;
                 char c = sb[off++];
                 switch (c) {
                 case '\b':
@@ -103,101 +90,124 @@ public class GenerateSBCS {
                     out.format("\\u%04X", c & 0xffff);
                 }
             }
-            if (off == end)
-               out.format("\" %s      // 0x%02x - 0x%02x%n", closure, off-8, off-1);
-            else
-               out.format("\" +      // 0x%02x - 0x%02x%n", off-8, off-1);
+            if (comment) {
+                if (off == end)
+                    out.format("\" %s      // 0x%02x - 0x%02x%n",
+                               closure, off-8, off-1);
+                else
+                    out.format("\" +      // 0x%02x - 0x%02x%n",
+                               off-8, off-1);
+            } else {
+                if (off == end)
+                    out.format("\"%s%n", closure);
+                else
+                    out.format("\" +%n");
+            }
         }
     }
 
     static Pattern sbmap = Pattern.compile("0x(\\p{XDigit}++)\\s++U\\+(\\p{XDigit}++)(\\s++#.*)?");
-    private static int genB2C(InputStream in, StringBuilder out)
+
+    private static void genClass(String srcDir, String dstDir,
+                                 String template,
+                                 String clzName,
+                                 String csName,
+                                 String hisName,
+                                 String pkgName,
+                                 boolean isASCII)
         throws Exception
     {
-        char[] sb = new char[0x100];
-        int[] indexC2B = new int[0x100];
+        StringBuilder b2cSB = new StringBuilder();
+        StringBuilder b2cNRSB = new StringBuilder();
+        StringBuilder c2bNRSB = new StringBuilder();
 
-        for (int i = 0; i < sb.length; i++)
-            sb[i] = UNMAPPABLE_DECODING;
+        char[] sb = new char[0x100];
+        char[] c2bIndex = new char[0x100];
+        int    c2bOff = 0;
+        Arrays.fill(sb, UNMAPPABLE_DECODING);
+        Arrays.fill(c2bIndex, UNMAPPABLE_DECODING);
 
-        // parse the b2c mapping table
+        // (1)read in .map to parse all b->c entries
+        FileInputStream in = new FileInputStream(
+                                 new File(srcDir, clzName + ".map"));
         Parser p = new Parser(in, sbmap);
         Entry  e = null;
-        int    off = 0;
+
         while ((e = p.next()) != null) {
             sb[e.bs] = (char)e.cp;
-            if (indexC2B[e.cp>>8] == 0) {
-                off += 0x100;
-                indexC2B[e.cp>>8] = 1;
+            if (c2bIndex[e.cp>>8] == UNMAPPABLE_DECODING) {
+                c2bOff += 0x100;
+                c2bIndex[e.cp>>8] = 1;
             }
         }
 
-        Formatter fm = new Formatter(out);
+        Formatter fm = new Formatter(b2cSB);
         fm.format("%n");
 
         // vm -server shows cc[byte + 128] access is much faster than
         // cc[byte&0xff] so we output the upper segment first
-        toString(sb, 0x80, 0x100, fm, "+");
-        toString(sb, 0x00, 0x80,  fm, ";");
-
+        toString(sb, 0x80, 0x100, fm, "+", true);
+        toString(sb, 0x00, 0x80,  fm, ";", true);
         fm.close();
-        return off;
-    }
 
-    // generate non-roundtrip entries from xxx.nr file
-    private static String genNR(InputStream in) throws Exception
-    {
-        StringBuilder sb = new StringBuilder();
-        Formatter fm = new Formatter(sb);
-        Parser p = new Parser(in, sbmap);
-        Entry  e = null;
-        fm.format("// remove non-roundtrip entries%n");
-        fm.format("        b2cMap = b2cTable.toCharArray();%n");
-        while ((e = p.next()) != null) {
-            fm.format("        b2cMap[%d] = UNMAPPABLE_DECODING;%n",
-                      (e.bs>=0x80)?(e.bs-0x80):(e.bs+0x80));
-        }
-        fm.close();
-        return sb.toString();
-    }
+        // (2)now the .nr file which includes "b->c" non-roundtrip entries
+        File f = new File(srcDir, clzName + ".nr");
+        if (f.exists()) {
+            in = new FileInputStream(f);
+            fm = new Formatter(b2cNRSB);
+            p = new Parser(in, sbmap);
+            e = null;
 
-    // generate c2b only entries from xxx.c2b file
-    private static String genC2BNR(InputStream in) throws Exception
-    {
-        StringBuilder sb = new StringBuilder();
-        Formatter fm = new Formatter(sb);
-        Parser p = new Parser(in, sbmap);
-        ArrayList<Entry> es = new ArrayList<Entry>();
-        Entry  e = null;
-        while ((e = p.next()) != null) {
-            es.add(e);
+            fm.format("// remove non-roundtrip entries%n");
+            fm.format("        b2cMap = b2cTable.toCharArray();%n");
+            while ((e = p.next()) != null) {
+                fm.format("        b2cMap[%d] = UNMAPPABLE_DECODING;%n",
+                          (e.bs>=0x80)?(e.bs-0x80):(e.bs+0x80));
+            }
+            fm.close();
         }
 
-        fm.format("// non-roundtrip c2b only entries%n");
-        fm.format("        c2bNR = new char[%d];%n", es.size() * 2);
-        int i = 0;
-        for (Entry entry: es) {
-            fm.format("        c2bNR[%d] = 0x%x; c2bNR[%d] = 0x%x;%n",
-                      i++, entry.bs, i++, entry.cp);
+        // (3)finally the .c2b file which includes c->b non-roundtrip entries
+        f = new File(srcDir, clzName + ".c2b");
+        if (f.exists()) {
+            in = new FileInputStream(f);
+            fm = new Formatter(c2bNRSB);
+            p = new Parser(in, sbmap);
+            e = null;
+            ArrayList<Entry> es = new ArrayList<Entry>();
+            while ((e = p.next()) != null) {
+                if (c2bIndex[e.cp>>8] == UNMAPPABLE_DECODING) {
+                    c2bOff += 0x100;
+                    c2bIndex[e.cp>>8] = 1;
+                }
+                es.add(e);
+            }
+            fm.format("// non-roundtrip c2b only entries%n");
+            if (es.size() < 100) {
+                fm.format("        c2bNR = new char[%d];%n", es.size() * 2);
+                int i = 0;
+                for (Entry entry: es) {
+                    fm.format("        c2bNR[%d] = 0x%x; c2bNR[%d] = 0x%x;%n",
+                              i++, entry.bs, i++, entry.cp);
+                }
+            } else {
+                char[] cc = new char[es.size() * 2];
+                int i = 0;
+                for (Entry entry: es) {
+                    cc[i++] = (char)entry.bs;
+                    cc[i++] = (char)entry.cp;
+                }
+                fm.format("        c2bNR = (%n");
+                toString(cc, 0, i,  fm, ").toCharArray();", false);
+            }
+            fm.close();
         }
-        fm.close();
-        return sb.toString();
-    }
 
-    private static void genSBCSClass(String srcDir,
-                                     String dstDir,
-                                     String template,
-                                     String clzName,
-                                     String csName,
-                                     String hisName,
-                                     String pkgName,
-                                     boolean isASCII,
-                                     String b2c,
-                                     String b2cNR,
-                                     String c2bNR,
-                                     int    c2blen)
-        throws Exception
-    {
+        // (4)it's time to generate the source file
+        String b2c = b2cSB.toString();
+        String b2cNR = b2cNRSB.toString();
+        String c2bNR = c2bNRSB.toString();
+
         Scanner s = new Scanner(new File(srcDir, template));
         PrintStream out = new PrintStream(new FileOutputStream(
                               new File(dstDir, clzName + ".java")));
@@ -239,16 +249,16 @@ public class GenerateSBCS {
                 line = line.replace("$B2CTABLE$", b2c);
             }
             if (line.indexOf("$C2BLENGTH$") != -1) {
-                line = line.replace("$C2BLENGTH$", "0x" + Integer.toString(c2blen, 16));
+                line = line.replace("$C2BLENGTH$", "0x" + Integer.toString(c2bOff, 16));
             }
             if (line.indexOf("$NONROUNDTRIP_B2C$") != -1) {
-                if (b2cNR == null)
+                if (b2cNR.length() == 0)
                     continue;
                 line = line.replace("$NONROUNDTRIP_B2C$", b2cNR);
             }
 
             if (line.indexOf("$NONROUNDTRIP_C2B$") != -1) {
-                if (c2bNR == null)
+                if (c2bNR.length() == 0)
                     continue;
                 line = line.replace("$NONROUNDTRIP_C2B$", c2bNR);
             }