The following types of character code sets can be used by the source and target databases:

• EBCDIC/KEIS, EBCDIK/KEIS
  These character code sets are used by a source database that is a mainframe database. There are four available EBCDIC/KEIS and EBCDIK/KEIS character code sets:
  
  • EBCDIC/KEIS78
  • EBCDIK/KEIS78
  • EBCDIC/KEIS83
  • EBCDIK/KEIS83
  These four character code sets are referred to collectively in this manual as the EBCDIK/KEIS character code set.
  
• EBCDIK
  This character code set constitutes a mainframe's one-byte codes.
  
• JIS8/Shift JIS
  These character code sets are used when the source or target database is HiRDB. They can be used when use of shift JIS Kanji encoding is specified in HiRDB's pdsetup command.
  You must specify the appropriate HiRDB character code set with the dblocale operand in the extraction system definition for the source Datareplicator and with the dblocale operand in the import system definition for the target Datareplicator.
  
• EUC
  This character encoding set is used when the source or target database is HiRDB. EUC is used when EUC Japanese Kanji encoding is specified in HiRDB's pdsetup command.
  You must specify the appropriate HiRDB character code set with the dblocale operand in the extraction system definition for the source Datareplicator and with the dblocale operand in the import system definition for the target Datareplicator.
  
• UCS2
  UCS2 is the Unicode UCS2 character set. Datareplicator supports only rules for encoding to UTF-8; it does not take into account mapping between codes and characters. The supported range of Unicode is UCS2 BMP (basic multi-language), and encoding of UCS2 to UTF-8 format is supported.
  
• UTF-8
  This character string encoding is applicable when the source or target database is HiRDB and is used when UTF-8 is specified in HiRDB's pdsetup command. UTF-8 is one of the character encoding schemes that represent Unicode characters, and expresses a character as a combination of 8-bit information. The applicable HiRDB character encoding is specified in the dblocale operand of the extraction system definition for the source Datareplicator, and in the dblocale operand of the import system definition for the target Datareplicator.
  

If the source and target databases do not use the same character code set, Datareplicator converts character codes in accordance with a provided definition. If the source and target databases use the same character code set, there is no need to convert character codes.

(a) Character code conversion by the target Datareplicator

The target Datareplicator executes character code conversion in accordance with the correspondence between the character code sets at the source and target databases. You use the dblocale operand in the import system definition to specify the target database's character code set.

If the source database is a mainframe database, use the ebcdic_type operand in the import environment definition to specify the type of EBCDIK/KEIS character code set for the source database. The following table shows the correspondence of character code sets between the source and target databases.

Table 4-15 Correspondence of character code sets between the source and target databases

Source database's character code set	Target database's character code set
	EBCDIK/KEIS#1	EBCDIK	JIS8/Shift JIS	EUC#2	UTF-8
EBCDIK/KEIS#1	--	--	Y	Y	Y
EBCDIK	--	--	Y	--	--
JIS8/Shift JIS	Y	Y	--	Y	Y
EUC#2	Y	--	Y	--	Y
UTF-8	Y	--	Y	Y	--

Y: The character code set used in the source database can be converted to the character code set used in the target database.

--: There is no need to convert the character code set.

#1

Only XDM/DS and VOS3 Database Datareplicator support EBCDIK/KEIS.

#2

Conversion results might not be correct if the specification of the character code set is not in the range 0 to 2.

Datareplicator uses the following character code conversion rules:

Character code conversion method

Uses a mapping table for converting character codes.

Definition of Gaiji conversion method

For conversion of Gaiji characters, you must use the hdsccnvedt command to edit and define a mapping table for converting character codes.

(a) Single-byte codes

Datareplicator converts a single-byte code to the corresponding Shift JIS character code.

(b) Double-byte codes (standard character codes)

• Datareplicator converts a double-byte code to the corresponding Shift JIS character code.
• Function characters ((0A42)₁₆ and (0A41)₁₆) are deleted and shifted up. If this results in a remainder, it is converted to a space character ((20)₁₆).
• Codes (00)₁₆ to (40)₁₆ enclosed by function characters are converted to the corresponding Shift JIS character codes.
• If the last character is the first byte of a double-byte code, Datareplicator converts it to a space character ((20)₁₆).

(c) Double-byte codes (Gaiji)

Datareplicator uses the provided mapping table for converting character codes to convert double-byte Gaiji codes. If a Gaiji character is not defined or no mapping table for converting character codes has been provided, Datareplicator converts the Gaiji code to a space character according to the value of the undefcode_cnv operand specified in the import environment definition.

(d) Space character

Datareplicator converts the space character in shift code. The following table shows the conversion rules for the space character (from EBCDIK/KEIS to JIS8/Shift JIS).

Table 4-16 Conversion rules for the space character (from EBCDIK/KEIS to JIS8/Shift JIS)

EBCDIK/KEIS character code	JIS8/Shift JIS character code
Double-byte space character (A1A1)16	Double-byte space character (8140)16
Two consecutive single-byte space characters (40)16(40)16	Converted to one double-byte space character or two consecutive single-byte space characters, depending on the shiftspace_cnv operand value specified in the import environment definition
Single-byte space character (40)16	Single-byte space character (20)16

(a) Single-byte codes

• Datareplicator converts a single-byte code, excluding kana characters, to the corresponding EUC character code.
• Datareplicator converts a kana character to a double-byte code.

(b) Double-byte codes (standard character codes)

• Datareplicator converts a double-byte code to the corresponding EUC character code.
• Function characters ((0A42)₁₆ and (0A41)₁₆) are deleted and shifted up. If this results in a remainder, it is converted to a space character ((20)₁₆).
• Codes (00)₁₆ to (40)₁₆ enclosed by function characters are converted to the corresponding Shift JIS character codes.
• If the last character is the first byte of a double-byte code, Datareplicator converts it to a space character ((20)₁₆).

(c) Double-byte codes (Gaiji)

Datareplicator assigns the last 8,836 characters in the 9,024-character Gaiji area as the Gaiji area for EUC character codes for conversion purposes.

Datareplicator converts any other code in accordance with the user-created mapping table for converting character codes. If a Gaiji character is not defined or no mapping table for converting character codes has been provided, Datareplicator converts the Gaiji code to a space character according to the value of the undefcode_cnv operand specified in the import environment definition.

Note that code set 3 might result in an error when an SQL statement is issued.

(d) Space character

Datareplicator converts the space character. The following table shows the conversion rules for the space character (from EBCDIK/KEIS to EUC).

Table 4-17 Conversion rules for the space character (from EBCDIK/KEIS to EUC)

EBCDIK/KEIS character code	JIS8/Shift JIS character code
Double-byte space character (A1A1)16	Double-byte space character (A1A1)16
Two consecutive single-byte space characters (40)16(40)16	Converted to one double-byte space character or two consecutive single-byte space characters, depending on the shiftspace_cnv operand value specified in the import environment definition
Single-byte space character (40)16	Single-byte space character (20)16

(e) Handling of overflow

Datareplicator converts a single-byte kana character to a double-byte character. If the source table contains single-byte kana characters, the data length increases after conversion and data overflow might occur. The following table shows how overflow is handled (from EBCDIK/KEIS to EUC).

Table 4-18 Handling of overflow (from EBCDIK/KEIS to EUC)

Location of overflow	Datareplicator's action	User's action
Literal or update data	SQL error (applicable if the data exceeds the defined length after conversion of all bytes)	Change the defined length and re-execute import processing.

Note:

If an identifier exceeds 30 bytes or update data exceeds 32,000 bytes, Datareplicator is unable to continue processing. Before starting data linkage, make sure that source system identifiers and update data will not exceed the permitted maximum length.

(a) Single-byte codes

• Datareplicator converts single-byte codes, excluding kana characters, to the corresponding UTF-8 character codes.
• Datareplicator converts a kana character to a three-byte code.

(b) Double-byte codes (standard character codes)

• Datareplicator converts double-byte codes to the corresponding UTF-8 character codes.
• Function characters ((0A42)₁₆ and (0A41)₁₆) are deleted and shifted up. If this results in a remainder, the remainder is converted to a space character ((20)₁₆).
• Codes (00)₁₆ to (40)₁₆ enclosed by function characters are converted to the corresponding UTF-8 character codes.
• If the last character is the first byte of a double-byte code, Datareplicator converts it to a space character ((20)₁₆).

(c) Double-byte codes (Gaiji)

Datareplicator uses the provided mapping table for converting character codes to convert double-byte Gaiji codes.

If a Gaiji character is not defined or no mapping table for converting character codes has been provided, Datareplicator converts the Gaiji code to a space character according to the value of the undefcode_cnv operand specified in the import environment definition.

(d) Space character

Datareplicator converts the space character in shift code. The following table shows the conversion rules for the space character (from EBCDIK/KEIS to UTF-8).

Table 4-19 Conversion rules for the space character (from EBCDIK/KEIS to UTF-8)

EBCDIK/KEIS character code	UTF-8 character code
Double-byte space character (A1A1)16	Double-byte space character (E38080)16
Two consecutive single-byte space characters (40)16(40)16	Converted to one double-byte space character or two consecutive single-byte space characters, depending on the shiftspace_cnv operand value specified in the import environment definition
Single-byte space character (40)16	Single-byte space character (20)16

(e) Handling of overflow

Datareplicator converts a single-byte kana character to a three-byte character and a double-byte standard kanji character to a three-byte character. If the source table contains single-byte kana characters or standard kanji characters, the data length increases after conversion and data overflow might occur.

The following table shows how overflow is handled (from EBCDIK/KEIS to UTF-8).

Table 4-20 Handling of overflow (from EBCDIK/KEIS to UTF-8)

Location of overflow	Datareplicator's action	User's action
Literal or update data	SQL error (applicable if the data exceeds the defined length after conversion of all bytes)	Change the defined length and re-execute import processing.

Note:

If an identifier exceeds 30 bytes or update data exceeds 32,000 bytes, Datareplicator is unable to continue processing. Before starting data linkage, make sure that source system identifiers and update data will not exceed the permitted maximum lengths.

(f) Notes

If the target database is HiRDB using UTF-8, a column of the NCHAR or NVARCHAR type cannot be created in the table. To store such data in a different data type, such as MCHAR, use a column data editing UOC routine.

(a) Single-byte codes

Datareplicator converts such single-byte codes to the corresponding JIS8 character codes (single-byte codes). Because only single-byte codes can be specified in a column for which EBCDIK is specified in the character set specification, the following might result:

• All characters are treated as being single-byte codes. If double-byte codes are converted, garbled characters might result.
• In single-byte code conversion, the length of the data remains the same before and after the conversion.
• No character code conversion error occurs because all characters are processed as being single-byte codes.

(a) Single-byte codes

Datareplicator converts single-byte codes to the corresponding EBCDIK character codes.

(b) Double-byte codes (standard character codes)

Datareplicator treats bytes 1 and 2 of a double-byte code as single-byte codes and converts them to single-byte EBCDIK code characters. Therefore, the following might result:

• Converting double-byte codes results in garbled characters.
• In single-byte code conversion, the length of the data remains the same before and after the conversion.
• No character code conversion error occurs because all characters are processed as being single-byte codes.

(a) Single-byte codes

Datareplicator converts a single-byte code to the corresponding EBCDIK/KEIS character code.

(b) Double-byte codes (standard character codes)

Datareplicator converts a double-byte code to the corresponding EBCDIK/KEIS character code. If the last character is the first byte of a double-byte code, Datareplicator converts it to a space character ((40)₁₆).

(c) Double-byte codes (Gaiji)

Datareplicator converts a double-byte Gaiji code in accordance with the mapping table for converting character codes. If a Gaiji character is not defined or no mapping table for converting character codes has been provided, Datareplicator converts the Gaiji code to the space character ((4040)₁₆).

(d) Space character

Datareplicator converts a double-byte space character ((8140)₁₆) to the corresponding double-byte space character ((A1A1)₁₆) and a single-byte space character ((20)₁₆) to the corresponding single-byte space character ((40)₁₆).

(a) Single-byte codes

• Datareplicator converts each single-byte code, excluding kana characters, to the corresponding UTF-8 character code.
• Datareplicator converts a kana character to a three-byte code.

(b) Double-byte codes (standard character codes)

Datareplicator converts double-byte codes to the corresponding UTF-8 character codes.

(c) Double-byte codes (Gaiji)

Datareplicator uses the provided mapping table for converting character codes to convert double-byte Gaiji codes.

If a Gaiji character is not defined or no mapping table for converting character codes has been provided, Datareplicator converts the Gaiji code to a space character according to the value of the undefcode_cnv operand specified in the import environment definition.

(d) Space character

Datareplicator converts the space character in shift code. The following table shows the conversion rules for the space character (from JIS8/Shift JIS to UTF-8).

Table 4-21 Conversion rules for the space character (from JIS8/Shift JIS to UTF-8)

JIS8/SJIS character code	UTF-8 character code
Double-byte space character (8140)16	Double-byte space character (E38080)16
Single-byte space character (20)16	Single-byte space character (20)16

(e) Handling of overflow

Datareplicator converts a single-byte kana character to a three-byte character and a double-byte standard kanji character to a three-byte character. If the source table contains single-byte kana characters or standard kanji characters, the data length increases after conversion and data overflow might occur. The following table shows how overflow is handled (from JIS8/Shift JIS to UTF-8).

Table 4-22 Handling of overflow (from JIS8/Shift JIS to UTF-8)

Location of overflow	Datareplicator's action	User's action
Literal or update data	SQL error (applicable if the data exceeds the defined length after conversion of all bytes)	Change the defined length and re-execute import processing.

Note:

If an identifier exceeds 30 bytes or update data exceeds 32,000 bytes, Datareplicator is unable to continue processing. Before starting data linkage, make sure that source system identifiers and update data will not exceed the permitted maximum lengths.

(f) Notes

• If the target database is HiRDB using UTF-8, a column of the NCHAR or NVARCHAR type cannot be created in the table. To store such data in a different data type, such as MCHAR, use a column data editing UOC routine.
• There are two types of mapping from SJIS Kanji encoding to Unicode (UCS2), as shown in the following table.
  

  Table 4-23 Character mapping in SJIS Kanji encoding and Unicode

  Mapping method			Description
  Method	Target	Range of kanji
  JIS method	SJIS to JIS X0221	JIS standard level 1	Depends on the mapping stipulated by JIS X0221.
  		JIS standard level 2
  MS method	Windows-the customer support center character set to MS-The customer support center	JIS standard level 1	Depends on the mapping defined by Microsoft.
  		JIS standard level 2
  		Vendor-extended characters

  In the MS method, vendor-extended characters have been added; when characters are converted to UCS2, the code point that is set is not the same as with the JIS method. Datareplicator uses the MS method.
  
• If data of the NCHAR type is converted and the length of UTF-8 data obtained after conversion consists of an odd number of bytes, a single-byte space is added to the resulting data. What is set in the collen member (column data length) is the length of the data obtained after conversion (in bytes) + 1 2.
  If the length of UTF-8 data obtained after conversion consists of an even number of bytes, the resulting data is used as is. What is set in the collen member (column data length) is the length of the data obtained after conversion (in bytes) 2.
  
• If the definition length n of the NCHAR type in the source table is an even number, the definition length of the MCHAR type in the target table becomes n x 3.
  If the definition length n of the NCHAR type in the source table is an odd number, the definition length of the MCHAR type in the target table becomes (n x 3) + 1.
  

(a) Single-byte codes

Datareplicator converts a single-byte code to the corresponding Shift JIS character code.

(b) Double-byte codes (standard character codes)

Datareplicator converts a double-byte code to the corresponding Shift JIS character code. If the last character is the first byte of a double-byte code, Datareplicator converts it to a space character ((20)₁₆).

(c) Double-byte codes (Gaiji)

Datareplicator converts a double-byte Gaiji code in accordance with the mapping table for converting character codes. If a Gaiji character is not defined or no mapping table for converting character codes has been provided, Datareplicator converts the Gaiji code to a space character according to the value of the undefcode_cnv operand specified in the import environment definition.

(d) Space character

Datareplicator converts a double-byte space character ((A1A1)₁₆) to the corresponding code ((8140)₁₆). Datareplicator converts two consecutive single-byte space characters ((40)₁₆) to a double-byte space character ((8140)₁₆). One single-byte space character ((40)₁₆) is converted to the corresponding code ((20)₁₆).

(a) Single-byte codes

Datareplicator converts a single-byte code, excluding kana characters, to the corresponding UTF-8 character code.

(b) Double-byte codes (standard character codes)

• Datareplicator converts a kana character to a three-byte code.
• Datareplicator converts a double-byte code to the corresponding UTF-8 character code.

(c) Three-byte codes (Gaiji)

Datareplicator converts a double-byte code (Gaiji) in accordance with the user-created mapping table for converting character codes.

If a Gaiji character is not defined or no mapping table for converting character codes has been provided, Datareplicator converts the Gaiji code to a space character according to the value of the undefcode_cnv operand specified in the import environment definition.

(d) Space character

Datareplicator converts the space character in shift code. The following table shows the conversion rules for the space character (from EUC to UTF-8).

Table 4-24 Conversion rules for the space character (from EUC to UTF-8)

EUC character code	UTF-8 character code
Double-byte space character (A1A1)16	Double-byte space character (E38080)16
Single-byte space character (20)16	Single-byte space character (20)16

(e) Handling of overflow

Datareplicator converts a double-byte kana character to a three-byte character and a double-byte standard kanji character to a 3-byte character. If the source table contains single-byte kana characters or standard kanji characters, the data length increases after conversion and data overflow might occur. The following table shows how overflow is handled (from EUC to UTF-8).

Table 4-25 Handling of overflow (for conversion from EUC to UTF-8)

Location of overflow	Datareplicator's action	User's action
Literal or update data	SQL error (applicable if the data exceeds the defined length after conversion of all bytes)	Change the defined length and re-execute import processing.

Note:

If an identifier exceeds 30 bytes or update data exceeds 32,000 bytes, Datareplicator is unable to continue processing. Before starting data linkage, make sure that source system identifiers and update data will not exceed the permitted maximum lengths.

(f) Notes

• If the target database is HiRDB using UTF-8, a column of the NCHAR or NVARCHAR type cannot be created in the table. To store data in a different data type, such as MCHAR, use a column data editing UOC routine.
• If data of the NCHAR type is converted and the length of UTF-8 data obtained after conversion consists of an odd number of bytes, a single-byte space is added to the resulting data. What is set in the collen member (column data length) is the length of the data obtained after conversion (in bytes) + 1 2.
  If the length of UTF-8 data obtained after conversion consists of an even number of bytes, the resulting data is used as is. What is set in the collen member (column data length) is the length of the data obtained after conversion (in bytes) 2.
  
• If the definition length n of the NCHAR type in the source table is an even number, the definition length of the MCHAR type in the target table becomes n x 3.
  If the definition length n of the NCHAR type in the source table is an odd number, the definition length of the MCHAR type in the target table becomes (n x 3) + 1.
  

(a) Single-byte codes

Datareplicator converts a single-byte code to the corresponding character code.

(b) Double-byte and three-byte codes (standard kanji)

Datareplicator converts a double-byte or three-byte code to the corresponding UTF-8 character code.

(c) Three-byte codes (Gaiji)

Datareplicator converts a three-byte code in accordance with the user-created mapping table for converting character codes.

If a Gaiji character is not defined or no mapping table for converting character codes has been provided, Datareplicator converts the Gaiji code to a space character according to the value of the undefcode_cnv operand specified in the import environment definition.

You can suppress character code conversion for a column at the target Datareplicator by specifying suppression of character code conversion for the desired extracted column. For details about the specification to suppress character code conversion, see the update information field definition in 5.10 Import definition.

(a) Conversion rules for JIS8/Shift JIS codes

The following table shows the conversion rules for JIS8/Shift JIS codes.

Table 4-26 Conversion rules for JIS8/Shift JIS codes

1-byte	2-byte	3-byte	Code conversion rule
0x00 to 0x80	--	--	Treats as JIS8 and converts it to the corresponding code.
0x81 to 0x9F	0x40 to 0xFC (excluding 0x7F)	--	Treats as SJIS (Kanji) and converts it to the corresponding code.
	Other	--	Converts according to the specification of undefcode_cnv.
	--	--	Treats as incomplete code and skips it without converting.
0xA0 to 0xDF	--	--	Treats as JIS8 and converts it to the corresponding code.
0xE0 to 0xEF	0x40 to 0xFC (excluding 0x7F)	--	Treats as SJIS (Kanji) and converts it to the corresponding code.
	Other	--	Converts according to the specification of undefcode_cnv.
	--	--	Treats as incomplete code and skips it without converting.
0xF0 to 0xFC	0x40 to 0xFC (excluding 0x7F)	--	Treats as SJIS (Gaiji) and converts it to the corresponding code.
	Other	--	Converts according to the specification of undefcode_cnv.
	--	--	Treats as incomplete code and skips it without converting.
0xFD to 0xFF	--	--	Treats as JIS8 and converts it to the corresponding code.

(b) Conversion rules for EUC codes

The following table shows the conversion rules for EUC codes.

Table 4-27 Conversion rules for EUC codes

1-byte	2-byte	3-byte	Code conversion rule
0x00 to 0x8D	--	--	Treats as code set 0 and converts it to the corresponding code.
0x8E	0xA0 to 0xFF	--	Treats as code set 2 (kana characters) and converts it to the corresponding code.
	Other	--	Converts according to the specification of undefcode_cnv.
	--	--	Treats as incomplete code and skips it without converting.
0x8F	0xA1 to 0xFE	0xA1 to 0xFE	Treats as code set 3 (Gaiji) and converts it to the corresponding code.
		Other	Converts according to the specification of undefcode_cnv.
		--	Treats as incomplete code and skips it without converting.
	Other	0xA1 to 0xFE	Converts according to the specification of undefcode_cnv.
		Other
		--	Treats as incomplete code and skips it without converting.
	--	--
0x90 to 0x9F	--	--	Treats as code set 0 and converts it to the corresponding code.
0xA0	--	--	Converts to 0x20.
0xA1 to 0xFE	0xA1 to 0xFE	--	Treats as code set 1 (Kanji) and converts it to the corresponding code.
	Other	--	Converts according to the specification of undefcode_cnv.
	--	--	Treats as incomplete code and skips it without converting.

(c) Conversion rules for EBCDIK/KEIS codes

The following table shows the conversion rules for EBCDIK/KEIS codes.

Table 4-28 Conversion rules for EBCDIK/KEIS codes

1-byte		2-byte	3-byte	Code conversion rule
Single-byte shift on	0x00-0x09	--	--	Converts to the corresponding code.
	0x0A	0x41	--	Toggles single-byte shift without performing conversion.
		0x42	--	Toggles double-byte shift without performing conversion.
		Other	--	Converts byte 1 or byte 2 as a single-byte code.
		--	--	Treats as incomplete code and skips it without converting.
	0x0B to 0xFF	--	--	Converts to the corresponding code.
Double-byte shift on	0x00 to 0x40	--	--	Converts byte 1 or byte 2 as a single-byte code.
	0x41 to 0xA0	0xA1 to 0xFE	--	Converts to the corresponding code.
		Other	--	Converts according to the specification of undefcode_cnv.
		--	--	Treats as incomplete code and skips it without converting.
	0xA1 to 0xFE	0xA1 to 0xFE	--	Converts to the corresponding code.
		Other	--	Converts according to the specification of undefcode_cnv.
		--	--	Treats as incomplete code and skips it without converting.
	0xFF	0x01 to 0xFF	--	Converts according to the specification of undefcode_cnv.
		--	--	Treats as incomplete code and skips it without converting.

(d) Conversion rules for UTF-8 codes

The following table shows the conversion rules for UTF-8 codes.

Table 4-29 Conversion rules for UTF-8 codes

1-byte	2-byte	3-byte	Code conversion rule
0x00 to 0x7F	--	--	Treats as single-byte code and converts it to the corresponding code.
0x80 to 0xBF	0x00 to 0xFF	--	Converts to 0x20.
	--	--
0xC2 to 0xDE	0x80 to 0xFF	--	Treats as double-byte code and converts it to the corresponding code.
	Other	--	Converts according to the specification of undefcode_cnv.
	--	--	Treats as incomplete code and skips it without converting.
0xDF	0x80 to 0xBF	--	Treats as double-byte code and converts it to the corresponding code.
	Other	--	Converts according to the specification of undefcode_cnv.
	--	--	Treats as incomplete code and skips it without converting.
0xE0	0xA0 to 0xFF	0x80 to 0xFF	Treats as three-byte code and converts it to the corresponding code.
		Other	Converts according to the specification of undefcode_cnv.
		--	Treats as incomplete code and skips it without converting.
	Other	0x80 to 0xFF	Converts according to the specification of undefcode_cnv.
		Other
		--	Treats as incomplete code and skips it without converting.
	--	--
0xE1 to 0xEE	0x80 to 0xFF	0x80 to 0xFF	Treats as three-byte code and converts it to the corresponding code.
		Other	Converts according to the specification of undefcode_cnv.
		--	Treats as incomplete code and skips it without converting.
	Other	0x80 to 0xFF	Converts according to the specification of undefcode_cnv.
		--	Treats as incomplete code and skips it without converting.
	--	--
0xEF	0x80 to 0xBF	0x80 to 0xBF	Treats as three-byte code and converts it to the corresponding code.
		Other	Converts according to the specification of undefcode_cnv.
		--	Treats as incomplete code and skips it without converting.
	Other	0x80 to 0xBF	Converts according to the specification of undefcode_cnv.
		Other
		--	Treats as incomplete code and skips it without converting.
	--	--
Other	--	--	Converts according to the specification of undefcode_cnv.