# skf conversion table
# this table is for input side, and specify a certain code to
# unic*de conversion.
# by S.Kaneko. This file is in public domain.
#
# put skf conversion table into this directory.
#
1. file name (input table)

example: KS-C_5601 case

              iIM44C.stb
               |||||
I: iso-2022----+|||+--- capitailized iso-2022 calling identify character
N: not     ----+|||
                ||+---- 6: 96-character set ('7' if small calling identify char)
M: Multibyte  --+|+---- 4: 94 character set ('5' if small calling identify char)
S: Single byte -+|
                 |
4: 4byte seq. ---+
3: 3byte seq. ---+

See source(in_code_table.c) about 'iso-2022 calling identify 
character' for non-iso-2022 codeset, for this part requires
code modification anyway. Third character is +1'd iff iso-2022
calling identify charatcter is small letter.

2. file name (output table)

example: KS-C_5601 case

              oIMMKC.stb
               |||||
I: iso-2022----+|||+--- capitalized iso-2022 calling identify character
N: not     ----+|||
                ||+---- L: latin code set
M: Multibyte  --+|+---- G: Graphic char set
S: Single byte -+|+---- N: kana code set 
                 |+---- A: CJK extension A charset
M: Multibase ----++---- K: Kanji code set
S: Singlebase ---++---- H: Hangle code set
		  +---- Y: Y code set
		  +---- C: Compatible area charset
		  +---- B: CJK extension B charset
		  +---- X: CJK compatible extension charset
		  +---- S: ascii code set

See source(out_code_table.c) about 'iso-2022 calling identify 
character' for non-iso-2022 codeset, for this part requires
code modification anyway. Third character is +1'd iff iso-2022
calling identify charatcter is small letter.

3. format

Byte 0-9: magic "skf_ctable"
Byte 10:  table entry byte-length '2': 2byte(UCS2), '4': 4byte(UCS4)
Byte 11:  Table revision. Currently '0'
Byte 12-: binary - unsigned big-endian UCS2 code sequence to specify
	  consecutive unic*de location for the code set.

