Print

Print


----------------------------Original message----------------------------
|   There has been developed an encoding scheme for (all?) known
|   characters into plain ascii characters. The work has been done by
|   Keld Simonsen <[log in to unmask]>.
 
I have tried to use this encoding scheme.  I was in the working group
(IAB/IETF 822 WG) which rejected it on technical merit, and on several
procedural factors.  The encoding scheme uses characters from the
invariant set from ISO 646, usually in pairs, to encode other
characters.  It's called "mnemonic".  It works well for the accented
characters of ISO Latin 1 (ISO 8859-1).  Included in the tables are
approximately 2000 characters, with sharply decreasing "mnemonic"
characteristics in their encoding.  Two characters isn't enough, and
can't be enough.  Apart from the ideographic characters of Chinese,
Japanese, and Korean, ISO DIS 10646-1.2 contains 7500 characters.  It is
obvious that with 84 characters, only 7056 combinations are available,
95% of which are far from mnemonic.  The encoding scheme is static, with
a fixed binding between a given "mnemonic" character sequence and a
character.
 
SGML, on the other hand, has established an _enabling_ notation, in
which it is possible to choose _any_ name for a character, according to
conventions and user needs.  The key is the "definitional character
entity set", which maps an entity name to a _character_ (meaning), as
opposed to the "display character entity set", which maps it to a _glyph
id_ or _coded character_ for display or other processing purposes.
 
The definitional character entity set has usually seen entity
declarations of the form
 
	<!ENTITY foobar SDATA "[foobar]" --foo with bar above-->
 
where the comment was intended to embody some kind of standardized
definitional name of the character, but the entity itself was just a
bracketed version of the entity name.  These have been largely useless,
as the entity name is not part of the ESIS, and a display version of the
definitional character entity set had to be remapped manually.
 
Much effort has been put into adopting a standardized set of entity
_names_, whereas the purpose has been to standardize an encoding for the
_characters_ some entity is used to access.
 
With the adoption of ISO 10646, this can change.  ISO 10646 contains all
(all!) known characters in the universe, and part 1 (ISO 10646-1), the
Basic Multilingual Plane, is already here, with around 40 thousand
characters, all with unique names.  We can use the unique names in the
definitional character entity sets, and then the mapping to a new
display version can be done mechanically, by describing the native
character set or glyph set in terms of these unique names:
 
	<!ENTITY foobar SDATA "FOO WITH BAR ABOVE">
 
If we also use the same names to describe characters in public character
sets (and private versions, too), we don't need to worry about mapping
and conversion tables, anymore.  They can be created on the fly.
 
I attach a fairly long article, which has two functions: expose the
weaknesses of Keld Simonsen's design, and to suggest a solution using
SGML's character set declarations.  I present our versions of the
complete, encoded ISO Latin 1 (ISO 8859-1:1987) as an example.  (This is
what makes the article so long).
 
As Glenn correctly states, ISO 10646 has already done a complete
encoding of IPA.  Although I couldn't care less which numbers are
assigned to a given character, I'm deeply appreciative of the unique
names that SC 2/WG 2 have assigned to them.  It makes the number a
matter of convention, one which is only a handle to the name and meaning
of a character in the context of coded representation.  Transformation
between coded representations is thereby possible by meaning, rather
than by hand-prepared number-to-number conversion tables.
 
My work is complete for character set registered according to ISO 2375.
I'm working on the IBM code pages, and the assorted randomness from
other vendors.  The work will not be published until I can get some
funding for this work.  I can give away many individual hours every
week, but I can't give away four months of my time without some
remuneration.  If, from the following article, the TEI would wish to
take part in the funding, I think much work in the character set WG can
be saved, and a cleaner design can be used.  Included in my work is a
complete set of library functions to access the tables, a complete ISO
2022 data stream parser and translator to ISO 10646, and utilities to
convert SGML documents between any two document character sets, given
appropriate and available entity sets.  Apart from the character set
declarations, also definitional character entity sets from ISO 8879,
with the same names.  The complete character set declaration for ISO
10646 is 1.4M in size, and contains names for 29,000 characters.
 
I plan to write the specifications for and to implement a "character set
manager" for SGML, to be located between the entity manager and the
parser proper, so a parser can always work with ISO 10646, and the
application can provide the parser with its own character set, in which
it will receive the document data.
 
A forthcoming artice in <TAG> will describe the design.  It will be
submitted to ISO for consideration in the revised SGML, and new entity
set declarations will be suggested defined in this scheme.  I think this
will be of interest to the TEI, too.
 
Note: Many ask me why I haven't published this, yet.  I don't like to
published half-finished work with errors and omissions, and I have so
far only published the fact that I'm working on this, in order to fill
in the picture when reinventions of the wheel are marketed as round.
I don't intend to publish this until ISO 10646-1 is published, which
will hopefully be this year.
 
Best regards,
</Erik>
--
Erik Naggum             |  ISO  8879 SGML     |      +47 295 0313
                        |  ISO 10744 HyTime   |
<[log in to unmask]>        |  ISO 10646 UCS      |      Memento, terrigena.
<[log in to unmask]>       |  ISO  9899 C        |      Memento, vita brevis.
 
------------------------------------------------------------------------
Newsgroups:
 comp.fonts,comp.protocols.iso,comp.os.os2.programmer,comp.os.ms-windows.program
 mer.misc,comp.os.misc,sci.lang
Path: enag
From: Erik Naggum <[log in to unmask]>
Organization: Department of Informatics, University of Oslo, Norway
Message-ID: <[log in to unmask]>
Date: 29 Sep 1992 00:06:10 +0100
References: <jon.717440434@ada> <[log in to unmask]>
Subject: Re: Character Sets (AGAIN)
Lines: 355
 
Harald Tveit Alvestrand <[log in to unmask]> writes:
|
|   <To avoid using up bandwidth: Erik Naggum does not like this document.
 
[[ RFC 1345, Keld Simonsen's published work. ]]
 
|   He says he is working on an alternative document that he likes, but has not
|   yet (as far as I know) published it.>
 
I think it would be fair to include the reasons I don't like it, rather
than imply that it's just a matter of "liking" a format or not: the
tables are generally unreadable, they're full of errors, and they can't
be debugged by inspection, so you can't even find the errors without
doing very time-consuming comparisons with the original material.  I
started doing this time-consuming work, but found it easier to go to the
original sources myself, and start over.  That's why I don't "like" RFC
1345.  All other users of it will also have to do this painstaking
checking all over, because the RFC's content can't be trusted.  (The
author has announced a new, improved edition, but again, we have to
trust it, since its correctness and accuracy is extremely hard to
inspect, even for character sets you know well by heart.)
 
Compare the following two definitions of ISO 8859-1 (ISO Latin 1):
 
From RFC 1345:
 
  &charset ISO_8859-1:1987
  &rem source: ECMA registry
  &alias iso-ir-100
  &g1esc x2d41 &g2esc x2e41 &g3esc x2f41
  &alias ISO_8859-1
  &alias ISO-8859-1
  &alias latin1
  &alias l1
  &alias IBM819
  &alias CP819
  &code 0
  NU SH SX EX ET EQ AK BL BS HT LF VT FF CR SO SI
  DL D1 D2 D3 D4 NK SY EB CN EM SB EC FS GS RS US
  SP ! " Nb DO % & ' ( ) * + , - . / 0 1 2 3 4 5 6 7 8 9 : ; < = > ?
  At A B C D E F G H I J K L M N O P Q R S T U V W X Y Z <( // )> '> _
  '! a b c d e f g h i j k l m n o p q r s t u v w x y z (! !! !) '? DT
  PA HO BH NH IN NL SA ES HS HJ VS PD PU RI S2 S3
  DC P1 P2 TS CC MW SG EG SS GC SC CI ST OC PM AC
  NS !I Ct Pd Cu Ye BB SE ': Co -a << NO -- Rg '-
  DG +- 2S 3S '' My PI .M ', 1S -o >> 14 12 34 ?I
  A! A' A> A? A: AA AE C, E! E' E> E: I! I' I> I:
  D- N? O! O' O> O? O: *X O/ U! U' U> U: Y' TH ss
  a! a' a> a? a: aa ae c, e! e' e> e: i! i' i> i:
  d- n? o! o' o> o? o: -: o/ u! u' u> u: y' th y:
 
(Note: this is really four different character sets, including two
control character sets, and two graphic character sets: ASCII, and the
right half of ISO 8859-1, which is known as ISO registration number 100,
and it's only this right half which has ISO 2022 escape code (g1esc)
"ESC 2/13 4/1" (or x2d41).  The complete ISO 2022 escape code sequence
is "ESC 2/0 4/3 ESC 2/1 4/0 ESC 2/8 4/2 ESC 2/2 4/3 ESC 2/13 4/1".  We
see that this RFC does a major disservice to the community by pretending
that an escape sequence is more than it is, and lists four character
sets where only one is actually identified (ISO #100).  It should also
be said that ISO #100 is intended to be used with ISO #6 (ASCII, or the
new IRV), but it is still identified as only 96 graphic characters, not
256 graphic and control characters.  It should also be noted that not
all control characters in C1 (PA, HO, BH...) had been standardized at
the time of publication of the RFC, and many are in fact withdrawn by
now.  In general, C1 control codes are used as escape sequences, too.
E.g., CSI (CI in the above) is uniformly used as "ESC 5/11", not as
"9/11".  ISO 10646 requires that escape sequences shall be used.)
 
Compare this with my encoding of the ISO 2375 Register of Character Sets
to Be Used with Escape Sequences (according to ISO 2022), of all four
character sets.  (The descriptions are according to ISO 8879 (SGML)
character set declarations, and weremainly intended for use with SGML,
but have greater utility than that.  Comments are surrounded by "--".)
 
-- "ISO Registration Number 1//CHARSET C0 Set of ISO 646//ESC 2/1 4/0" --
-- "ISO 646:1991//CHARSET C0 Set//ESC 2/1 4/0" --
BASESET "ISO 2022//CHARSET Empty C0 Set//ESC 2/1 7/14"
DESCSET
    0 -- 0000 --     1 -- NUL -- "NULL"
    1 -- 0001 --     1 -- SOH -- "START OF HEADING"
    2 -- 0002 --     1 -- STX -- "START OF TEXT"
    3 -- 0003 --     1 -- ETX -- "END OF TEXT"
    4 -- 0004 --     1 -- EOT -- "END OF TRANSMISSION"
    5 -- 0005 --     1 -- ENQ -- "ENQUIRY"
    6 -- 0006 --     1 -- ACK -- "ACKNOWLEDGE"
    7 -- 0007 --     1 -- BEL -- "BELL"
    8 -- 0008 --     1 -- BS  -- "BACKSPACE"
    9 -- 0009 --     1 -- HT  -- "HORIZONTAL TABULATION"
   10 -- 000A --     1 -- LF  -- "LINE FEED"
   11 -- 000B --     1 -- VT  -- "VERTICAL TABULATION"
   12 -- 000C --     1 -- FF  -- "FORM FEED"
   13 -- 000D --     1 -- CR  -- "CARRIAGE RETURN"
   14 -- 000E --     1 -- SO  -- "SHIFT OUT"
   15 -- 000F --     1 -- SI  -- "SHIFT IN"
   16 -- 0010 --     1 -- DLE -- "DATA LINK ESCAPE"
   17 -- 0011 --     1 -- DC1 -- "DEVICE CONTROL ONE"
   18 -- 0012 --     1 -- DC2 -- "DEVICE CONTROL TWO"
   19 -- 0013 --     1 -- DC3 -- "DEVICE CONTROL THREE"
   20 -- 0014 --     1 -- DC4 -- "DEVICE CONTROL FOUR"
   21 -- 0015 --     1 -- NAK -- "NEGATIVE ACKNOWLEDGE"
   22 -- 0016 --     1 -- SYN -- "SYNCRONOUS IDLE"
   23 -- 0017 --     1 -- ETB -- "END OF TRANSMISSION BLOCK"
   24 -- 0018 --     1 -- CAN -- "CANCEL"
   25 -- 0019 --     1 -- EM  -- "END OF MEDIUM"
   26 -- 001A --     1 -- SUB -- "SUBSTITUTE"
   27 -- 001B --     1 -- ESC -- "ESCAPE"
   28 -- 001C --     1 -- IS4 -- "INFORMATION SEPARATOR FOUR"
   29 -- 001D --     1 -- IS3 -- "INFORMATION SEPARATOR THREE"
   30 -- 001E --     1 -- IS2 -- "INFORMATION SEPARATOR TWO"
   31 -- 001F --     1 -- IS1 -- "INFORMATION SEPARATOR ONE"
 
-- "ISO Registration Number 6//CHARSET ISO 646:1991 IRV//ESC 2/8 4/2" --
-- "ISO 646:1991//CHARSET IRV//ESC 2/8 4/2" --
BASESET "ISO 2022//CHARSET Empty G0 Set//ESC 2/8 7/14"
DESCSET
   32 -- 0020 --     1 "SPACE"
   33 -- 0021 --     1 "EXCLAMATION MARK"
   34 -- 0022 --     1 "QUOTATION MARK"
   35 -- 0023 --     1 "NUMBER SIGN"
   36 -- 0024 --     1 "DOLLAR SIGN"
   37 -- 0025 --     1 "PERCENT SIGN"
   38 -- 0026 --     1 "AMPERSAND"
   39 -- 0027 --     1 "APOSTROPHE"
   40 -- 0028 --     1 "LEFT PARENTHESIS"
   41 -- 0029 --     1 "RIGHT PARENTHESIS"
   42 -- 002A --     1 "ASTERISK"
   43 -- 002B --     1 "PLUS SIGN"
   44 -- 002C --     1 "COMMA"
   45 -- 002D --     1 "HYPHEN-MINUS"
   46 -- 002E --     1 "PERIOD"
   47 -- 002F --     1 "SOLIDUS"
   48 -- 0030 --     1 "DIGIT ZERO"
   49 -- 0031 --     1 "DIGIT ONE"
   50 -- 0032 --     1 "DIGIT TWO"
   51 -- 0033 --     1 "DIGIT THREE"
   52 -- 0034 --     1 "DIGIT FOUR"
   53 -- 0035 --     1 "DIGIT FIVE"
   54 -- 0036 --     1 "DIGIT SIX"
   55 -- 0037 --     1 "DIGIT SEVEN"
   56 -- 0038 --     1 "DIGIT EIGHT"
   57 -- 0039 --     1 "DIGIT NINE"
   58 -- 003A --     1 "COLON"
   59 -- 003B --     1 "SEMICOLON"
   60 -- 003C --     1 "LESS-THAN SIGN"
   61 -- 003D --     1 "EQUALS SIGN"
   62 -- 003E --     1 "GREATER-THAN SIGN"
   63 -- 003F --     1 "QUESTION MARK"
   64 -- 0040 --     1 "COMMERCIAL AT"
   65 -- 0041 --     1 "LATIN CAPITAL LETTER A"
   66 -- 0042 --     1 "LATIN CAPITAL LETTER B"
   67 -- 0043 --     1 "LATIN CAPITAL LETTER C"
   68 -- 0044 --     1 "LATIN CAPITAL LETTER D"
   69 -- 0045 --     1 "LATIN CAPITAL LETTER E"
   70 -- 0046 --     1 "LATIN CAPITAL LETTER F"
   71 -- 0047 --     1 "LATIN CAPITAL LETTER G"
   72 -- 0048 --     1 "LATIN CAPITAL LETTER H"
   73 -- 0049 --     1 "LATIN CAPITAL LETTER I"
   74 -- 004A --     1 "LATIN CAPITAL LETTER J"
   75 -- 004B --     1 "LATIN CAPITAL LETTER K"
   76 -- 004C --     1 "LATIN CAPITAL LETTER L"
   77 -- 004D --     1 "LATIN CAPITAL LETTER M"
   78 -- 004E --     1 "LATIN CAPITAL LETTER N"
   79 -- 004F --     1 "LATIN CAPITAL LETTER O"
   80 -- 0050 --     1 "LATIN CAPITAL LETTER P"
   81 -- 0051 --     1 "LATIN CAPITAL LETTER Q"
   82 -- 0052 --     1 "LATIN CAPITAL LETTER R"
   83 -- 0053 --     1 "LATIN CAPITAL LETTER S"
   84 -- 0054 --     1 "LATIN CAPITAL LETTER T"
   85 -- 0055 --     1 "LATIN CAPITAL LETTER U"
   86 -- 0056 --     1 "LATIN CAPITAL LETTER V"
   87 -- 0057 --     1 "LATIN CAPITAL LETTER W"
   88 -- 0058 --     1 "LATIN CAPITAL LETTER X"
   89 -- 0059 --     1 "LATIN CAPITAL LETTER Y"
   90 -- 005A --     1 "LATIN CAPITAL LETTER Z"
   91 -- 005B --     1 "LEFT SQUARE BRACKET"
   92 -- 005C --     1 "REVERSE SOLIDUS"
   93 -- 005D --     1 "RIGHT SQUARE BRACKET"
   94 -- 005E --     1 "CIRCUMFLEX ACCENT"
   95 -- 005F --     1 "LOW LINE"
   96 -- 0060 --     1 "GRAVE ACCENT"
   97 -- 0061 --     1 "LATIN SMALL LETTER A"
   98 -- 0062 --     1 "LATIN SMALL LETTER B"
   99 -- 0063 --     1 "LATIN SMALL LETTER C"
  100 -- 0064 --     1 "LATIN SMALL LETTER D"
  101 -- 0065 --     1 "LATIN SMALL LETTER E"
  102 -- 0066 --     1 "LATIN SMALL LETTER F"
  103 -- 0067 --     1 "LATIN SMALL LETTER G"
  104 -- 0068 --     1 "LATIN SMALL LETTER H"
  105 -- 0069 --     1 "LATIN SMALL LETTER I"
  106 -- 006A --     1 "LATIN SMALL LETTER J"
  107 -- 006B --     1 "LATIN SMALL LETTER K"
  108 -- 006C --     1 "LATIN SMALL LETTER L"
  109 -- 006D --     1 "LATIN SMALL LETTER M"
  110 -- 006E --     1 "LATIN SMALL LETTER N"
  111 -- 006F --     1 "LATIN SMALL LETTER O"
  112 -- 0070 --     1 "LATIN SMALL LETTER P"
  113 -- 0071 --     1 "LATIN SMALL LETTER Q"
  114 -- 0072 --     1 "LATIN SMALL LETTER R"
  115 -- 0073 --     1 "LATIN SMALL LETTER S"
  116 -- 0074 --     1 "LATIN SMALL LETTER T"
  117 -- 0075 --     1 "LATIN SMALL LETTER U"
  118 -- 0076 --     1 "LATIN SMALL LETTER V"
  119 -- 0077 --     1 "LATIN SMALL LETTER W"
  120 -- 0078 --     1 "LATIN SMALL LETTER X"
  121 -- 0079 --     1 "LATIN SMALL LETTER Y"
  122 -- 007A --     1 "LATIN SMALL LETTER Z"
  123 -- 007B --     1 "LEFT CURLY BRACKET"
  124 -- 007C --     1 "VERTICAL LINE"
  125 -- 007D --     1 "RIGHT CURLY BRACKET"
  126 -- 007E --     1 "TILDE"
 
-- "ISO Registration Number 77//CHARSET C1 Control Set//ESC 2/2 4/3" --
-- "ISO 6429:1983//CHARSET C1 Control Set//ESC 2/2 4/3" --
BASESET "ISO 2022//CHARSET Empty C1 Set//ESC 2/2 7/14"
DESCSET
  128 -- 0080 --     4 UNUSED
  132 -- 0084 --     1 -- IND -- "INDEX"
  133 -- 0085 --     1 -- NEL -- "NEXT LINE"
  134 -- 0086 --     1 -- SSA -- "START OF SELECTED AREA"
  135 -- 0087 --     1 -- ESA -- "END OF SELECTED AREA"
  136 -- 0088 --     1 -- HTSD-- "CHARACTER TABULATION SET"
  137 -- 0089 --     1 -- HTJ -- "CHARACTER TABULATION WITH JUSTIFICATION"
  138 -- 008A --     1 -- VTS -- "LINE TABULATION SET"
  139 -- 008B --     1 -- PLD -- "PARTIAL LINE FORWARD"
  140 -- 008C --     1 -- PLU -- "PARTIAL LINE BACKWARD"
  141 -- 008D --     1 -- RI  -- "REVERSE LINE FEED"
  142 -- 008E --     1 -- SS2 -- "SINGLE-SHIFT TWO"
  143 -- 008F --     1 -- SS3 -- "SINGLE-SHIFT THREE"
  144 -- 0090 --     1 -- DCS -- "DEVICE CONTROL STRING"
  145 -- 0091 --     1 -- PU1 -- "PRIVATE USE ONE"
  146 -- 0092 --     1 -- PU2 -- "PRIVATE USE TWO"
  147 -- 0093 --     1 -- STS -- "SET TRANSMIT STATE"
  148 -- 0094 --     1 -- CCH -- "CANCEL CHARACTER"
  149 -- 0095 --     1 -- MW  -- "MESSAGE WAITING"
  150 -- 0096 --     1 -- SPA -- "START OF GUARDED AREA"
  151 -- 0097 --     1 -- EPA -- "END OF GUARDED AREA"
  152 -- 0098 --     3 UNUSED
  155 -- 009B --     1 -- CSI -- "CONTROL SEQUENCE INTRODUCER"
  156 -- 009C --     1 -- ST  -- "STRING TERMINATOR"
  157 -- 009D --     1 -- OSC -- "OPERATING SYSTEM COMMAND"
  158 -- 009E --     1 -- PM  -- "PRIVACY MESSAGE"
  159 -- 009F --     1 -- APC -- "APPLICATION PROGRAM COMMAND"
 
-- "ISO Registration Number 100//CHARSET Latin 1//ESC 2/13 4/1" --
-- "ISO 8859-1:1987//CHARSET Latin 1, right half//ESC 2/13 4/1" --
BASESET "ISO 2022//CHARSET Empty G1 Set//ESC 2/13 7/14"
DESCSET
   32 -- 0020 --     1 "NO-BREAK SPACE"
   33 -- 0021 --     1 "INVERTED EXCLAMATION MARK"
   34 -- 0022 --     1 "CENT SIGN"
   35 -- 0023 --     1 "POUND SIGN"
   36 -- 0024 --     1 "CURRENCY SIGN"
   37 -- 0025 --     1 "YEN SIGN"
   38 -- 0026 --     1 "BROKEN BAR"
   39 -- 0027 --     1 "SECTION SIGN"
   40 -- 0028 --     1 "DIAERESIS"
   41 -- 0029 --     1 "COPYRIGHT SIGN"
   42 -- 002A --     1 "FEMININE ORDINAL INDICATOR"
   43 -- 002B --     1 "LEFT-POINTING DOUBLE ANGLE QUOTATION MARK"
   44 -- 002C --     1 "NOT SIGN"
   45 -- 002D --     1 "SOFT HYPHEN"
   46 -- 002E --     1 "REGISTERED SIGN"
   47 -- 002F --     1 "OVERLINE"
   48 -- 0030 --     1 "DEGREE SIGN"
   49 -- 0031 --     1 "PLUS-MINUS SIGN"
   50 -- 0032 --     1 "SUPERSCRIPT DIGIT TWO"
   51 -- 0033 --     1 "SUPERSCRIPT DIGIT THREE"
   52 -- 0034 --     1 "ACUTE ACCENT"
   53 -- 0035 --     1 "MICRO SIGN"
   54 -- 0036 --     1 "PILCROW SIGN"
   55 -- 0037 --     1 "MIDDLE DOT"
   56 -- 0038 --     1 "CEDILLA"
   57 -- 0039 --     1 "SUPERSCRIPT DIGIT ONE"
   58 -- 003A --     1 "MASCULINE ORDINAL INDICATOR"
   59 -- 003B --     1 "RIGHT-POINTING DOUBLE ANGLE QUOTATION MARK"
   60 -- 003C --     1 "VULGAR FRACTION ONE QUARTER"
   61 -- 003D --     1 "VULGAR FRACTION ONE HALF"
   62 -- 003E --     1 "VULGAR FRACTION THREE QUARTERS"
   63 -- 003F --     1 "INVERTED QUESTION MARK"
   64 -- 0040 --     1 "LATIN CAPITAL LETTER A WITH GRAVE"
   65 -- 0041 --     1 "LATIN CAPITAL LETTER A WITH ACUTE"
   66 -- 0042 --     1 "LATIN CAPITAL LETTER A WITH CIRCUMFLEX"
   67 -- 0043 --     1 "LATIN CAPITAL LETTER A WITH TILDE"
   68 -- 0044 --     1 "LATIN CAPITAL LETTER A WITH DIAERESIS"
   69 -- 0045 --     1 "LATIN CAPITAL LETTER A WITH RING ABOVE"
   70 -- 0046 --     1 "LATIN CAPITAL LETTER AE"
   71 -- 0047 --     1 "LATIN CAPITAL LETTER C WITH CEDILLA"
   72 -- 0048 --     1 "LATIN CAPITAL LETTER E WITH GRAVE"
   73 -- 0049 --     1 "LATIN CAPITAL LETTER E WITH ACUTE"
   74 -- 004A --     1 "LATIN CAPITAL LETTER E WITH CIRCUMFLEX"
   75 -- 004B --     1 "LATIN CAPITAL LETTER E WITH DIAERESIS"
   76 -- 004C --     1 "LATIN CAPITAL LETTER I WITH GRAVE"
   77 -- 004D --     1 "LATIN CAPITAL LETTER I WITH ACUTE"
   78 -- 004E --     1 "LATIN CAPITAL LETTER I WITH CIRCUMFLEX"
   79 -- 004F --     1 "LATIN CAPITAL LETTER I WITH DIAERESIS"
   80 -- 0050 --     1 "LATIN CAPITAL LETTER ETH"
   81 -- 0051 --     1 "LATIN CAPITAL LETTER N WITH TILDE"
   82 -- 0052 --     1 "LATIN CAPITAL LETTER O WITH GRAVE"
   83 -- 0053 --     1 "LATIN CAPITAL LETTER O WITH ACUTE"
   84 -- 0054 --     1 "LATIN CAPITAL LETTER O WITH CIRCUMFLEX"
   85 -- 0055 --     1 "LATIN CAPITAL LETTER O WITH TILDE"
   86 -- 0056 --     1 "LATIN CAPITAL LETTER O WITH DIAERESIS"
   87 -- 0057 --     1 "MULTIPLICATION SIGN"
   88 -- 0058 --     1 "LATIN CAPITAL LETTER O WITH STROKE"
   89 -- 0059 --     1 "LATIN CAPITAL LETTER U WITH GRAVE"
   90 -- 005A --     1 "LATIN CAPITAL LETTER U WITH ACUTE"
   91 -- 005B --     1 "LATIN CAPITAL LETTER U WITH CIRCUMFLEX"
   92 -- 005C --     1 "LATIN CAPITAL LETTER U WITH DIAERESIS"
   93 -- 005D --     1 "LATIN CAPITAL LETTER Y WITH ACUTE"
   94 -- 005E --     1 "LATIN CAPITAL LETTER THORN"
   95 -- 005F --     1 "LATIN SMALL LETTER SHARP S"
   96 -- 0060 --     1 "LATIN SMALL LETTER A WITH GRAVE"
   97 -- 0061 --     1 "LATIN SMALL LETTER A WITH ACUTE"
   98 -- 0062 --     1 "LATIN SMALL LETTER A WITH CIRCUMFLEX"
   99 -- 0063 --     1 "LATIN SMALL LETTER A WITH TILDE"
  100 -- 0064 --     1 "LATIN SMALL LETTER A WITH DIAERESIS"
  101 -- 0065 --     1 "LATIN SMALL LETTER A WITH RING ABOVE"
  102 -- 0066 --     1 "LATIN SMALL LETTER AE"
  103 -- 0067 --     1 "LATIN SMALL LETTER C WITH CEDILLA"
  104 -- 0068 --     1 "LATIN SMALL LETTER E WITH GRAVE"
  105 -- 0069 --     1 "LATIN SMALL LETTER E WITH ACUTE"
  106 -- 006A --     1 "LATIN SMALL LETTER E WITH CIRCUMFLEX"
  107 -- 006B --     1 "LATIN SMALL LETTER E WITH DIAERESIS"
  108 -- 006C --     1 "LATIN SMALL LETTER I WITH GRAVE"
  109 -- 006D --     1 "LATIN SMALL LETTER I WITH ACUTE"
  110 -- 006E --     1 "LATIN SMALL LETTER I WITH CIRCUMFLEX"
  111 -- 006F --     1 "LATIN SMALL LETTER I WITH DIAERESIS"
  112 -- 0070 --     1 "LATIN SMALL LETTER ETH"
  113 -- 0071 --     1 "LATIN SMALL LETTER N WITH TILDE"
  114 -- 0072 --     1 "LATIN SMALL LETTER O WITH GRAVE"
  115 -- 0073 --     1 "LATIN SMALL LETTER O WITH ACUTE"
  116 -- 0074 --     1 "LATIN SMALL LETTER O WITH CIRCUMFLEX"
  117 -- 0075 --     1 "LATIN SMALL LETTER O WITH TILDE"
  118 -- 0076 --     1 "LATIN SMALL LETTER O WITH DIAERESIS"
  119 -- 0077 --     1 "DIVISION SIGN"
  120 -- 0078 --     1 "LATIN SMALL LETTER O WITH STROKE"
  121 -- 0079 --     1 "LATIN SMALL LETTER U WITH GRAVE"
  122 -- 007A --     1 "LATIN SMALL LETTER U WITH ACUTE"
  123 -- 007B --     1 "LATIN SMALL LETTER U WITH CIRCUMFLEX"
  124 -- 007C --     1 "LATIN SMALL LETTER U WITH DIAERESIS"
  125 -- 007D --     1 "LATIN SMALL LETTER Y WITH ACUTE"
  126 -- 007E --     1 "LATIN SMALL LETTER THORN"
  127 -- 007F --     1 "LATIN SMALL LETTER Y WITH DIAERESIS"
 
Slightly more verbose :-), but also easily debuggable: it's parsable,
and the character number is explicitly identified with the character
name.  (Missing characters and resulting "shifts" account for about 400
errors in RFC 1345.)  The names are drawn from ISO/IEC DIS 10646-1.2,
and will be updated to include the official names from the published
standard.  The fact that these are actually delimited strings also makes
it possible to construct conversion tables by name lookup, instead of
typing in (with errors) a pre-composed conversion table.
 
If this peaks your interest, please drop me a line.
 
Best regards,
</Erik>
--
Erik Naggum             |  ISO  8879 SGML     |      +47 295 0313
                        |  ISO 10744 HyTime   |
<[log in to unmask]>        |  ISO 10646 UCS      |      Memento, terrigena.
<[log in to unmask]>       |  ISO  9899 C        |      Memento, vita brevis.