Gedanken wrote:
the content returned by my agent always fail to find the euro symbol.

As far as encoding is concerned, I can says this about the euro symbol (monetary symbol for Europe since 1/1/2002), for 8bit and wide chars:

latin1 (160-255)
----------------
The defacto std for 8bit chars (replacing 7bit ASCII) is latin1
(ie ISO-8859-1), to accept west-european languages chars,
in codes 160 to 255 (0xA0 to 0xFF, 240-777 in octal).
$ man latin1

cp1252 (128-255)
----------------
NB In latin1, chars 128-159 are invalid. In HTML4, these chars
are allowed, taken from windows charset cp1252. See section 24.4 on
http://www.w3.org/TR/1998/REC-html40-19980424/sgml/entities.html

latin1 currency (164)
---------------------
But the euro symbol is not in latin1. The latin1 character code
164 (0xA4, octal 244) is "CURRENCY SIGN", looking like a little sun.
http://www.w3.org/TR/REC-html32#charset

latin0 aka ISO-8859-15 euro (164)
---------------------------------
So a special latin0 charset has been created to handel it (ISO-8859-15).
It (re)defined that char 164 (0xA4, octal 244) to be "EURO SIGN"
(and (re)defined some other chars too)
$ man iso_8859-15

cp1252 euro (128)
-----------------
In windows-specific encoding windows cp1252, the char for euro is
"EURO SIGN", code 128 (0x80, octal 200 --unicode 8364)
http://www.kostis.net/charsets/cp1252.htm

HTML € (€)
---------------------
The '€' sgml entity is defined since HTML4 as
<!ENTITY euro   CDATA "&#8364;"  -- euro sign, U+20AC NEW -->
See http://www.w3.org/TR/1998/REC-html40-19980424/sgml/entities.html

Unicode euro (20AC, 20A0)
-------------------------
In the wide char world, Unicode (ISO/IEC 10646) can represent it
20A0;EURO-CURRENCY
20AC;EURO SIGN
http://www.unicode.org/Public/UNIDATA/UnicodeData.txt

Hope this helps, even if it sounds complicated.

Christophe



Reply via email to