Three wishes

David Hopwood <>
Mon Jul 18 20:33:02 CEST 2005


Roger Price wrote:
> Wish 3: ISO latin 9
> 
> If you are using the ISO latin 1 character set and can see the Euro
> symbol, then you are not using ISO latin 1, but rather ISO latin 9 as
> defined by ISO/IEC 8859-15 [1].   For example, on my Dell laptop the
> following function call displays a Euro:
> 
> 3> io:format ("~c~n", [$\244]) .
> ¤
> ok
> 
> Confusion between the two is very common. I would like Erlang to be an ISO
> latin 9 language.

Please don't. If strings are to be in a single ASCII-compatible charset, it
should be UTF-8. (The difficulties in using a charset with variable-length
character encodings are typically overstated.) Better would be a thorough
review of the language and libraries to provide full Unicode support.

Specific reasons not to use ISO-Latin-9 are that it's not compatible with
ISO-Latin-1, and that in practice hardly anyone uses it. A more common,
mostly ISO-Latin-1-compatible charset that includes the € character is
Windows CP1252+euro (http://www.microsoft.com/globaldev/reference/sbcs/1252.mspx),
not that I'm suggesting using that either.


[Note that this email got encoded automatically in UTF-8 without me having
to do anything special. Besides €, here are a few other currency characters:
₡ ₢ ₥ ₦ ₨ ₩ ₪ ₫. Try doing that in Latin-9.]

-- 
David Hopwood <>





More information about the erlang-questions mailing list