RE: [bug-gnu-libiconv] iconv bug: LATIN-9

[Nellis, Kenneth]

Bruno,

I appreciate your informative response.

An alternative you might consider is to add LATIN9 as an alias for LATIN-9. 
That would seem to satisfy the original intent (without the hyphen, according 
to the references you supplied), as well as the standard (with the hyphen).

Ken 

—
Kenneth Nellis
Systems Development Principal
Transportation Management Solutions
Affiliated Computer Services, Inc. 
A Xerox Company
7160 Riverwood Drive
Columbia, MD  21046

Tel: 443-259-7222

> -----Original Message-----
> From: Bruno Haible [mailto:address@hidden
> Sent: Saturday, June 26, 2010 07:53
> To: address@hidden
> Cc: Nellis, Kenneth
> Subject: Re: [bug-gnu-libiconv] iconv bug: LATIN-9
> 
> Hi,
> 
> Kenneth Nellis wrote:
> > In iconv when listing all supported encodings (iconv -l), LATIN9
> appears
> > with a hyphen (LATIN-9), but none of the other LATINs do.
> 
> Yes. This is how the names of the encodings have been standardized by
> the
> IANA (Internet Assigned Numbers Authority), see
>   http://www.iana.org/assignments/character-sets
> 
> $ grep -i '^Alias: latin' character-sets
> Alias: latin1
> Alias: latin2
> Alias: latin3
> Alias: latin4
> Alias: latin5
> Alias: latin6
> Alias: latin1-2-5
> Alias: latin8
> Alias: Latin-9
> Alias: latin10
> 
> > I would think consistency would be desired.
> 
> After reading the arguments in the discussion that started at [1],[2],
> in 1998,
> I only see "latin9" mentioned. I don't see where "Latin-9" comes from.
> Maybe
> it was unintended. But that's the way it is standardized for 12 years
> now.
> 
> Bruno
> 
> [1] http://mail.apps.ietf.org/ietf/charsets/msg00466.html
> [2] http://mail.apps.ietf.org/ietf/charsets/msg00475.html