Re: [Lcms-user] Using cmsMLUsetWide()

[Richard H. <hug...@gm...>]

On 20 March 2013 14:31, Marti Maria <mar...@li...> wrote:
> ICC have this note: http://color.org/unicode.xalter
> Which is very scarce.

Right, agreed.

> Regarding serbian latin, sorry I have no idea. This is beyond lcms scope and
> probably have to do with ICC spec itself. You could contact ICC by using the  "ask
> Phil" mail as...@co...

I'll do that, thanks. I did think it rather odd that the codes were
cmsUInt16Number's and that we pass a char char[3] -- is the idea the
last char is '\0'?

Note, the LCMS docs state: "The localization trick is done by using
the lenguage and country codes, which  you  are  supposed  to  supply.
 Those  are  two  or  three  ASCII  letters." -- but surely using a
three letter code is going to be truncated?

> Instead of cmsGetProfileInfo(), which is a simplified interface, use
> cmsReadTag().
> This will return a cmsMLU object with all strings in all locales.

Yes, this is what I was doing, but I think we need something like the
attached (untested) patch. If this looks okay to you, I'm happy to
wire up all the other bits and add some unit tests. Or, feel free to
take my patch, munge it, and apply something completely different :)

Thanks,

Richard.