Hi Vic,
On Tue, Mar 15, 2011 at 11:40 PM, Vic Lee <ll...@163.com> wrote:
> Hi Marc,
>
> Regarding uppercase, just think of one special situation in Turkish. In
> English, the letter 'i' in uppercase is 'I', as everyone knows. However
> in Turkish, 'i' in uppercase is 'İ', and 'ı' in uppercase is 'I'.
>
> So this uppercase conversion is locale-related, not just a one-one
> mapping table. This actually gives a lot of trouble for many systems
> that did not take care of the situation when used in Turkish-spoken
> countries.
>
Yeah, I was reading about unicode uppercasing, and I saw turkish being
mentioned:
http://www.unicode.org/faq/casemap_charprop.html
<http://www.unicode.org/faq/casemap_charprop.html>Also, uppercasing is not
always a 1:1 relationship, which further complicates things. However, the
unicode consortium provides files with maps for uppercasing.
http://www.unicode.org/Public/UNIDATA/UnicodeData.txt (1:1 uppercase
relationships)
http://www.unicode.org/Public/UNIDATA/SpecialCasing.txt (1:1+ uppercase
relationships)
Otherwise there is this file which appears to be a combination of
UnicodeData.txt and SpecialCasing.txt:
http://www.unicode.org/Public/UNIDATA/CaseFolding.txt
I'm looking through it... with that data I can probably manage to build
something good enough
>
> Vic
>
> On 03/16/2011 11:18 AM, Marc-André Moreau wrote:
> > Hi FreeRDP developers,
> >
> > I have just moved more stuff to the new libfreerdputils library,
> > including datablob.c, mem.c and unicode utilities. What I did finally
> > for the unicode utils was to wrap iconv under an "UNICONV" structure
> > that can be initialized anywhere and then passed back to the function.
> > This way, it's easy for any module to get its own handle to a unicode
> > converter.
> >
> > I am currently working on getting NLA to work with unicode characters.
> > It should now use the new unicode utils, but NLA will still fail, and I
> > just realized why: NTLMSSP requires the uppercasing of characters as
> > part of its cryptographic hashing functions. My previous uppercasing was
> > based on the fact that it's trivial in ASCII to check for a character
> > between 'a' and 'z' and change it to a character between 'A' and 'Z',
> > but that doesn't hold in unicode: there are tons of letters from many
> > different languages that are not in US ASCII that have uppercase
> > equivalents.
> >
> > From my understanding, it is possible to have usernames and passwords
> > in just about any language supported by unicode. We can probably make a
> > conversion table to cover most "special" letters but properly covering
> > all of them would not be trivial. Does anyone know of an actual good
> > method of uppercasing unicode characters?
>
>
>
> ------------------------------------------------------------------------------
> Colocation vs. Managed Hosting
> A question and answer guide to determining the best fit
> for your organization - today and in the future.
> http://p.sf.net/sfu/internap-sfd2d
> _______________________________________________
> Freerdp-devel mailing list
> Freerdp-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/freerdp-devel
>
------------------------------------------------------------------------------
Colocation vs. Managed Hosting
A question and answer guide to determining the best fit
for your organization - today and in the future.
http://p.sf.net/sfu/internap-sfd2d
_______________________________________________
Freerdp-devel mailing list
Freerdp-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/freerdp-devel
