Re: [hatari-devel] GEMDOS filename handling |
[ Thread Index |
Date Index
| More lists.tuxfamily.org/hatari-devel Archives
]
- To: hatari-devel@xxxxxxxxxxxxxxxxxxx
- Subject: Re: [hatari-devel] GEMDOS filename handling
- From: David Savinkoff <dsavnkff@xxxxxxxxx>
- Date: Sat, 30 Aug 2014 12:05:34 -0600 (MDT)
- Thread-index: ZMC0TC+CqVXrjbnRlfkI5xQzAvAJZQ==
- Thread-topic: GEMDOS filename handling
Hi,
What is the objective of GEMDOS filename handling?
Maybe a GEMDOS filename handling specification
should be made before pushing out code.
The specification should consider that Hatari has
hard drive and floppy disk image capabilities, and that
the GEMDOS directory is an Atari ST plug in cartridge.
Maybe NATFEATS should be considered.
Sincerely,
David Savinkoff
----- Max Böhm <mboehm3@xxxxxxxxx> wrote:
> Hi,
>
> 2014-08-29 22:58 GMT+02:00 Eero Tamminen <oak@xxxxxxxxxxxxxx>:
> > On lauantai 16 elokuu 2014, Max Böhm wrote:
> >> > I'm going to look into having correct TOS error codes /
> >> > mapping for invalid characters and doing conversions in
> >> > fewer places (higher in call hierarchy).
> >>
> >> ok. Some comments for consideration:
> >>
> >> 1. The OSX UTF-8 normalization calls (Str_DecomposedToPrecomposedUtf8)
> >> may be wrapped by #ifdefs __MACOSX__.
> >
> > Is there some reason why similar UTF-8 strings couldn't be encountered
> > on other operation systems?
>
> The only system I know which uses the decomposed form of UTF-8 is OSX.
> It won't hurt to call the Str_DecomposedToPrecomposedUtf8() function also
> for other systems which use UTF-8, but if the string encoding is not UTF-8
> then the function should not be called, as it replaces certain UTF-8 encoded
> sequences of letter + diacritical mark by its UTF-8 encoded precomposed
> unicode character representation, see
> http://en.wikipedia.org/wiki/Combining_character.
>
> - Max
>
>