SD card suddenly read-only

[Jonas Norlander]

2008/10/15 Emanoil Kotsev <deloptes at yahoo.com>:
> --- On Wed, 10/15/08, Jonas Norlander <jonorland at gmail.com> wrote:
>
>> From: Jonas Norlander <jonorland at gmail.com>
>> Subject: Re: SD card suddenly read-only
>> To: deloptes at yahoo.com, "Kubuntu Help and User Discussions" <kubuntu-users at lists.ubuntu.com>
>> Date: Wednesday, October 15, 2008, 1:02 PM
>> 2008/10/15 Emanoil Kotsev <deloptes at yahoo.com>:
>> > --- On Tue, 10/14/08, Dotan Cohen
>> <dotancohen at gmail.com> wrote:
>> >
>> >> From: Dotan Cohen <dotancohen at gmail.com>
>> >> Subject: Re: SD card suddenly read-only
>> >> To: "Kubuntu Help and User Discussions"
>> <kubuntu-users at lists.ubuntu.com>
>> >> Date: Tuesday, October 14, 2008, 4:27 PM
>> >> 2008/10/14 Jason Straight
>> >> <jason at jeetkunedomaster.net>:
>> >> > you can format it with linux as well - sudo
>> mkfs.vfat
>> >> -F32 /dev/[sdcard]
>> >> >
>> >> > or -F16 if you want fat16.
>> >> >
>> >>
>> >> Thanks. I've found that Windows machines are
>> picky
>> >> about USB storage
>> >> formated on Linux, especially when using UTF-8
>> filenames.
>> >>
>> >> --
>> >> Dotan Cohen
>> >>
>> >
>> > I don't remember that FAT has UTF support. There
>> for you specify it as mount option. The formatting has
>> nothing to do with it.
>> >
>> > Correct me if I'm wrong.
>> >
>> > regards
>> >
>>
>> I think you are wrong :-)
>>
>> If i remember right VFAT/FAT32 Long File Names can have up
>> to 255
>> UTF-16 codec characters. I don't know if you can have
>> Long file names
>> in FAT16 but I'm sure wikipedia has the answer.
>>
>> / Jonas
>
> Why should I then specify the codepage and iocharset as a mount option if it was all about UTF?

It's not all about UTF, FAT is an _old_ filesystem and it's still
needs to read and store the filenames in the "old" format to be
backward compatibility besides the vfat UTF-16 encoded names.

You need the iocharset mount option in order for non-ASCII characters
in file names to be interpreted properly and it should be the same as
the character set of your locale. So if you have a UTF-8 locale it
should be set to utf8

You need the codepage to do the conversion from long to short file
names and it should be set to the codepage number used under MS-DOS in
your country.


> besides utf is not unicode - the specification says unicode. Windows is talking also about unicode.
UTF (Unicode Transformation Format) is a variable-length character
encoding for Unicode. There are several ways to encode Unicode, to
name a few UTF-1, UTF-7, UTF-8 and UCS-4 there are all unicode just
different ways how to encode them.

> I think Dotans problem is useing default values which corresponds to western chars ... but of course I'm not sure about it.
> most file names are may be in the iso8859 table but having one in hebrew breaks his task ... as far as I understood
>
> regards
>

Yes i think so to. Perhaps someday all layers involving characters
will work with unicode and we dont have to care about all this
conversation.

/ Jonas