2005-03-29 20:34:00

by Xuân Baldauf

[permalink] [raw]
Subject: vfat: why is shortname=lower the default?

Hi Hirofumi,

Why is shortname=lower the default mount option for vfat filesystems?
Because, with "shortname=lower", copying one FAT32 filesystem tree to
another FAT32 filesystem tree using Liux results in semantically
different filesystems. (E.g.: Filenames which were once "all uppercase"
are now "all lowercase").

With "shortname=mixed", such semantic changes would not occur. That's
why I'd consider "shortname=lower" as default mount option as a bug. I'd
propose "shortname=mixed" as default as a fix for this bug.

What do you think?

ciao,
Xu?n.


P.S.: "man mount" says:

shortname=[lower|win95|winnt|mixed]

Defines the behaviour for creation and display of
filenames which fit into 8.3 characters. If a long name for a file
exists, it will always be preferred display. There are four modes:

lower Force the short name to lower case upon display;
store a long name when the short name is not all upper case.

win95 Force the short name to upper case upon display;
store a long name when the short name is not all upper case.

winnt Display the shortname as is; store a long name when
the short name is not all lower case or all upper case.

mixed Display the short name as is; store a long name
when the short name is not all upper case.

The default is "lower".


2005-03-30 13:12:27

by OGAWA Hirofumi

[permalink] [raw]
Subject: Re: vfat: why is shortname=lower the default?

Xu?n Baldauf <[email protected]> writes:

> Why is shortname=lower the default mount option for vfat filesystems?
> Because, with "shortname=lower", copying one FAT32 filesystem tree to
> another FAT32 filesystem tree using Liux results in semantically
> different filesystems. (E.g.: Filenames which were once "all
> uppercase" are now "all lowercase").

The reason is only it's very long-standing behavior. When this
behavior was changed before, it seems an one user was confused at
least.

http://marc.theaimsgroup.com/?t=97041869500002&r=1&w=2

Personally I agree that "winnt" or "mixed" is proper.

However, if we want to change the default behavior, it would need to
be tested for some months, and if anyone has no objection it can
change I think.

Thanks.
--
OGAWA Hirofumi <[email protected]>

2005-03-30 13:26:12

by Xuân Baldauf

[permalink] [raw]
Subject: Re: vfat: why is shortname=lower the default?

OGAWA Hirofumi wrote:

>Xu?n Baldauf <[email protected]> writes:
>
>
>
>>Why is shortname=lower the default mount option for vfat filesystems?
>>Because, with "shortname=lower", copying one FAT32 filesystem tree to
>>another FAT32 filesystem tree using Liux results in semantically
>>different filesystems. (E.g.: Filenames which were once "all
>>uppercase" are now "all lowercase").
>>
>>
>
>The reason is only it's very long-standing behavior. When this
>behavior was changed before, it seems an one user was confused at
>least.
>
> http://marc.theaimsgroup.com/?t=97041869500002&r=1&w=2
>
>Personally I agree that "winnt" or "mixed" is proper.
>
>However, if we want to change the default behavior, it would need to
>be tested for some months, and if anyone has no objection it can
>change I think.
>
>
One could make a slow transition, starting now with a warning like
"vfat: warning: You are using "shortname=lower" as default. This may not
be what you want. This default will change to "shortname=mixed" after
2005-07-01." if the shortname behaviour is not explicitly selected.

>Thanks.
>
>
ciao,
Xu?n. :-)

P.S.: I'm now trying to recover about 4M files from "lost case"...

2005-03-30 14:38:59

by OGAWA Hirofumi

[permalink] [raw]
Subject: Re: vfat: why is shortname=lower the default?

Xu?n Baldauf <[email protected]> writes:

> One could make a slow transition, starting now with a warning like
> "vfat: warning: You are using "shortname=lower" as default. This may
> not be what you want. This default will change to "shortname=mixed"
> after 2005-07-01." if the shortname behaviour is not explicitly
> selected.

Yes. But it is easy ignored. So, maybe peoples doesn't complain until
it is changed in fact.

Probably you can post the patch for -mm tree.

And it is tested in -mm tree. And at starting of dev-cycle of 2.6.13
or 2.6.14, it will be merged to Linus's tree if anyone has no objection.
--
OGAWA Hirofumi <[email protected]>