[Freedos-devel] Freeware CHKDSK with FAT32 support

[Luchezar Georgiev]

Hello from Bulgaria, and big thanks to Alain for providing me with an
e-mail address for FreeDOS!

I'm back just to announce the availability of a freeware FAT32-enabled
CHKDSK written by some of our most gifted graduate students. It's at
http://linux.tu-varna.acad.bg/~lig/freedos/CHKDSK.ZIP

I has simple all-permitting MIT-style freeware license. Like ROMDSK, the
source code is commented in Bulgarian and will hardly be ever translated
to English. Unlike ROMDSK, I don't have permission to make it available.
And before you ask, graduate students can't support their work, nor can I
or anyone else, so your bug reports and wishes will most likely go to NUL
- sorry but I can't help...

Anyway, as the Italians say, "A caval donato non si guarda in bocca"
("Never look a gift horse in the mouth"), and also "O la va, o la spacca"
(both proverbs exist in Bulgarian too). So, don't complain, but spread and
enjoy instead!

Regards,
Lucho

[Jim Hall]

Hi Lucho!

Could you see about getting permission to release the source code for
this CHKDSK? It isn't much use to the FreeDOS community if we don't
have access to the source code to make modifications, improvements, etc.
Even if the code is commented in Bulgarian, I'm sure the code will be
readable by someone who would like to make improvements.

You say that the grad students can't support this version of CHKDSK -
the best way to continue improving the program is therefore to open up
the code.

Thanks.
-jh

Post by Luchezar Georgiev
Hello from Bulgaria, and big thanks to Alain for providing me with an
e-mail address for FreeDOS!
I'm back just to announce the availability of a freeware FAT32-enabled
CHKDSK written by some of our most gifted graduate students. It's at
http://linux.tu-varna.acad.bg/~lig/freedos/CHKDSK.ZIP
I has simple all-permitting MIT-style freeware license. Like ROMDSK, the
source code is commented in Bulgarian and will hardly be ever translated
to English. Unlike ROMDSK, I don't have permission to make it available.
And before you ask, graduate students can't support their work, nor can
I or anyone else, so your bug reports and wishes will most likely go to
NUL - sorry but I can't help...
Anyway, as the Italians say, "A caval donato non si guarda in bocca"
("Never look a gift horse in the mouth"), and also "O la va, o la
spacca" (both proverbs exist in Bulgarian too). So, don't complain, but
spread and enjoy instead!
Regards,
Lucho

[Luchezar Georgiev]

Hello Jim,

Post by Jim Hall
Could you see about getting permission to release the source code for
this CHKDSK?

I'll see what I can do, but chances are small. Our department decided that
all graduate thesises are its valuable intellectual property not to be
released in whole or in part. I wonder how it was possible to get
permission to release even the executable! There is strong influence of
Microsoft and even a specialisation called "Microsoft Information
Technologies" supplied with M$ materials. We've turned into brain
exporters and this is both a big shame and a sure sign that Bulgaria is
really on its death bed without young blood. Oh, sorry but it seems we
can't escape from politics!

Post by Jim Hall
It isn't much use to the FreeDOS community if we don't have access to
the source code to make modifications, improvements, etc.

Sure, but just a minor note - it may not be of much use to the
*developers* community, but I hope it *can* be to the *users* community,
even without a source code. After all, the goal of CHKDSK is to... check
disks and fix them (even with FAT32 :)

Perhaps I should have posted my announcement in the users mailing list
instead? I'll do it now.

Post by Jim Hall
Even if the code is commented in Bulgarian, I'm sure the code will be
readable by someone who would like to make improvements.

You obviously haven't seen our students' source code :) It's usually a
strange mix of Bulgarian and English :( Even if it was readable, the
assembly parts of it (it's written in 1/3 C and 2/3 Assembler) are
spaghetti-code and very difficult to understand. And that's not only
students. Take UDMA code for example. Its author Jack R. Ellis has
commented everything but it's still very hard to understand - that's one
of the reasons why there are no volunteers to continue its development.

Post by Jim Hall
You say that the grad students can't support this version of CHKDSK -
the best way to continue improving the program is therefore to open up
the code.

There is source code and source code. Some isn't understandable even by
its authors in 6 months :)

Regards,
Lucho

[Johnson Lam]

On Mon, 26 Jul 2004 21:11:06 +0300, you wrote:

Hi Lucho,

Glad to see you.

Post by Luchezar Georgiev
I'm back just to announce the availability of a freeware FAT32-enabled
CHKDSK written by some of our most gifted graduate students. It's at
http://linux.tu-varna.acad.bg/~lig/freedos/CHKDSK.ZIP
I has simple all-permitting MIT-style freeware license. Like ROMDSK, the
source code is commented in Bulgarian and will hardly be ever translated
to English. Unlike ROMDSK, I don't have permission to make it available.
And before you ask, graduate students can't support their work, nor can I
or anyone else, so your bug reports and wishes will most likely go to NUL
- sorry but I can't help...

Too bad! :-(

Post by Luchezar Georgiev
Anyway, as the Italians say, "A caval donato non si guarda in bocca"
("Never look a gift horse in the mouth"), and also "O la va, o la spacca"
(both proverbs exist in Bulgarian too). So, don't complain, but spread and
enjoy instead!

Still better than the existing one ... at least FAT32 support.
We need your comment, if you have time please join our discussion.

Thanks.


Rgds,
Johnson.

[Eric Auer]

Hi, I would like to comment that I am planning to improve the
DOS port of dosfstools DOSFSCK 2.10 next month, which should
give us an open source FAT12-FAT16-FAT32 filesystem checker
(including "drop" and "undelete" functionality for single files and
directories, but without surface scan). It has a very non-MS-ish
user interface, but it works. It can even give you choices how to
fix particular filesystem problems, and run a second pass with the
changes SIMULATED before writing them to disk. This could be improved
by wrapping everything into a SCANDISK-style user interface and a
function to store a binary change log to an "undo file" if the user
wants to create one.

I myself do not care for undo files or eye candy or surface scan,
so I recommend that the FreeDOS CHKDSK surface scan gets a progress
meter (like printf("%3.3d%% done (sector %d of %d)\r", ..., ..., ...);
being called every 100 sectors), although that only supports FAT12/FAT16.
By the way, FreeDOS DEFRAG bug: "just move away empty clusters" kind of
compression has broken progress bar, and if you start DEFRAG from > 80x25
mode, then the screen display will get messed up (DEFRAG switches to 80x25
but some parts of it still use the old geometry).

Whatever. I have experience with DJGPP, so I should be able to merge
my dosdosfsck-2.8-fat32, Imre's dosfsck-2.10, Lucho's patch for > 4 GB
support, and some other spices, creating a nice and tasty 2.10.x DOSFSCK.
Good enough for me to check AND fix FAT32 disks on 386+ systems, even if
the screen output differs from CHKDSK style.

Eric

[Alain]

I am VEEERY glad to hear this :)))

Alain

Post by Eric Auer
Hi, I would like to comment that I am planning to improve the
DOS port of dosfstools DOSFSCK 2.10 next month, which should
give us an open source FAT12-FAT16-FAT32 filesystem checker
This could be improved
by wrapping everything into a SCANDISK-style user interface and a
function to store a binary change log to an "undo file" if the user
wants to create one.

That is sugar. Not Needed. If the standard interface is 100% functional
it is THE most important thing.