Re: [libredwg] reed solomon

[Till Heuschmann]

Hi Mike,

I'am working on the R2007 reader which for now just extracts the encoded 
data and ignores the rs parity bytes.
So I can provide you some data and give you some hints how to read it.
Attached is the rs data found at offset 0x80 in a .dwg file. The data is
0x3d8 bytes long and contains 3 blocks of rs data. One block contains 
239 bytes of data and 16 parity bytes.
The blocks are interleaved with a factor of 3.

The given data starts like this:
B3 01 68 D2 58 01 70 67 51 9E 00 7F 26 01 01 80 CD F0 68 A5

...and so the first rs encoded block starts with:
B3  D2  70  9E  26  80

the rest of the data (0x3d8 - (3 * 0xFF) = 0xDB) are padded and not rs 
encoded.

If you need some more data feel free to ask. I will push a first version 
of the reader in the next weeks.

Till

Am 30.08.2010 08:02, schrieb address@hidden:
> hi all,
> I have started to research into this reed solomon stuff, it does not
> look that hard.
>
> http://en.wikipedia.org/wiki/Reed–Solomon_error_correction
> http://en.wikipedia.org/wiki/BCH_code
> http://en.wikipedia.org/wiki/Berlekamp-Massey_algorithm
> also MIT OCW :
> http://www.youtube.com/watch?v=Nnj9lHePqKM
>
> can you please provide me with a block of pure data that is in RS
> format and the encoding parameters. I am going to work on a decoder.
>
> mike
>
>

rsdata.dat
Description: Binary data