I've been reading through the forums and there seemed to be a few people requesting somekind of error recovery for the par files themselves.
Well if you think about it, why not use the Data section of the PAR index file as a Parity volumn FOR the PXX set. So basicly if you where missing one PXX you could recover it from the .PAR index file.
I know that section is supposed to be for comments, but why not just tag the data onto the end of the file? I mean you have a 'start offset for data' and a 'data size' (and in the case of the index file 'data' should read 'comment')soooo... it could actually work. The Parity checksums for the PXX set could be writen to the end of the .PAR file and still be backward compatable.
just an idea... i guess it wouldn't even require changing the PAR spec now that i think about it.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
We don't want yet another par file, if we did, we could just MAKE one from the par files. Of course even this would be pointless, since just posting another NORMAL par file would accomplish MORE, as it could restore one of the RARs that we want rather than restore a PAR that would restore one of the RARs that we want.....
What would be nice, is a low-overhead error correction system that could repair small errors without causing the poster any extra greif.
The idea of placing this in the comment section might be useful here, though that isn't at all what it was intended for. What might be better is to use one of the undefined flags to signify the presence of recovery info at the end of the file, after the parity data.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I've been reading through the forums and there seemed to be a few people requesting somekind of error recovery for the par files themselves.
Well if you think about it, why not use the Data section of the PAR index file as a Parity volumn FOR the PXX set. So basicly if you where missing one PXX you could recover it from the .PAR index file.
I know that section is supposed to be for comments, but why not just tag the data onto the end of the file? I mean you have a 'start offset for data' and a 'data size' (and in the case of the index file 'data' should read 'comment')soooo... it could actually work. The Parity checksums for the PXX set could be writen to the end of the .PAR file and still be backward compatable.
just an idea... i guess it wouldn't even require changing the PAR spec now that i think about it.
We don't want yet another par file, if we did, we could just MAKE one from the par files. Of course even this would be pointless, since just posting another NORMAL par file would accomplish MORE, as it could restore one of the RARs that we want rather than restore a PAR that would restore one of the RARs that we want.....
What would be nice, is a low-overhead error correction system that could repair small errors without causing the poster any extra greif.
The idea of placing this in the comment section might be useful here, though that isn't at all what it was intended for. What might be better is to use one of the undefined flags to signify the presence of recovery info at the end of the file, after the parity data.
Good point. Didn't think it all the way through.