You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
What exactly is the problem with storing a generic header in the WavPack file? The purpose of this is to allow very old versions of wvunpack to unpack WavPack files to valid WAVs (at one point it could not generate a header if one was not provided).
But more generally, if you force WavPack to not put on a generic header (as you have), the current version of wvunpack will simply generate the same exact header itself, so there's actually no practical difference. As far as I know, no other programs even look at those stored headers except to see if there's other RIFF metadata in there (which in this case there would not be).
Of course, it adds 80+ bytes to the file. Is that the issue?
BTW, the reason that the header is bigger than the standard 44-byte version is that it reserves space for the RF64 header which is required with 2+ GB files (which might not be known in advance if the source is stdin). I could improve this, but it's a little tricky because this must be updated in-place.
Yes, I would like to get rid of the extra 80 bytes. :)
Also, the description for -r option says "do not store the headers in the resulting WavPack file", but in the end, there is a header stored in the file, so that's a little bit misleading. I would very much like to have no headers at all when -r has been given. Maybe you could consider another option for that purpose?
If I compress a RIFF-WAV file, I end up with the header stored in WavPack file. This is expected:
When I add
-r
option, WavPack creates its own RIFF header, which is longer than the original one:I would expect
-r
didn't add any file wrappers at all. To achieve that, I removed this section from the source code:Now, the file is what I expected:
The text was updated successfully, but these errors were encountered: