Moritz Borcherding
Moritz Borcherding
I'm not sure what you are tying to accomplish here. To count the decoded bytes you can customize the source which is already generic over the `Read` trait. The reason...
Very exciting to see some activity in this direction! I realize this is pretty early but I have some experience in efficient encoding/decoding stuff: I see you have a few...
That is definitely interesting! If you need any help I'd be glad to assist. These kinds of differences might indicate a bug
That's great!
I'll take a look this weekend :)
I think I found the error. Your code seems to ignore this bit of the standard (https://github.com/facebook/zstd/blob/dev/doc/zstd_compression_format.md#frame_content_size): ``` Frame_Content_Size format is little-endian. When FCS_Field_Size is 1, 4 or 8 bytes,...
yeah zstd has some weird edges like this to save a few more bytes here and there
Very nice!
Just one thing that came to my mind: I think it would be pretty cool to expand the interop fuzzer to additionally use this encoder
I'm not sure what error handling you mean? If you mean the error from creating the frame descriptor I think you should check the descriptor before creating the encoder so...