Bulat-Ziganshin
https://github.com/Bulat-Ziganshin/ECC-Benchmark I made ECC benchmark that includes mainly your libraries :) The reason is that I don't know better alternatives for them, so if anyone has better ideas, I'm all...
It will be great if Leopard can in runtime detect whether CPU supports AVX2/SSSE3/none and executes the appropriate code
65536 data+parity blocks limit is too low for some PAR3-style applications. I will be be glad to see it supported.
It's permanent topic for discussion and comparison of other ECC libraries. Fast GF(256) libraries with O(N2) speed: - [ISA-L](https://github.com/01org/isa-l): hands-down optimized asm code for SSSE3, AVX2 and AVX512. No support...
This topic is dedicated to construction of public API that the library should provide. I think, it should meet the following criteria, in the order of importance: - minimize memory...
Adding fazip functions is great because - Updates in my compression libs will bring faster to fazip users - In future, it allows to implement advanced features such as Lua...
Options -d, -df, as well as -ac, implemented via onSuccessfullyCompressed() callback
Ideally, it should be implemented via Virtual Lua I/O, that supports the following operations: - OpenInputArchive, OpenOutputArchive, OpenFileToCompress, OpenFileToExtract, OpenFile (this allows to check prior to operation which operations were...
-sctANSI by default is better for Windows. Also, -sct/l ANSI/OEM is useful for non-Unicode aware MultiArc plugins. JFYI: default codepages in arc/win: ``` haskell -- |Default charsets for various domains...
This option is critical for full MultiArc suppport. -dp is optional but will be nice too.