I think Christian meant adding a compression abstraction library above zlib, bz2, zip, etc?
That's even worse, as we would be throwing away twenty five years of optimizations in the decompressor, much of which is written in assembly language for various architectures. Is he proposing we rewrite all of
I actually really meant this lol. I will say, I too even struggled getting b2 to pick up zlib on my system but that's not something that's impossible to solve. You just log into slack and complain in #boost until Peter solves your build problems for you. It's 2024 and the best solution we have for adding compression to a user's application is: "just learn zlib" or other similar C-like APIs. I didn't have anything particularly concrete in mind. I was just vocalizing my feelings. that?
This is not a sound engineering practice to me.
Well, get prepared to have your mind blown! https://github.com/zlib-ng/zlib-ng Seems like someone already beat us to the punch and went ahead and re-implemented all of zlib for some decent gains, it seems. - Christian