Do i know somehow with good C knowledge?
I searching for features of C versions that are originated from C++.
I know C23 got [[nodiscard]], but I'm searching also for
C99
C11
C17
(reposting is appreciate)
@DanielaKEngert@sickeroni@erisceleste Function prototypes (aka void f(int a); that actually used types) was indeed something C stole from C++. I had a highly upvoted tweet with the screenshot and linked source but I've since cleaned out my Twitter so I'd have to find it again, but I'm willing to stake my Project Editor hat on it.
@sickeroni (They were standardized in-parallel but the motion of the feature was "C++ did it, then someone (P.J. Plauger/Hans Boehm) translated it into a C interface".
@DanielaKEngert@sickeroni@erisceleste FWIW I believe it's words from Brian Kernighan reflecting directly on C itself, but it's hard to find again since I think he's cleaned up his homepage. Plus Dennis's was turned into a memorial some time ago, so it's hard to find their writing on the C programming language and its earliest changes/standardization.
@whitequark@Patricia Haskell is just straight up readonly to me. if I wrote Haskell call my friends and tell them to come look for me because something must've happened.
@krans Hey, that's fine! I appreciate that you even took the time to read it at all; helped me remember to make a bunch more edits before sending it over again to be published!! 🫡
@hackymix Some feedback directly from implementers that implement their own stdlib on tiny devices, saying the interface was okay (and that it was nice that it didn't require any static data stored in the stdlib for the conversions). But otherwise, no other outreach.
Not sure why customization is a problem: the paper never got rid of the char/wchar_t bits. Those are implementation-defined, runtime-changed execution and wide execution encoding-based conversion routines, of which the implementation can stuff full of things (or perhaps nothing) that suits its specific needs.
@doty I'm not actually sure why it's not updating on the website itself, the local version seems to generate a proper feed.xml fine. I'll have to look into it...!
@gob I haven't taken a full opinion on it, because it's outside my wheel house. I am not an Optimizer Guy™, I just know enough to barely be able to not blow my own leg off (and measure when I think I might).
There's people working hard on the Memory Model and Provenance Models for C and C++ right now. Hell, someone just put out a paper about it recently: https://wg21.link/P3292R0
Me, ignorant: "Going over CVEs is a great way to learn what exploited weaknesses C has and the things human beings truly struggle with that are at risk."
Linux kernel, snorting a fat line: "Weeeeeoohaaaaa!! What if we turned every fix commit into its own CVE using grep???"