3 ms·
The main problem for rust that comes to mind is the already huge codebase written in c. Personally I tend to use opencv and openssl a lot. I know there are rust
by accurrent 10y ago
The main problem for rust that comes to mind is the already huge codebase written in c. Personally I tend to use opencv and openssl a lot. I know there are rust wrappers around openssl and opencv it would be nice to see pure implementations. I think an application like tls would benefit from rust a lot.
There are other things like advanced static analyzers such as frama c which exist only for c. Similar tooling would be helpful for rust.
Then there's of course embedded where rust could really shine. However llvm doesn't have backends to many of the 8bit architectures. Furthermore industry standards such as misra c dont seem to have rust equivalents, although rust may be better suited for safety critical code.
Finally, explaining the need of a borrow checker to a beginner is very difficult unless they already have experienced enough pain with segfaults. It took me ten minutes with a pen and paper to explain pointers to my 12yo brother. How can I explain the need for all of rusts safety checks unless he has experience with segfaults. Any suggestions here would be welcome.
- steveklabnik 10y ago> the already huge codebase written in c. This is something you can work with!. A big example happening right now is the work going on in the GNOME project. > I think an application like tls would benefit from rust a lot. https://crates.io/crates/rustls https://crates.io/crates/rustls still has some C at its core, in ring, but it will be eventually. > Similar tooling would be helpful for rust. Rustc already does a lot of things this kind of tooling does. More tools are good though! 8 bit is far off. Teaching is hard. I don't have easy answers for those ones :)