50
DARPA suggests turning legacy C code automatically into Rust • The Register
(www.theregister.com)
A community for high quality news and discussion around technological advancements and changes
Things that fit:
Things that don't fit
How are they going to tackle heavy use of the C preprocessor?
What about pointer arithmetic without decorating everything with "unsafe"? As the whole point is making the code safer.
Can't have code vulnerabilities if the code doesn't run. *taps forehead*
The point is to reduce the amount of work by doing the boring work automatically. Manually copying struct definitions, functions, etc etc would be very boring but error prone work which is kind of the worst combination.
If that's done you can start improving the codebase module by module.
I guess it will translate the majority of the code which does not contain unsafe parts, and leave the pointer manipulation for manual rewrite.
Or maybe it'll write the unsafe parts in unsafe blocks?
Then there's no point. Converting the code to Rust as-is will also transfer all the bugs.
There is. It's a starting point to start converting the code to safe Rust. Some obstacles are removed.
~~you can just wrap a whole C code into a macro~~
yeah, preprocessor is going to blow everything, I forgot how good it is. Maybe there's a way by preprocessing and then leaving it as the code, but that wouldn't be as useful and configurable
Rust also has powerful macros, are you sure that those can't mimic at least most uses of the preprocessor?
You're right, but:
There is the project c2rust which is being used to transpile c projects to rust. I didn't find any mentions on restrictions in this regard but I obviously agree it's a complicated issue.
I'm just not sure this is true. Maybe you and I just don't see that there is always an equivalent solution in rust macros even if it's going to be unconventional/unidiomatic use of rust macros?
Maybe so, but large part of where it will be impossible is reproducing bugs introduced by unsanitized macros in C /half joking