Live data from Hacker News

Think twice before translating all C to Rust

agarriga.substack.com

1–5 of 5 posts

Re: Think twice before translating all C to Rust

#2
"Another downside is that Fil-C is not compatible with libraries in binary form, it requires a re-compile."

My gut tells me this is only going to work for a relatively small part of the existing C code in production. Recompiling old C with new compiler is fraught with risks.

Re: Think twice before translating all C to Rust

#3

"Another downside is that Fil-C is not compatible with libraries in binary form, it requires a re-compile." My gut tells me this is only going to work for a relatively small part of the existing C code in production. Recompiling old C with new compiler is fraught with risks.

Possible! But fixing this seems easier than re-writing in Rust with reasonable expectation that the software behaves the same?

Re: Think twice before translating all C to Rust

#4
Rust is clearly not stable in either toolchain or the lang, and any substantial component you make today will be an anachronism to the common Rust man 18 mos from now.

Migrating critical c code to an unstable toolchain is utter insanity, and that madness will come back to haunt everyone.

Re: Think twice before translating all C to Rust

#5
post #3

"Another downside is that Fil-C is not compatible with libraries in binary form, it requires a re-compile." My gut tells me this is only going to work for a relatively small part of the existing C code in production. Recompiling old C with new compiler is fraught with risks.

Possible! But fixing this seems easier than re-writing in Rust with reasonable expectation that the software behaves the same?

It depends. There's a lot of code out there old enough that reproducing is effectively lost knowledge. But also, there is some evidence that bugs are more common in new code, so rewriting that is a good idea.