Conversation

Notices

1. Embed this notice
   Dan Drake 🦆 (ddrake@mathstodon.xyz)'s status on Tuesday, 11-Nov-2025 05:07:59 JST Dan Drake 🦆

   I used to teach at KAIST and this is a great development from their CS folks -- provably correct translation from C to Rust.

   https://news.kaist.ac.kr/newsen/html/news/?mode=V&mng_no=54270

   #kaist #c #rust

   • Embed this notice
     Paul Cantrell (inthehands@hachyderm.io)'s status on Tuesday, 11-Nov-2025 05:09:25 JST Paul Cantrell

     in reply to

     • DougMerritt (log😅 = 💧log😄)
     • Benjamin Geer

     @dougmerritt @benjamingeer @ddrake
     That, and:

     Refactoring semi-idiomatic Rust → idiomatic Rust is surely a far less error-prone trip than translating C → Rust.

   • Embed this notice
     DougMerritt (log😅 = 💧log😄) (dougmerritt@mathstodon.xyz)'s status on Tuesday, 11-Nov-2025 05:09:27 JST DougMerritt (log😅 = 💧log😄)

     in reply to

     • Benjamin Geer

     @benjamingeer @ddrake
     I haven't read it yet, but the news article points to the open access paper; at first blush they appear to care about idiomatic Rust :

     Automatically Translating C to Rust
     Automatic C-to-Rust translation tools are helpful, but they produce unsafe and unidiomatic code. What can be done to address these issues?
     Authors: Jaemin Hong, Sukyoung RyuAuthors Info & Claims
     Communications of the ACM, Volume 68, Issue 11
     https://dl.acm.org/doi/10.1145/3737696

   • Embed this notice
     Benjamin Geer (benjamingeer@piaille.fr)'s status on Tuesday, 11-Nov-2025 05:09:28 JST Benjamin Geer

     in reply to

     @ddrake This is very cool, but I wonder what the Rust code looks like: is it idiomatic Rust, which in many cases would require drastically changing the design of the original code? Is it something that a human would find maintainable? And (considering that it’s entirely possible to write unsafe code in Rust), is it provably safer than the original code? If not, maybe it would be better to pay human programmers to rewrite C code in Rust.