On Sat, Apr 17, 2021 at 6:24 AM Willy Tarreau <w@xxxxxx> wrote: > > My concern was to know what field to look at to reliably detect an error > from the C side after a sequence doing C -> Rust -> C when the inner C > code uses NULL to mark an error and the upper C code uses NULL as a valid > value and needs to look at an error code instead to rebuild a result. But I see, thanks for clarifying. I don't think we want to change anything on either of the C sides (at least for the foreseeable future). So the Rust code in-between must respect whatever conventions both C sides already use, even if they happen to be different on each side. Thus the C side will not know there was a `Result` inside the Rust side and so it does not need to worry about which field to look at. Cheers, Miguel
