Update code to use newer rust versions and eliminate use of obsolete methods #33
Conversation
Signed-off-by: Michael Lodder <[email protected]>
Signed-off-by: Michael Lodder <[email protected]>
|
|
|
This is mostly a straight across fix to bring to 2018 edition. If you want, we can update it to something else but this shows that its better than |
|
I’m not sure where you get from that page that it’s better than
(UB = undefined behavior), which is calling out the latter pattern as equally bad, not better. |
There was a problem hiding this comment.
Hi, I think I just re-did a bunch of what you do in this PR here, perhaps we can merge/rebase and pick the best changes.
Sorry I didn't review this before doing mine :)
I left some comments inline, my main complaint is that this PR could have been done in a bunch of separate commits, there were a few different things going on in each one IIUC.
| @@ -81,7 +82,7 @@ impl Mpf { | |||
|
|
|||
| pub fn new(precision: usize) -> Mpf { | |||
| unsafe { | |||
| let mut mpf = uninitialized(); | |||
| let mut mpf = MaybeUninit::uninit().assume_init(); | |||
There was a problem hiding this comment.
I think the pattern we want is
let mut mpf = MaybeUninit::uninit();
unsafe {
__gmpf_init2(mpf.as_mut_ptr(), precision as c_ulong);
let mpf = mpf.assume_init();
Mpf { mpf }
}
| use super::sign::Sign; | ||
| use ffi::*; | ||
| use libc::{c_char, c_double, c_int, c_ulong}; | ||
| use crate::mpz::{mpz_struct, Mpz, mpz_ptr, mpz_srcptr}; |
| @@ -206,7 +206,7 @@ impl Error for ParseMpqError { | |||
| "invalid rational number" | |||
| } | |||
|
|
|||
| fn cause(&self) -> Option<&'static Error> { | |||
| fn cause(&self) -> Option<&'static dyn Error> { | |||
There was a problem hiding this comment.
yep, seems you were using Clippy to steer this too :)
| @@ -256,6 +259,19 @@ impl Mpz { | |||
| } | |||
| } | |||
|
|
|||
| pub fn div_rem(&self, other: &Mpz) -> (Mpz, Mpz) { | |||
There was a problem hiding this comment.
I can't work out why you add this? Was there a clippy warning after changing to 2018 that I didn't see?
There was a problem hiding this comment.
I use it for certain arithmetic operations where I get both results with one call
| Mpz::from_str_radix("1010010011", 2).unwrap().popcount() == 5; | ||
| assert_eq!(Mpz::from_str_radix("1010010011", 2).unwrap().popcount(), 5); |
There was a problem hiding this comment.
Oh nice, I did this one wrong. I fixed my change in line with this :)
No description provided.