wrapping add #9
wrapping add #9
Conversation
|
I'm surprised you managed to figure out the inner workings, as basically nothing is documented right now. Wrapping, checked, and overflowing arithmetic is something that I wanted to implement for a while, and was thinking about recently. I will take a look into this soon! I imagine it'll take some thinking on my end. |
Panics if range_size greater than inner::MAX. This only applies on signed integers. There might be a way to read bytes as unsigned, wrap add, then read bytes as signed. Using something like to_le_bytes and from_le_bytes.
|
The implementation panics if the number is signed and the range is too large. Ideally this could be fixed with methods like checked_add_unsigned but the Although there isn't an equivalent rem_euclid for unsigned arguments so maybe that technique won't work anyway. |
|
As to |
Still clarifying/checking wrapping add method works. It seems to work for all input I can come up with but since I haven't finished verifying it might be wrong.
Easyoakland
force-pushed
the
wrapping-methods
branch
from
a4b689a to
fba6c18
Compare
- It was actually much simpler than previously realized. - The tests might be overkill
|
@jhpratt I think I finished the no_panic implementation with a valid explanation why the algorithm works. |
|
Thanks, I'll check it out! Now you know why I was putting off implementing it myself 😅 |
|
Quick status update: Apologies for the delay! Between RustConf and some other things, I've been quite busy. I intend to look at this in the next day or two. |
|
Thanks for the changes! Everything LGTM. Merging. |
Adding methods for wrapping.
I don't know if there is a better way to implement these.
Currently only
wrapping_addis implemented but I can work on the others if this is the correct way to go about it.