Remove MutVecInput and MappedInput in Favour of Impls on References #2783
Conversation
riesentoaster
changed the title
| self.as_mut() | ||
| } | ||
|
|
||
| fn resize(&mut self, new_len: usize, value: u8) { |
There was a problem hiding this comment.
Doesn't this make using Vec a bit of a headache whenever HasMutatorBytes is in scope?
There was a problem hiding this comment.
If, in the same file, one has HasMutatorBytes in scope and uses any of the duplicate functions on a Vec<u8> specifically, one would need to specify which implementation to use. Since one just forwards to the other, you can't really choose wrong, it's just annoying to write the extra code.
| @@ -36,28 +36,28 @@ impl Input for CustomInput { | |||
|
|
|||
| impl CustomInput { | |||
| /// Returns a mutable reference to the byte array | |||
| pub fn byte_array_mut(&mut self) -> MutVecInput<'_> { | |||
| (&mut self.byte_array).into() | |||
| pub fn byte_array_mut(&mut self) -> &mut Vec<u8> { | |||
|
Amazing, think we're almost there now! :) Much clener than before |
|
Careful, there's a difference between the mutators themselves ( Edit: Ah, wait. I misread. I don't really like those names tbh, I think they should at least have the word |
|
Sounds good. Anything shorter is good :) |
Turns out implementing the required traits directly on the borrows is possible and much cleaner.
This also allows removing
MappedInput, which caused significant headaches for mapping mutators.