Rename Word::zero to Word::zero_expr

[ChihChengLiang]

Description

Address #1746 (comment), we rename the constructors for zero word and one word.

Type of change

• Refactor

Contents

• Rename Word<Expression>::zero to zero_expr
• Rename Word<Expression>::one to one_expr
• Rename Word::zero_f to one
• Rename Word::one_f to one

[hero78119]

Another thoughts: how about define a trait with one(), zero() and let implement it for for some basic types ? Idea is to keep succinct Word::one() Word:zero() and let type system infer for us

[KimiWu123]

LGTM.

[KimiWu123]

Another thoughts: how about define a trait with one(), zero() and let implement it for for some basic types ? Idea is to keep succinct Word::one() Word:zero() and let type system infer for us

I tried to to have a trait for zero() and one(), but the compiler complained something like one() has been implemented.... Probably, there are some tricks to fix the error but I'm not familiar with it. Maybe you could give it a try!

[hero78119]

Another thoughts: how about define a trait with one(), zero() and let implement it for for some basic types ? Idea is to keep succinct Word::one() Word:zero() and let type system infer for us

I tried to to have a trait for zero() and one(), but the compiler complained something like one() has been implemented.... Probably, there are some tricks to fix the error but I'm not familiar with it. Maybe you could give it a try!

I tried to implement it in another PR and seems to work #1754

If new PR is acceptable I think we can close this one.

[ChihChengLiang]

@hero78119 Nice approach on #1754. I built on top of #1754 to get #1758, which removes the need to type hint the word.

@hero78119 @KimiWu123 Can you share your opinion on #1758?

[hero78119]

@hero78119 Nice approach on #1754. I built on top of #1754 to get #1758, which removes the need to type hint the word.

@hero78119 @KimiWu123 Can you share your opinion on #1758?

Just different topic

For unifying compress_f() to compress(), due to it got different implementation on different generic type, so it's can't be resolve like the same way. We might require feature like

• autoref-specialization => honestly I dislike the reference appear in syntax but it works on stable rusts
• specialization => still unstable feature

Maybe we keep compress_f() first and left it to future work

[ChihChengLiang]

Close in favor of #1758