refactor: Roll our own hex utils #4470
Conversation
Veykril
force-pushed
the
veykril/push-kmnmkuvmsnzw
branch
2 times, most recently
from
db4b2f9
to
9811f5a
Compare
lib/src/hex_util.rs
Outdated
| @@ -14,7 +14,17 @@ | |||
|
|
|||
| #![allow(missing_docs)] | |||
|
|
|||
| fn to_reverse_hex_digit(b: u8) -> Option<u8> { | |||
| pub fn to_reverse_hex_digit(b: u8) -> u8 { | |||
There was a problem hiding this comment.
nit: might be simpler to do hex encoding by ourselves. It won't gain SIMD optimization, but the performance wouldn't matter.
There was a problem hiding this comment.
As in roll all the hex stuff ourselves and put it into this module? Fine by me
There was a problem hiding this comment.
Oh, I just meant to_reverse_hex(&str) -> Option<String> could be reimplemented as encode_reverse_hex(&[u8]) -> String.
It might be okay to self-host all hex encoding functions if that helps reduce dependency, though.
Veykril
force-pushed
the
veykril/push-kmnmkuvmsnzw
branch
3 times, most recently
from
233f3cf
to
b746a32
Compare
Veykril
changed the title
hex crate with faster_hex
Veykril
force-pushed
the
veykril/push-kmnmkuvmsnzw
branch
3 times, most recently
from
0f5419f
to
e699f90
Compare
`hex` does not seem to be maintained any longer and we already transitively depend on `faster_hex` anyways.
Veykril
force-pushed
the
veykril/push-kmnmkuvmsnzw
branch
from
e699f90
to
0d2a154
Compare
|
Why is CI running nightly rustfmt 😕 |
The current state of the repository seems compatible with |
Veykril
force-pushed
the
veykril/push-kmnmkuvmsnzw
branch
from
0d2a154
to
0ef4d22
Compare
|
Ah wait, this is rustfmt working differently depending on how its invoked again I think (bare |
hexdoes not seem to be maintained any longer and we already transitively depend onfaster_hexanyways.Checklist
If applicable:
CHANGELOG.md