Mo :ferris: :tux: @ FOSDEM on Nostr: npub1cdxdq…ak3h7 Here are things I assume: - split has to search for the commas and ...
npub1cdxdqcxmelr9jjtcu2d6lccelefyfu3hy3sthlfg9q9pl58xndds9ak3h7 (npub1cdx…k3h7) Here are things I assume:
- split has to search for the commas and then give you the slice between two. This means that it has to check the characters in the slice too. I tried using split. But it made my parsing about twice slower. A single scan is more efficient, but not always practical.
- Why don't you have the state field of the hasher as a u64? You could save some castings that way.
- split has to search for the commas and then give you the slice between two. This means that it has to check the characters in the slice too. I tried using split. But it made my parsing about twice slower. A single scan is more efficient, but not always practical.
- Why don't you have the state field of the hasher as a u64? You could save some castings that way.