Big integers (Uint*) from byte slice array with from_be_bytes

[dr-orlovsky]

No description provided.

[elichai]

Could you please implement this without heap allocations?
also, you can add a slice_to_u64_be here and use that:

rust-bitcoin/src/util/endian.rs

Line 54 in f4e26ca

define_slice_to_be!(slice_to_u32_be, u32);

[stevenroose]

Hmm, it reads correctly, but I'd be happier with a unit test.

[dr-orlovsky]

also, you can add a slice_to_u64_be here and use that:

@elichai I did, but it fails to compile since this function is not being used from anywhere (i.e. triggers dead_code error):

error: function is never used: `slice_to_u64_be`
  --> src/util/endian.rs:4:9
   |
4  |         pub fn $name(slice: &[u8]) -> $type {
   |         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
...
55 | define_slice_to_be!(slice_to_u64_be, u64);
   | ------------------------------------------ in this macro invocation
   |
note: the lint level is defined here
  --> src/lib.rs:43:9
   |
43 | #![deny(dead_code)]
   |         ^^^^^^^^^
   = note: this error originates in a macro (in Nightly builds, run with -Z macro-backtrace for more info)

[elichai]

I meant that you'll use that instead of: u64::from_be_bytes(...) (for MSRV reasons)

[dr-orlovsky]

Oh yes, true

[dr-orlovsky]

done

[dr-orlovsky]

Now, with #435 merged, it will be nice to merge this one as well before the next version release

[sgeisler]

Looks good. I think it could be more readable/idiomatic if it used iterators.

[dr-orlovsky]

Btw, trying to understand how util::endianis different from https://github.com/rust-bitcoin/bitcoin_hashes/blob/master/src/util.rs#L149-L210 and why not to move there all these functions (old and new ones)? It looks like just a copy-past from bitcoin_hashes::util

Updt: created an issue for that #467 This can be fixed independently from this PR

[apoelstra]

utACK