Chore: Buffered_Computation can deal with generic containers

[reneme]

This should be a transparent change for the most part but enables modern support for all sorts of containers without resorting to the C-style pointer APIs.

Allows passing all sorts of contiguous ranges to ::update() and ::process() as well as storing results of Buffered_Computation in user-defined contiguous containers.

The latter is most-useful for the strong types:

using MessageHash = Strong<std::vector<uint8_t>, struct MessageHash_>;

auto sha = HashFunction::create("SHA-256");
sha->update("some message");
auto message_hash = sha->final<MessageHash>();

[reneme]

Sorry, I should have marked this as a draft while I was still working on it.

We added another overload for ::process<>() that allows custom output containers (same as ::final<>()). And also there's a test now, that makes sure all this convenience is actually working (aka compiling).

[reneme]

I deliberately left the virtual methods (below) untouched to get the low hanging fruits first.

virtual void add_data(const uint8_t input[], size_t length) = 0;
virtual void final_result(uint8_t out[]) = 0;

Though, both are good candidates for std::span<>. Particularly with sub-classes being forced to trust their base class(s) and write into an unbounded array. 😨

What's your take on this? Should we take this on?

[codecov-commenter]

Codecov Report

Base: 88.08% // Head: 88.09% // Increases project coverage by +0.01% 🎉

Coverage data is based on head (c4582aa) compared to base (e1b3226).
Patch coverage: 98.30% of modified lines in pull request are covered.

📣 This organization is not using Codecov’s GitHub App Integration. We recommend you install it so Codecov can continue to function properly for your repositories. Learn more

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #3294      +/-   ##
==========================================
+ Coverage   88.08%   88.09%   +0.01%     
==========================================
  Files         610      611       +1     
  Lines       68481    68577      +96     
  Branches     6830     6832       +2     
==========================================
+ Hits        60322    60414      +92     
- Misses       5303     5310       +7     
+ Partials     2856     2853       -3     

Impacted Files	Coverage Δ
src/tests/test_bufcomp.cpp	98.30% <98.30%> (ø)
src/lib/pbkdf/argon2/argon2pwhash.cpp	69.69% <0.00%> (-9.10%)	⬇️
src/lib/utils/cpuid/cpuid_x86.cpp	54.05% <0.00%> (-0.62%)	⬇️
src/lib/math/numbertheory/numthry.cpp	83.81% <0.00%> (-0.58%)	⬇️
src/lib/tls/tls12/tls_record.cpp	91.80% <0.00%> (-0.41%)	⬇️
src/lib/block/aes/aes_ni/aes_ni.cpp	99.80% <0.00%> (-0.20%)	⬇️
src/lib/math/numbertheory/dsa_gen.cpp	89.28% <0.00%> (-0.19%)	⬇️
src/tests/test_hash.cpp	84.96% <0.00%> (-0.10%)	⬇️
src/tests/test_bigint.cpp	90.81% <0.00%> (-0.03%)	⬇️
src/lib/block/sm4/sm4.cpp	100.00% <0.00%> (ø)
... and 12 more

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report at Codecov.
📢 Do you have feedback about the report comment? Let us know in this issue.

[randombit]

What's your take on this? Should we take this on?

Sure, using span internally here would be a lot cleaner overall. But also a lot of work since you'll have to touch every hash/MAC. Those kind of changes tend to be a slog.

[reneme]

@randombit I didn't merge this yet because I pushed some API changes after you approved the PR. Most notably: c4582aa

[randombit]

👍