std::span for Botan::Cipher_Mode

[reneme]

Two things:

• allow Cipher_Mode::process() and Cipher_Mode::start() for arbitrary contiguous containers
• allow concepts::resizable_byte_buffer for the in/out parameters of Cipher_Mode::update() and Cipher_Mode::finish().

To do that properly a "virtual entrypoint" indirection was introduced. Namely Cipher_Mode::process_msg() and Cipher_Mode::finish_msg().

Note that Cipher_Mode::finish() with anything but Botan::secure_vector<uint8_t> will introduce additional data copies. We might be able to address this later on (by generalizing the virtual entrypoint for Cipher_Mode). But that felt entirely out of scope here. FWIW: Users that had their data stored in some other container were previously forced to do the same copy ceremony before in their application code. That's a hot take, maybe. I'm open for better approaches.

[codecov-commenter]

Codecov Report

Patch coverage: 100.00% and no project coverage change.

Comparison is base (eb51e40) 88.13% compared to head (9f63da9) 88.14%.

📣 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    #3392   +/-   ##
=======================================
  Coverage   88.13%   88.14%           
=======================================
  Files         617      617           
  Lines       70358    70350    -8     
  Branches     6984     6987    +3     
=======================================
- Hits        62011    62010    -1     
+ Misses       5402     5398    -4     
+ Partials     2945     2942    -3     

Impacted Files	Coverage Δ
src/lib/modes/aead/ccm/ccm.cpp	90.37% <100.00%> (ø)
...b/modes/aead/chacha20poly1305/chacha20poly1305.cpp	94.31% <100.00%> (ø)
src/lib/modes/aead/eax/eax.cpp	97.72% <100.00%> (ø)
src/lib/modes/aead/gcm/gcm.cpp	92.77% <100.00%> (ø)
src/lib/modes/aead/ocb/ocb.cpp	98.38% <100.00%> (ø)
src/lib/modes/aead/siv/siv.cpp	93.13% <100.00%> (ø)
src/lib/modes/cbc/cbc.cpp	92.10% <100.00%> (ø)
src/lib/modes/cfb/cfb.cpp	95.49% <100.00%> (ø)
src/lib/modes/xts/xts.cpp	98.36% <100.00%> (ø)
src/lib/tls/tls12/tls_cbc/tls_cbc.cpp	83.41% <100.00%> (ø)

... and 9 files with indirect coverage changes

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 in Codecov by Sentry.
📢 Do you have feedback about the report comment? Let us know in this issue.

[randombit]

I'm open for better approaches.

TBH I'm pretty unhappy with the whole cipher API as it exists. For context, originally it was designed to handle arbitrary transformations with minimal copies. (Until #475 cipher modes and compression shared a base class.) But as a cipher-specific interface it leaves a bit to be desired, and is not so easy to use, especially in incremental mode.

I think a new cipher API is happening at some point, and we can try to apply some lessons learned.

[reneme]

I think a new cipher API is happening at some point, and we can try to apply some lessons learned.

I read that you don't foresee this for Botan 3.0, right?

With today's PRs I'm trying to reach a somewhat consistent API re: the adoption of std::span and concepts::resizable_byte_buffer.

[randombit]

I read that you don't foresee this for Botan 3.0, right?

No, not at this stage of things certainly. I don't even know what exactly a new API would look like at this point.

With today's PRs I'm trying to reach a somewhat consistent API re: the adoption of std::span and concepts::resizable_byte_buffer.

👍 absolutely, this PR series is a good improvement over status quo