feat: Cmake port of Media crypto

[cosmin]

Rebasing #1148 on top of current cmake branch

Related to #1047

[google-cla]

Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).

View this failed invocation of the CLA check for more information.

For the most up to date status, view the checks section at the bottom of the pull request.

[joeyparrish]

@cosmin, could you fix the linter errors?

To run the linter against the changes since your local cmake branch, run:

python3 packager/tools/git/check_formatting.py cmake

To fix the same errors, run:

git clang-format --style Chromium cmake

For both of those, make sure your local cmake branch is up-to-date with the upstream cmake branch.

Thanks!

[cosmin]

Now that I reverted the change with dealt with calling CryptInternal with an insufficiently large ciphertext size I'm seeing a test failure related to this. Taking a look at that.

[cosmin]

https://github.com/cosmin/shaka-packager/blob/75987351976cd4a84b24cf1590c77482aab9250a/packager/media/crypto/encryption_handler.cc#L302

ok, I think this is an example of a callsite that sets up the cipher text buffer to be the same size as the plaintext, but when using AesCbcEncryptor the required ciphertext size is larger due to padding. This then fails the check, which the earlier commit worked around by re-allocating a larger buffer.

[cosmin]

@joeyparrish take a look at the most recent commit, the only way I could think to workaround this issue is to introduce a function to return the expected ciphertext size so the caller can allocate a properly sized buffer for it. Open to suggestions though if there's a better way to solve this.

[joeyparrish]

Such an improvement, thanks!

[joeyparrish]

I have now fixed the HTTP tests, so I'm going to merge your branch with cmake and update the PR again. Then we should see a fairly straightforward test pass and know that the crypto tests enabled by this PR are passing.

[joeyparrish]

Build failure on Windows:

packager\media\crypto\subsample_generator.cc(273,28): error C2220: the following warning is treated as an error
packager\media\crypto\subsample_generator.cc(273,28): warning C4457: declaration of 'frame' hides function parameter
packager\media\crypto\subsample_generator.cc(262,20): message : see declaration of 'frame'

That sounds like it could be a real bug.

[cosmin]

As far as I can tell this code has not changed recently. I don't think it's a bug in this case because by the time

  for (const VPxFrameInfo& frame : vpx_frames) {

happens which shadows the frame function parameter it was already parsed into vpx_frames by

  if (!vpx_parser_->Parse(frame, frame_size, &vpx_frames))

however it's trivial to fix while I'm at it.

[cosmin]

All tests seem to pass now except for Arch which should be fixed by #1233