[PATCH] D149259: [analyzer][NFC] Use std::optional instead of custom "empty" state

[Donát Nagy via Phabricator via cfe-commits]

donat.nagy added a comment.

> It's gonna be bumpy.

Yup :) but it's not impossible...

> Also, note that this is in production here, so it will take me time to verify the patches. So, I'd probably prefer to have a patch stack, and do the evaluation for some selected patches - basically bundling the stack into a few (1-2) checkpoints.
> It would mean that the reviews would be done for each revision, but the verification (prior to landing anything) would be done at the selected stages.
> WDYT?

It would be effective to adopt a workflow where the individual commits are quickly reviewed for design directions and visible code quality issues (to avoid dead ends and keep the rebasing on a manageable level); but the complete verification and the upstreaming is done later, in bundles (to avoid wasting time on repeated verifications). Feel free to give "Seems fine for now; start working on the next commit; this will be verified and merged later." reviews after handling the bulk of obvious issues.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D149259/new/

https://reviews.llvm.org/D149259