[PATCH] D131438: [clang][dataflow] Analyze constructor bodies

[Sam Estep via Phabricator via cfe-commits]

samestep added a comment.

In D131438#3710697 <https://reviews.llvm.org/D131438#3710697>, @xazax.hun wrote:

> I feel like this is a repeated pattern. The CSA solved a very similar issue by introducing the CallEvent class hierarchy. I also remember seeing many disparate code snippets littered throughout the clang codebase that tries to deal with the problem of not having facilities to treat call-like nodes uniformly. At some point, I believe there were even some AST changes or supporting structures proposed to ameliorate this problem, but I can't find those at the moment. While I think it might be OK to introduce yet another workaround here, this is a cleanup that is long overdue, and I hope someone will have the time to actually improve the situation. Sorry for the rant, I will actually look at the code but had to vent this.

No worries at all for the rant, I appreciate the broader context. I was definitely surprised to learn that the two types don't share a common ancestor.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D131438/new/

https://reviews.llvm.org/D131438