Billy O'Neal

How is the polyfill setting baked into the installed bits? See https://github.com/microsoft/vcpkg/blob/master/docs/maintainers/maintainer-guide.md#a-feature-may-replace-polyfills-with-aliases-provided-that-replacement-is-baked-into-the-installed-tree for more details.

> I didn't get it > > mongo-cxx-driver already have built-in defines for selecting polyfills, no need to patch it > > just need to add a vcpkg feature for...

> But look further into mongo-cxx-driver upstream cmake files, they already done the baked part I believe what upstream is doing is trying to inspect __cplusplus, which means the user's...

> polyfill choose by `BSONCXX_POLY_USE_${BSONCXX_POLY}` and `${BSONCXX_STANDARD}`, not auto detected. and these two variables set by vcpkg feature at around line 21 > as you can see, there were already...

> I think the visibility patch is covered by existing upstream PRs Can you link to them somewhere? We generally want to avoid 'putting words in upstream's mouth' and/or otherwise...

> > Can you link to them somewhere? We generally want to avoid 'putting words in upstream's mouth' and/or otherwise doing things of which they are unaware. > > The...

> I don't believe we can accept this because it's using a feature to control alternatives: https://github.com/microsoft/vcpkg/blob/master/docs/maintainers/maintainer-guide.md#do-not-use-features-to-control-alternatives-in-published-interfaces > > It may make sense to do unconditionally on Windows (which is...

> @BillyONeal does it work for you? > > ``` > if(VCPKG_TARGET_IS_WINDOWS) > list(APPEND OPTIONS -DENABLE_UNICODE=ON) > endif() > ``` It works for me! @bagder @MarcelRaad @apique Is there any...

Thanks for your contribution @rozdaybeda !

This is becoming a relatively large build system replacement. Do you know why we aren't just using upstream's meson one?