[Lldb-commits] [PATCH] D71306: [RFC] Change how we deal with optional dependencies

Jonas Devlieghere via Phabricator via lldb-commits lldb-commits at lists.llvm.org
Wed Dec 18 14:48:47 PST 2019


JDevlieghere added inline comments.


================
Comment at: lldb/cmake/modules/LLDBConfig.cmake:33
+    set(maybe_required)
+  elseif("${${variable}}" STREQUAL "ON")
+    set(maybe_required REQUIRED)
----------------
mgorny wrote:
> JDevlieghere wrote:
> > mgorny wrote:
> > > Now you broke it the other way around. Is there any reason you can't just `${${variable}}` here?
> > Can you please describe the desired behavior rather than the symptom? I must be missing what you're trying to achieve. :-) 
> > 
> > Maybe there's a gap in my CMake knowledge, but wouldn't `if(${${variable}})` evaluate to `true` for **any** string value? 
> The desired behavior is to keep standard CMake behavior for any value other than `AUTO`, and by standard behavior I mean: https://cmake.org/cmake/help/v3.0/command/if.html the first definition on the list.
> 
> Particularly, in Gentoo we default to passing `yes`/`no` for historical reasons. I would find it irritating if different places in LLVM accepted different constants.
Got it, thanks for the explanation!


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D71306/new/

https://reviews.llvm.org/D71306





More information about the lldb-commits mailing list