Additions to CMakeLists.txt to better organize information in IDEs

[jdsierral]

In general is just adding some organization instructions so that folders in an IDE show up properly. Notable details include adding header files to source files: This is important so that an IDE can build "Counterpart" relationships between .h files and .cpp files Formats were also separated as they are later used to move all those targets into a general Targets folder

Tested this in Xcode (It also seems to be working properly in VS2019 but couldn't build because of Faust dependencies)

Most of the changes come from suggestions on this discussion.

[jdsierral]

I did a general code review suggesting changes to implement more modern C++ practices such as smart pointers. The rest were general things that should be showing as warnings in any case.

[jdsierral]

I also removed empty argv in favor nullptr and whenever there are conditionals on returns simplified them to match the real intention of the code

[glocq]

Thanks a lot for the detailed info, it's great for learning to have a personalized code review like that. I still have a few issues and questions.

• Why remove AUv3? Does it not compile?
• Why add the header files to the list of source files? Don't they get included in the .cpp files anyway?
• Unfortunately the plugin now crashes when I attempt to compile a Faust program. I think it is linked to the nullptr --- I had previously tried using an empty array, and it crashed too. Seems like we need at least an array with an empty string (might be something we should tell the Faust folks?)

[jdsierral]

No prob :)... I thought about doing a general first pull request before diving directly in the parameter stuff to see how it would work XD (learning a lot myself with all this :) )

• AUv3 -> whoops hehe... I didn't remove it intentionally but I think that one is a bit harder to support (just a hypothesis here but because of the sandboxing I wouldn't be surprised if embedding a JIT could cause problems... ill try to test it myself since I'm guessing it is a Mac thing anyway ;) )
• HeaderFiles-> that one is important to make them available to the IDE, it doesn't even impede the compilation on the IDE but it complains because it doesn't really know where the declarations are... From what I could test it shouldn't affect anything on the makefile build right?
• yes! I though I had tested that but I can confirm that making the argv a nullptr causes a crash... I'm guessing it tries to dereference the pointer at some point causing the crash? (Maybe in the future we could actually allow passing flags to the compilation process so that we could take advantage of this option in the JIT, e.g. double precision internal calculations?)

[glocq]

Ah yes, I just realized you were the one who added AUv3 in the first place. I am letting you deal with the MacOS specific stuff indeed. I just merged your changes! (except for the last commit) Thank you for your work. Looking forward to the parameter stuff if you do it :)