Bring spdlog into OpenSim.

[chrisdembia]

Brief summary of changes

This PR is an offshoot of #2551. I built on @aymanhab's original work to build and install spdlog.

The destination for this PR is feature_logging, not master.

Testing I've completed

I can build and run an OpenSim C++ example from the OpenSim installation on Mac. I did not test on Windows.

CHANGELOG.md (choose one)

• no need to update because...this feature is not complete yet

---

This change is 

[aymanhab]

Reviewed 6 of 6 files at r1.
Reviewable status: all files reviewed, 2 unresolved discussions (waiting on @chrisdembia)

---

cmake/OpenSimConfig.cmake.in, line 80 at r1 (raw file):

    # We pre-define these variables because finding Simbody redefines variables
    # that we need for finding spdlog.
    set(_SIMBODY_PATH "@PACKAGE_OPENSIM_INSTALL_SIMBODYDIR@")

Can you clarify this comment?

---

OpenSim/Common/CMakeLists.txt, line 18 at r1 (raw file):

    # Clients of osimCommon need not link to BTK.
    LINKLIBS PUBLIC ${Simbody_LIBRARIES} spdlog::spdlog PRIVATE ${BTK_LIBRARIES}
    INCLUDES ${INCLUDES}

Can we move the explicit library name to a CMake variable for future maintenance?

[chrisdembia]

Reviewable status: all files reviewed, 2 unresolved discussions (waiting on @aymanhab)

---

cmake/OpenSimConfig.cmake.in, line 80 at r1 (raw file):

Previously, aymanhab (Ayman Habib) wrote…

Can you clarify this comment?

Done.

---

OpenSim/Common/CMakeLists.txt, line 18 at r1 (raw file):

Previously, aymanhab (Ayman Habib) wrote…

Can we move the explicit library name to a CMake variable for future maintenance?

The Simbody_LIBRARIES and BTK_LIBRARIES variables are defined by the projects' SimbodyConfig.cmake, etc. files, not by OpenSim's CMake files. spdlog does not define a similar variable, so I would prefer to not invent one ourselves.

[chrisdembia]

Ready for review.

[aymanhab]

Reviewable status: all files reviewed, 1 unresolved discussion (waiting on @chrisdembia)

---

cmake/OpenSimConfig.cmake.in, line 80 at r1 (raw file):

Previously, chrisdembia (Christopher Dembia) wrote…

Done.

I don't think anything changed here, unless I'm looking at a stale file.The confusing part for me was the statement that "finding Simbody redefines variables that we need for finding spdlog"

[chrisdembia]

Reviewable status: 5 of 6 files reviewed, 1 unresolved discussion (waiting on @aymanhab)

---

cmake/OpenSimConfig.cmake.in, line 80 at r1 (raw file):

Previously, aymanhab (Ayman Habib) wrote…

I don't think anything changed here, unless I'm looking at a stale file.The confusing part for me was the statement that "finding Simbody redefines variables that we need for finding spdlog"

Oops; I hadn't pushed. Can you check again?

[aymanhab]

Reviewed 1 of 1 files at r2.
Reviewable status: all files reviewed, 1 unresolved discussion (waiting on @chrisdembia)

---

cmake/OpenSimConfig.cmake.in, line 80 at r1 (raw file):

Previously, chrisdembia (Christopher Dembia) wrote…

Oops; I hadn't pushed. Can you check again?

👍

[chrisdembia]

@aymanhab do you know why testStaticOptimization is failing on Mac?

[aymanhab]

No but very sporadic.

…

On Fri, Oct 11, 2019, 3:49 PM Christopher Dembia ***@***.***> wrote: @aymanhab <https://github.com/aymanhab> do you know why testStaticOptimization is failing on Mac? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#2591?email_source=notifications&email_token=AA6JY4CEYISCJ5VSK2FRBD3QOD7HJA5CNFSM4I7UPN2KYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEBBMZSQ#issuecomment-541248714>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AA6JY4C64MIKS7JVR4OVZB3QOD7HJANCNFSM4I7UPN2A> .

[chrisdembia]

Thanks for the review @aymanhab