Add Windows JACK support

[tresf]

JACK is surprisingly plug-and-play with regards to LMMS on Windows, thanks to the hard work of @falkTX and the rest of the JACK2 team combined with our already existing weakjack (thanks @x42) module (preventing the need for windows dev libs).

• Adds JACK build support for Windows systems
• Removes libjack-dev dependency unless specified
• Adds jack2 as a submodule, related Submodule candidates #3963
• Bumps weakjack submodule to add MSVC support.

To test:

• Install JACK for Windows from the website
• Launch JACK with ASIO support per JACK's website instructions
• Switch LMMS backend to JACK
• Configure using qjackctl, found in JACK install directory

Closes #2136

[LmmsBot]

🤖 Hey, I'm @LmmsBot from github.com/lmms/bot and I made downloads for this pull request, click me to make them magically appear! 🎩

Windows

• Windows 32-bit: lmms-1.2.3-752%2Bg52f780b5b-mingw-win32.exe (build link)
• Windows 64-bit: lmms-1.2.3-752%2Bg52f780b5b-mingw-win64.exe (build link)
• Windows 32-bit: lmms-1.2.2-msvc2017-win32.exe (build link)
• Windows 64-bit: lmms-1.2.2-msvc2017-win64.exe (build link)

Linux

• Linux (AppImage): lmms-1.2.3-752%2Bg52f780b-linux-x86_64.AppImage (build link)

macOS

• macOS : lmms-1.2.3-752%2Bg52f780b5b-mac10.13.dmg (build link)

🤖

{"platform_name_to_artifacts": {"Windows": [{"artifact": {"title": {"title": "32-bit", "platform_name": "Windows"}, "link": {"link": "https://10066-15778896-gh.circle-artifacts.com/0/lmms-1.2.3-752%2Bg52f780b5b-mingw-win32.exe"}}, "build_link": "https://circleci.com/gh/LMMS/lmms/10066?utm_campaign=vcs-integration-link&utm_medium=referral&utm_source=github-build-link"}, {"artifact": {"title": {"title": "64-bit", "platform_name": "Windows"}, "link": {"link": "https://10067-15778896-gh.circle-artifacts.com/0/lmms-1.2.3-752%2Bg52f780b5b-mingw-win64.exe"}}, "build_link": "https://circleci.com/gh/LMMS/lmms/10067?utm_campaign=vcs-integration-link&utm_medium=referral&utm_source=github-build-link"}, {"artifact": {"title": {"title": "32-bit", "platform_name": "Windows"}, "link": {"link": "https://ci.appveyor.com/api/buildjobs/5l409d92f1rnlf1l/artifacts/build/lmms-1.2.2-msvc2017-win32.exe"}}, "build_link": "https://ci.appveyor.com/project/Lukas-W/lmms/builds/36057155"}, {"artifact": {"title": {"title": "64-bit", "platform_name": "Windows"}, "link": {"link": "https://ci.appveyor.com/api/buildjobs/33armqsq0i5hnai1/artifacts/build/lmms-1.2.2-msvc2017-win64.exe"}}, "build_link": "https://ci.appveyor.com/project/Lukas-W/lmms/builds/36057155"}], "Linux": [{"artifact": {"title": {"title": "(AppImage)", "platform_name": "Linux"}, "link": {"link": "https://10069-15778896-gh.circle-artifacts.com/0/lmms-1.2.3-752%2Bg52f780b-linux-x86_64.AppImage"}}, "build_link": "https://circleci.com/gh/LMMS/lmms/10069?utm_campaign=vcs-integration-link&utm_medium=referral&utm_source=github-build-link"}], "macOS": [{"artifact": {"title": {"title": "", "platform_name": "macOS"}, "link": {"link": "https://10065-15778896-gh.circle-artifacts.com/0/lmms-1.2.3-752%2Bg52f780b5b-mac10.13.dmg"}}, "build_link": "https://circleci.com/gh/LMMS/lmms/10065?utm_campaign=vcs-integration-link&utm_medium=referral&utm_source=github-build-link"}]}, "commit_sha": "00c98a2a35b79549f117ce4b5bd91af842dcfdc2"}

[PhysSong]

The weakjack submodule should be updated to support MSVC, see x42/weakjack#9.

[tresf]

The weakjack submodule should be updated to support MSVC, see x42/weakjack#9.

Done via b1ab5ae.

[tresf]

All notes have been addressed, but there's a lingering issue from #5636 preventing AppVeyor to succeed. I'm fine waiting for that to be fixed to make sure AppVeyor is happy or merging as-is. Thoughts welcome. Pinging @Spekular since he authored #5636.

[Spekular]

Huh. Is this related to the issue fixed by #5678 somehow? Other builds since then have been fine, so I'm frankly not sure why an issue would pop up now.

[tresf]

Huh. Is this related to the issue fixed by #5678 somehow? Other builds since then have been fine, so I'm frankly not sure why an issue would pop up now.

Ok, sorry for the ping, I did search beforehand but must have missed it. I'll fast-forward and it should fix it, thanks.

Edit: Done via 7927bdd, just awaiting CI. 🤞

[tresf]

Ok, I fast-forwarded my branch didn't fix this @Spekular, can you comment on the AppVeyor error? It's the same as before.

[PhysSong]

I think (std::numeric_limits<int>::max)() may work. I'll test it out and fix lmms_basic.h as well if that works.

[tresf]

I'm frankly not sure why an issue would pop up now.

@Spekular as discussed in #devtalk on Discord, it appears this is as a result of adding jack.h into AppVeyor, which is new with this PR. So, although the symptom is the same, the cause is likley a 3rdparty header we don't manage. I've made my first attempt at patching this in d56db57. If that doesn't work, I might fire up a MSVC environment instead to avoid the guesswork.

I think (std::numeric_limits<int>::max)() may work.

@DomClark shared his sentiments on Discord, and I tend to agree with them that we shouldn't have to worry about adding parens to each call to max.

I'll test it out and fix lmms_basic.h as well if that works.

I started writing this on a separate branch, but was largely unsuccessful. I'll revisit if needed.

[tresf]

I'll test it out and fix lmms_basic.h as well if that works.

I decided to opt for a top-level...

ADD_DEFINITIONS(-DNOMINMAX)

... which I believe solves the problem proactively and holistically.

On the contrary, lmms_basics.h can't extend to headers which we don't manage, like weak_jack.h, so I believe this ADD_DEFINITIONS approach to be ideal.

Components which build using ExternalProject_Add may not import all flags, so RemoteVstPlugin is left as-is with its existing NOMINMAX flag.

[tresf]

Something's odd with Travis, it's not even queuing my builds (it's been hours and I've tried restarting it once). I'm not holding up the PR for that specifically. There's also a failure after rebase, but it's unrelated and explained here: 435dbc5#r43742718. Merging.