[Code Cleanup] Cleaned up some header files and move a bit of debugging logic to cmake

[Rossmaxx]

• Renamed lmms_basics.h to LmmsTypes.h and scoped it down for that purpose.
• Shifted the LMMS_STRINGIFY macro to it's own header.
• Removed the debug.h header file and use cmake to handle the macro logic.
• Remove some unused headers and include directives

[Rossmaxx]

Do note that the diff itself is small. Most of the thing is due to the file rename in the #include statement

[Rossmaxx]

Now i should hunt down any unused includes over here, but I don't really have the insanity to do that manually.

Edit: using iwyu, I could somewhat automate this process.

[Rossmaxx]

I'll fix the declaration later.

[Rossmaxx]

Seems like I got myself into a rabbit hole with this refactoring

[Rossmaxx]

I'm done now.

[Rossmaxx]

I'll undraft the PR once I manage to fix the history. The work is done for now.

[tresf]

@tresf any idea on fixing this?

When I get into a pickle like this, I take the lazy route and PR against a test branch that's even with master, squash everything down, force that commit back. For this reason, I'm probably not a good person to ask. Last time my efforts were scrutinized #7726. Perhaps someone with more experience in this area can help.

[Rossmaxx]

I found a way to fix it. I'll reorder commits using an interactive rebase. But it is a bit effort consuming tho.

[sakertooth]

Since you renamed lmms_basic.h to LmmsTypes.h, why not do something similar for lmms_constants.h and lmms_math.h?

[Rossmaxx]

Since you renamed lmms_basic.h to LmmsTypes.h, why not do something similar for lmms_constants.h and lmms_math.h?

I could, but mind if I defer, the PR is getting a bit too large already. I'll do the other changes now.

[szeli1]

I tested this PR and it doesn't seem like it causes any issues or bugs.

I couldn't find any problems with the code.

I think a future improvement could be if the Spectrum analyzer FFT code was moved to a common lmms class, since it is common between the eq and SpectrumAnalyzer

[Rossmaxx]

Thanks for the heads up @szeli1. I could do with another review from someone with write access, and I am safe to hit the button. @sakertooth mind following up? Need your opinion on deferring the other renames too

[sakertooth]

Should we change the name of the changed files from snake case to camel case as well?

[Rossmaxx]

Should we change the name of the changed files from snake case to camel case as well?

No need. That's a bit too much.

[Rossmaxx]

@rubiefawn can i get a green tick atleast now?

[tresf]

I tested this PR and it doesn't seem like it causes any issues or bugs.

I couldn't find any problems with the code.

I think a future improvement could be if the Spectrum analyzer FFT code was moved to a common lmms class, since it is common between the eq and SpectrumAnalyzer

@Rossmaxx Why was this recommendation ignored? I agree with @szeli1, just because a value is only used in one place doesn't mean it shouldn't be in a shared header. This moving things like audible frequencies into plugin-specific headers does not make sense.

[Rossmaxx]

This moving things like audible frequencies into plugin-specific headers does not make sense.

Well, you got a point too. I will wait for a 2nd opinion before reverting.

[Rossmaxx]

why was this recommendation ignored?

As to ignoring that comment, I don't really understand that code as of now, and though i could dabble a bit I don't want to create any regressions.

[tresf]

This moving things like audible frequencies into plugin-specific headers does not make sense.

Well, you got a point too. I will wait for a 2nd opinion before reverting.

You mean a 3rd?

[Rossmaxx]

@tresf I have reverted the spectrum analyser stuff, it's bloating the diff for no real benefit. Also I'm ignoring the fftw comment for now. The PR is ready for merge.