Shadow parser false positives

[David Healey]

I'm still not using this feature because I find it's too buggy, but someone else brought another false positive to my attention so thought it might be a good idea to collect them here so Christoph can play wack-a-mole

[ustk]

this one's because reduced has an optional y parameter. Not using it results in a warning:

[dannytaurus]

@David-Healey @ustk What are you both talking about? What am I missing out on here?!

[David Healey]

@dannytaurus said in Shadow parser false positives:

What am I missing out on here?

You're missing out on an annoying buggy feature :)

Right-click in the script editor to toggle "Enable Live Diagnostics" - but it's on by default so you've turned it off at some point previously which was the right decision :p

[dannytaurus]

@David-Healey Oh, you mean list this?

Claude says:

It's painting the warning in the wrong tab, and it's a false positive from conservative metadata.

The metadata classifies every sendRepaintMessage variant as unsafe wholesale; a finer classification ("warning: fine to call, just don't do it per-sample") would be more accurate.

The two points are: (1) ApiDiagnostic needs the callback id so the editor can place the marker in the right tab, and (2) sendRepaintMessage's callScope should probably be "warning" rather than "unsafe".

Glad I don't use the HISE script editor anymore. That would confuse the hell out of me!