Fix bugs in modulator handling

[derselbst]

This PR fixes two problems:

#1068: Previously, if a modulator had src1 set to "No Controller" it would always cancel src2.

#1389: If any src was set to "No Controller", its input value would not be transformed.

Also, previously, some modulators received a "special treatment" during soundfont loading, by setting their amount to zero. This was to aid the magic done in fluid_voice_get_lower_boundary_for_attenuation(). After looking through that function again, I don't see the changes introduced in fluid_defsfont.c will cause problems for the fluid_voice_get_lower_boundary_for_attenuation(): Bipolar modulators are correctly accounted for. And if src1 is set to FLUID_MOD_NONE it will no longer cancel the modulator, hence it would trigger the else branch here:

fluidsynth/src/synth/fluid_voice.c

Lines 1808 to 1812 in ce97c6d

	else
	{
	/* No negative value possible. But still, the minimum contribution is 0. */
	min_val = 0;
	}

The only theoretical problem that comes to my mind is if there was a modulator which sets both src1 and src2 to "No Controller", in which case it would never cause a "minimum" contribution.

But I might be missing something else. @mawe42 In case you have some time, could you have a look into the fluid_voice_get_lower_boundary_for_attenuation() as well?

[sonarcloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
100.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarCloud