BREAKING CHANGE: the order of these operations was inconsistent
across the different VMs. Go VM was the only one to first modulate
and then apply note tracking multiplication. But that made most
sense. So now all different VM versions work in this same way.
The exception to the rule is the dialog buttons (which use still
the default material buttons), because when there is a modal dialog
on screen, there is not much else the user would want to do.
Fixes#156
In addition to the oscilloscope and loudness/peak detections, this
commit refactors all the channels between components (i.e.
ModelMessages and PlayerMessages) etc. into a new class Broker. This
was done because now we have one more goroutine running: a Detector,
where the loudness / true peak detection is done in another thread.
The different threads/components are only aware of the Broker and
communicate through it. Currently, it's just a collection of
channels, so it's many-to-one communication, but in the future,
we could change Broker to have many-to-one-to-many communication.
Related to #61
Also includes a refactoring of the List: all methods that accepted
or returned a [from, to] range now return a Range, which is
non-inclusive i.e. [start,end).
Also the assignUnitIds was slightly refactored & a new function
called assignUnitIdsForPatch was added, to assign all unit IDs for
an patch at once.
Closes#157, #163.
The sample-based oscillators converted the samplepos to an integer
and did samplepos < loop_end comparison to check if we are past
looping. Unfortunately, the < comparison was done in signed math.
Normally, this should never happen, but if the x87 FPU stack
overflowed exactly at right position, we then got 0x80000000 in
samplepos, which is equal to -2147483648. Thus, we considered that
sample is not looping and read the sample table at position
-2147483648, well out of bound. TL;DR changing jl to jb makes sure
we always wrap within to sample table, no matter what.
Fixes#149.
