While checking how to import/export the mixer script definitions in Companion, I found that the mixer script data stored in the model lack something very important.
The inputs (either a source or a fixed value) are stored this way in memory:
union ScriptDataInput {
int16_t value;
source_t source;
};
The issue here is that the information deciding on whether it is a value or source depends on the mixer script itself. It is stored nowhere else. This is an issue as we don't have this information when converting the data from the binary model, and thus cannot map the source properly to a symbolic name. So what happens at the moment is that we only output value, and hope for the best (which necessarily breaks once the some sources are remapped).
While checking how to import/export the mixer script definitions in Companion, I found that the mixer script data stored in the model lack something very important.
The inputs (either a source or a fixed value) are stored this way in memory:
The issue here is that the information deciding on whether it is a
value
orsource
depends on the mixer script itself. It is stored nowhere else. This is an issue as we don't have this information when converting the data from the binary model, and thus cannot map the source properly to a symbolic name. So what happens at the moment is that we only outputvalue
, and hope for the best (which necessarily breaks once the some sources are remapped).