Did you know about #cloc `--by-line`? I just learned about it yesterday, and it kicks ass

I'm explicitly aiming to implement the most features in the fewest #LoC while simultaneously avoiding the code undergoing #spaghettification

The release build is 100x faster than the dev build (at least as PipeWire is concerned): 5msec vs 50usec to generate a chunk of 1024 audio samples (huge buffer I know)

That didn't stop me from optimizing the rendering today. Or fixing bugs, or making new ones