Since v0.2.0, we are using separate frame processing Framer API.
Before this version, we made all parsing inside readers\writers. But now I've separated frame processing into the standalone classes. This approach provides more modularity and ability for extending.
Currently, separate frame processing implemented for these tags:
APIC
frameCOMM
frameTXXX
frameWXXX
frameUSLT
frame
Another tags proceeds through default tag processing