Multi-room hand history pain points: what breaks parsers the most?

Hi, I’m a dev working at 2WinPower. Currently building an internal multi-room hand history parser/normalizer for personal use, and I’m trying to map the real edge cases before I waste weeks on the wrong assumptions.

What are the most annoying HH issues you hit across rooms/tools (missing actions, weird all-in side pots, fast-fold formats, anonymous tables, MTT specifics like PKO/ICM tags, re-entries, etc.)? If you have examples of hands that break PT/HM/replayers, I’d love to know what exactly fails (seat mapping, pot size, action order, showdown, etc.).