Earlier quoted context omitted.
To reduce the amount of allocation instead of: struct parsed_data * = parse (...); struct process_data * = process (..., parsed_data); struct foo_data * = do_foo (..., process_data); you can do parse (...) { ... process (...); ... } process (...) { ... do_foo (...); ... } It sounds like violating separation of concerns at first, but it has the benefit, that you can easily do procession and parsing in parallel, and al…
How testable is this, though?
It might be a problem when you can't afford side-effects that you later throw away, but I haven't experienced that yet. The functions still have return codes, so you still can test, whether a correct input results in no error check being followed and that incorrect input results in an error check being triggered.