Ondrej Medek

The first commit is about 10-15%, the second commit about 5%. The second commit moved caching to the instance of processor, so, if someone uses a new processor every time...

I've added `BaseType.size` caching to the first commit, which is kind of micro-optimization.

Update: I've run benchmarks on my Win32 Intel i5 2.5GHz machine previously. I've tried it on our Linux x64 (Ubuntu Xenial) Intel Xeon 2.6GHz server, and it's 5x faster. (all...

Just about the commit: **FitFileDataProcessor cache methods not just method names** I thing the `data_processor` should be removed from the `FitFile` completely. It may be designed as a wrapper around...

@pR0Ps thanks for CR. Meanwhile I've got mroe insight into the problem with the processors a bit. So I would discard the commit about the processors, make a issue for...

Duplicate of #8 some alpha code is in the PR #58 (not merged yet).

IMO #8 is mostly about saving fit file, maybe some simple CVS or JSON parsing. It's not intention of this tool to do TCX or any other elaborate data format...

@dtcooper `fitdump` is ready, but it _reads_ FIT. This feature request is about _writing_ a FIT file.

The Python fit library https://github.com/rembish/fit can create FIT files, but it's unmaintained for some time, though.

What about to split `profile.py` file to more files: `profile_types.py` and `profile_messages.py` (or just the old `profile.py` to keep backward compatibility.) Then, it would be a bit simpler to track...