- column_names of point maker no longer saved as title, but as attributes --> auto read out in OrcaNet
- properly implement h5shuffle in tools and add unittests
- concatenate copies over attributes
- add option to keep dtype of mcinfo
- now all infos in the blob are kept for the mc_info_extr, including classical reco
- add overwrite option to BaseProcessor

- overhaul concatenate.py and move to main repo
- Add 4line bin edges
- timing of modules
- subtract t0 option
- warning for when mcinfo maker produces more then one line

- orcasong is now called legacy
- orcasong_2 is now called orcasong

Other changes:
- Added one line 2017 bin edges to repo
- added unittests
- added versioning for files created with orcasong
- Expanded doc
- Added check to calib if calib has been done to file already

- Added option for calibration with detx file
- added bin stats plot to setup
- better doc

- Removed event skipper from concatenate_h5_files (use shuffle instead)
- Fix for shuffle_h5, and it also copies used_files datagroup if present to shuffled file
- Added authorship to orcasong_plag

- Added command line tools to setup.py

- changed data_to_images in order to use the KM3Pipe Pipeline concept instead of iterating over the HDF5Pump.
- Added output dirpath, chunksize, compression filter and compression strength arguments
- Added CI versioning and publishing of OrcaSong via Pypi