2018-JUN-15
- Improved version to work with public release DB
- Update dtypes to match Y3 tables (PR 149)
- Fix bytes to unicode issue for py3 when writing h5 files
- Fix bug with Oracle types NUMBER with no scale/digits
- Add desdr database and remove oldoper
- Fix HDF5 output files, (see 140)
- Add compression option to configuration, gzip is applied to .csv,.tab,.fits (latter only for 1 DB trip), bzip2 for hdf5
fits can be opened in append mode when they are compressed, workaround is to increase prefetch