FLACArray

The FLACArray package is my effort to centralize the use of FLAC compression for collections of scientific data timestreams. Many types of data acquisition systems record noisy timestreams with a fixed quantization (for example from a analog to digital converter). The FLAC algorithm can often produce good compression for such data, since it has properties similar to audio data. FLACArray provides nice interfaces for compressing data, doing I/O to either HDF5 or Zarr formats, and also supports fast seeking and decompression of slices of timestreams.

FLACArray is available both on PyPI and on Conda-Forge.

Recent Work

FLACArray consists of a python package with some lower-level compiled code. The latest release also provides a public C interface that can be used by external compiled projects that want to leverage the latest features, such as support for 64bit data stored in “stereo” 32bit FLAC channels.