Core functionality¶

This section describes the core functionality of bitflow.

Batch¶

Batch is currently the only keyword in the bitflow script language. Batch is used as a processing step inside a script and must be combined with a subpipeline like "input.csv" -> batch(sample-window-size=5){ avg() } -> out.csv

(TODO) Required batch modes: - flush after input closed (entire file) - separation tag (with optional timeout) - flush after fixed number of samples - flush after fixed period of time

Fork¶

TODO: describe fork semantics (distributors, etc)

Pipeline¶

TODO