PyDaF stands for Python Data Flows. It is an experimental Python package that implements data-driven processing pipelines.
Data-driven processing pipelines revolve around two major elements: data units, which encapsulate a type, a set of tags and actual data, and processors. Processors are fragments of code that get executed whenever the job's current data pool contains data units matching the processor's input specification.
IMPORTANT - PyDaF is pretty much an experiment. The API is not stable at this stage. At all.
The current version of PyDaF can be downloaded from the download page. The latest version is 0.3.0.
PyDaF's manual can be viewed on this site. It is also available for download as a PDF, and its TeX sources are included in the distribution.