Repository structure

The production software is divided up into the following structure:

• data-flow/lib/dflow: Helpfully named, contains modules useful for both dflow and processing (“gasp”), including the document.py module that defines document structures used by the data-flow / processing and production databases.
• data-flow/lib: More modules used by both dflow and production, but potentially also useful externally (e.g. lcgutils is a set of simple wrapper functions to call lcg-utils command line executables).
• data-flow/dflow: the dflow code.
• data-flow/gasp: the processing and production code:
  • data-flow/gasp/GangaSNOplus: the SNO+ ganga plugin.
  • data-flow/gasp/core: core stuff!
  • data-flow/gasp/modules: specific setup for different production and processing modules
• data-flow/database: database structure (design documents, attachments for monitoring), layout follows the requirements of stool to push the design document structure to a CouchDB database.
• data-flow/grid: contains examples on how to setup proxies etc.
• data-flow/doc: what you’re reading now.