Document-Driven Testing

This package provides tools for doing automated parsing, processing, and annotation of "test tables" in documents (typically HTML files).

To use the package, one creates requirements/design documents that contain tables describing "text fixtures" such as expected outputs given a set of inputs. Then, once you've created appropriate adapters for the tables, you can run a DDT processor over the document to create an annotated document. The annotated version will usually shade passed tests in green, failed tests in red, add a summary of passed/failed tests, etc.

You can then do things like generate the annotated version on a regular schedule and post it on a website, so that all of a project's stakeholders can see the development status relative to the requirements.

Goals:

• Support all the FIT idioms cleanly

• Support HTML generated by MS products, StructuredText, and reST, so that developers and non-developers alike can create testing documents

• Independent reimplementation, free of GPL restrictions

• Use PEAK development idioms (e.g. domain model objects separated from parse/format implementation, property namespace to look up fixture types, commands framework for command-line tools, etc.)