A project to clarify and eventually evolve the Universal Dive Data Format (UDDF) specification.

The current version of this repository mirrors UDDF 3.2.3 as published at https://www.streit.cc/resources/UDDF/v3.2.3/en/index.html. No content changes are layered on top yet — the goal at this stage is to package the spec in a buildable, reproducible form so that future iterations (errata, examples, clarifications, machine-readable changelogs) can be tracked in version control.

The published UDDF spec has known inconsistencies between its prose pages, the per-element HTML pages, and the XSD schema. The XSD is machine-validatable; the prose isn't. This repository treats the XSD as the source of truth and rebuilds everything else from it.

src/
├── schema/
│   └── uddf.xsd          authoritative schema (currently UDDF 3.2.3)
├── templates/
│   ├── element.md.j2     per-element markdown page
│   └── index.md.j2       element index
└── tools/
    └── generate.py       XSD → markdown generator

dist/ is the build output (gitignored). Each element gets a stub markdown page derived from its XSD definition, listing attributes and child elements with cardinality.

make install     # one-time: pip install -r requirements.txt
make generate    # rebuild dist/ from src/
make clean       # remove dist/

Requires Python 3.10+ and Jinja2.

This is the bootstrap commit. The next milestones, in order:

1. Layer hand-written prose (descriptions, examples, cross-references) on top of the XSD-derived structure.
2. Add a structured changelog that supersedes the XSD-header <!-- ... --> block and the disconnected HTML changes pages.
3. Decide on the publication target (rendered HTML site, MkDocs, or pure markdown for GitHub viewing).
4. Catalog known spec defects with proposed corrections — the spec inconsistencies should be tracked here, not only in downstream library notes.