Overview

The HDMF specification language defines formal structures for describing the organization of complex data using basic concepts, e.g., Groups, Datasets, Attributes, and Links. See the documentation for more information and release notes.

The HDMF specification language is used by the Hierarchical Data Modeling Framework (HDMF).

Customizing the language for specific applications

Certain keys of the hdmf-schema-language may be customized for a particular scientific domain or use case. For example, the data_types key in the namespace may be customized to be neurodata_types and the data_type_def and data_type_inc keys in the schema may be customized to be neurodata_type_def and neurodata_type_inc for use in a neuroscience-specific application of the hdmf-schema-language. This can be easily done by forking the hdmf-schema-language repo and updating source/namespace_map.yml with the appropriate customizations. The source/hdmf.schema.json and any references to HDMF in the documentation should also be modified appropriately. See the NWB schema language repo for an example of customizing the HDMF schema language for use as the NWB schema language.

For maintainers

The HDMF specification documentation is generated using Sphinx

Building the documentation

Install the latest version of Sphinx:

python -m pip install sphinx

To build the documentation, enter:

make <doctype>

where <doctype> is, e.g., html, latexpdf.

For a complete list of supported doc-types, enter:

make help

To remove previously generated documentation, enter:

make clean

Making a release

1. Update the release notes in source/release_notes.rst.
2. Update the version in source/namespace_map.yml, source/hdmf.schema.json, and source/conf.py