Have the UI vocabulary JSON files generated as part of the build

[dannymandel]

• Check out https://github.com/isamplesorg/navocab
• Run vocab uijson as part of the build
• Need to run one for each type of vocabulary
• Make sure that we include -e when we run the materials
• This should run with every change
• May also need to generate the database cache using the load command -- give it the URLs to GitHub for the 3 ttl files when we generate the database at build time https://github.com/isamplesorg/metadata/tree/develop/src/vocabularies --make sure to grab the extension files as well

[datadavev]

The navocab tool collates vocabularies from the .ttl files into a "triple store" using RDFLib, and supports various operations on the triple store. The process is a bit clunky, but basically involves building the store, then using the store to generate the json.

Building the store (can use local file path or url):

vocab load "https://raw.githubusercontent.com/isamplesorg/metadata/develop/src/vocabularies/materialtype.ttl"
vocab load "https://raw.githubusercontent.com/isamplesorg/metadata/develop/src/vocabularies/sampledfeature.ttl"
vocab load "https://raw.githubusercontent.com/isamplesorg/metadata/develop/src/vocabularies/specimentype.ttl"

Then adding the various extensions.

The webui JSON can be produced by:

vocab uijson VOCABULARY

Where VOCABULARY is one of the root vocabularies. These can be found by the command:

vocab vocabs

Then using one of the root vocabularies, and adding -e to include the extensions. e.g.:

vocab uijson mat:materialsvocabulary -e

Output is json to stdout.