gns-science / nshm-model-graphql-api Goto Github PK

we want to remove the 'baked' in inversion/background roles and support more flexiblity in source logic trees.

• Source Logic Tree Branch contains one-or-more source models... each with a toshi_id/URI
• GNS-Science/nzshm-model#15
• update the graphql API to reflect these changes
• import existing SLT json/python to store
• import existing NRML XML to store

Let's keep up the Q/A drive

install and setup tox with

• mypy
• pytest
• flake8
• black
• isort

also bump2version and CHANGELOG

The Object Identification pattern must be defined such that a unique graphql_relay object id (nodeId) can be guaranteed across the entire NSHM API domain.

NB unique IDs and mapping to service are essential to NSHM search service. and also the public API.

We need a document describing the ruel and how this is implemented in our django_graphql code and Elastic Search integration code,

if DEBUG then:
filename: /DEV/static/graphene_django/graphiql.js
else:
filename: /static/graphene_django/graphiql.js which FAILS

This happens since we picked up DEBUG from OS. For now setting DEBUG=1 gets things working again.

again, to NSHM API standards

actually this means some updating:

• look at new AWS OpenSearch platform and UI facilites (kabana etc)
• look at serverless version of OpenSearch
• verify library options
• look at client and API integration

depends on #14