ResInsight/scripts/ri_cloud_api

ResInsight Cloud API service

A localhost FastAPI/uvicorn service whose lifecycle is managed by ResInsight (RiaCloudApiService): it is started when Sumo authentication succeeds, bound to an auto-selected free port, health-checked on /alive, restarted if unresponsive, and killed when ResInsight closes.

Location

The package lives at scripts/ri_cloud_api in the repository root.

On install, the contents of this folder are copied to <install folder>/CloudServiceApi (see the Installation packaging Cloud Service API section in the top-level CMakeLists.txt).

RiaCloudApiService::serviceWorkingDirectory locates the package by probing, in order:

1. the configured Shared Script Folder(s) (Preferences → Scripting → Shared Script Folder(s)), and
2. <application dir>/CloudServiceApi/

and returning the first folder that contains a ri_cloud_api directory. For a development build, point one of the shared script folders at the repository scripts folder (e.g. C:/Git/ResInsight/scripts), since the executable lives in the build tree and not next to the package.

Deployment status

Target state: this service is intended to be distributed and installed as a regular pip package into the active Python environment. Once packaged, importing ri_cloud_api (and any ri_cloud_* workspace libraries) will work directly from the environment's site-packages, with no path manipulation.

Current state (interim): the service is not yet a pip package. It runs from the source tree inside a virtual environment, with the folder containing ri_cloud_api used as the process working directory. If workspace libraries are laid out as ri_cloud_api/libs/<lib>/src, ResInsight adds each such src folder to PYTHONPATH at launch time (see RiaCloudApiService::buildProcessEnvironment). This is a temporary measure that should be removed once the package is published/installed.

Requirements (interim setup)

• A Python virtual environment configured in ResInsight under Preferences → Scripting → Python Executable (e.g. C:/Git/.venvResInsight/Scripts/python.exe).
• uvicorn and fastapi installed in that environment, plus any third-party dependencies pulled in by the routers.

Contents

ri_cloud_api/
  main.py                       shim exposing `app` for `uvicorn ri_cloud_api.main:app`
  primary/
    main.py                     FastAPI application, router registration
    routers/health/router.py    `/alive` health endpoint

Running manually (for debugging)

From the scripts folder (the parent of this package):

cd scripts
<venv>/Scripts/python.exe -m uvicorn ri_cloud_api.main:app --host 127.0.0.1 --port 8000

If workspace libraries are present, put them on PYTHONPATH first:

set PYTHONPATH=ri_cloud_api/libs/<lib>/src

Then check the health endpoint:

http://127.0.0.1:8000/alive   ->   {"status": "alive"}

Interactive API docs are available at:

http://127.0.0.1:8000/docs