IRIS Class Explorer UI

iris-class-explorer is a frontend UI for the backend project iris-table-stats.

The UI was vibecoded in Lovable, then adapted to work as an IPM-installable IRIS web module. The result is a browser-based explorer for persistent classes, table data, and column population statistics running directly inside InterSystems IRIS.

What it does

This UI connects to the iris-table-stats backend API and provides:

a searchable list of persistent classes
optional inclusion of system classes and mapped classes
a data tab to browse rows from a selected class
pagination for class data browsing
a stats tab to inspect column population percentages, populated counts, and empty counts
a configurable backend API URL stored in browser local storage

Screenshots

Data view

IRIS Class Explorer data view

Stats view

IRIS Class Explorer stats view

By default, the UI expects the backend API at:

/iris-table-stats/api

Install with IPM

If the package is available in your registry, install it with:

zpm "install iris-table-stats-frontend"

The module installs the frontend as an IRIS web application at:

/iris-table-stats-ui

Local build and container flow

npm run build compiles the frontend into dist/
docker compose build builds the frontend first, then copies dist/ into the IRIS image
IPM creates the IRIS web application and deploys the compiled frontend files into the CSP directory

Once running, open:

http://localhost:52773/iris-table-stats-ui/

Development notes

This repo contains the frontend packaging pattern for IRIS/IPM:

Vite production assets are built under dist/
the app is configured to run under a non-root base path
React Router handles both /iris-table-stats-ui/ and /iris-table-stats-ui/index.html
module.xml uses WebApplication and FileCopy so the bundle is installed as an IRIS web app

If you want the reusable agent workflow for doing this to another frontend app, see codex-ipm-frontend-module-runbook.md.

Fork and contribute

Fork it. Use it as a template for your own IRIS frontend modules. Improve it and send a pull request.

Useful contributions include:

UI and UX improvements
more views over backend metadata
better filtering and navigation
support for more backend endpoints from iris-table-stats
packaging refinements for IRIS/IPM deployment

iris-class-explorer Works