ecstacy-tui 0.1.1

A Textual TUI for beautiful terminal data visualization from files, REST endpoints and more.

Ecstacy

Beautiful data, right in your terminal.

Ecstacy is a Textual TUI for visualizing data from local files, REST endpoints, SQL, and WebSockets -- with charts, tables, sparklines, gauges, heatmaps and a JSON explorer. Any source feeds any visualization through a typed DataSet contract, and dashboards compose multiple panels from a single YAML file.

Splash screen	Home screen
Sparkline chart	JSON explorer

Install

pip install ecstacy

Then launch it:

ecstacy

You can also run it directly with uv from a clone of this repo:

uv sync
uv run ecstacy

Quick start

Open a file directly:

ecstacy file sample.csv --chart line
ecstacy file data.csv --chart bar --x region --y value
ecstacy file metrics.parquet --chart table --max-rows 1000

Query a REST endpoint:

ecstacy rest https://api.example.com/items --json-path data.items --chart table

Run a DuckDB SQL query:

ecstacy sql "select 1 as a, 2 as b" --chart table
ecstacy sql "select * from 'data.csv'" --db :memory: --chart line

Stream from a WebSocket:

ecstacy socket ws://localhost:8765 --chart table

Open a multi-panel dashboard:

ecstacy dashboard ops.yaml

Inspect available themes, charts, or your config path:

ecstacy themes
ecstacy charts
ecstacy config path

Sources

Source	Formats / protocol
file	CSV, TSV, JSON, NDJSON, Parquet, log/text
rest	HTTP/HTTPS JSON endpoints with dotted json-path
sql	DuckDB queries (in-memory or file-backed)
socket	WebSocket streaming JSON records

Visualizations

table (sortable and searchable), line, bar, histogram, scatter, sparkline, gauge, heatmap (correlation matrix), json tree.

Keys

Home: o open, d dashboard, t theme, ? help, q quit. Chart: n/right next chart, p/left previous chart, t theme, esc back. Table: s sort by column, / focus search, type to filter, esc clear. Dashboard: m toggle grid/single layout, n/p cycle panels, esc back.

Configuration

Ecstacy layers configuration in this order (later wins):

1. Built-in defaults.
2. User config: $XDG_CONFIG_HOME/ecstacy/config.yaml (or ~/.config/ecstacy/config.yaml).
3. Project config: ./ecstacy.yaml in the current directory.
4. Environment variables prefixed ECSTACY_ (e.g. ECSTACY_THEME=ecstacy-light).
5. CLI flags.

Example config.yaml:

theme: ecstacy-dark
refresh: 0s
splash: true
max_rows: 1000

Dashboards

A dashboard YAML describes sources and panels. Each panel binds a source to a visualization with an optional column mapping. Source paths are resolved relative to the dashboard file, so dashboards are portable.

theme: ecstacy-dark
refresh: 5s
sources:
  - id: metrics
    kind: file
    path: ./sample.csv
panels:
  - source: metrics
    viz: line
    x: timestamp
    y: [revenue, margin]
  - source: metrics
    viz: bar
    category: region
    value: revenue

When refresh is set, dashboard panels re-fetch on that interval. Press m to switch between grid and single-panel layouts.

Architecture

• core -- DataSet contract, plugin registry, transforms, store, scheduler.
• sources -- file, rest, sql, socket -- all registered plugins.
• widgets -- visualization widgets, registered plugins.
• screens -- splash, home, chart, dashboard, modal, help.
• config -- defaults, Pydantic schema, layered config loader.
• theming -- Textual themes and the global stylesheet.

Sources resolve to a typed DataSet (DataFrame + schema + meta). Widgets bind to a DataSet and a column mapping, so any source can feed any visualization.

Extending

Add a source by subclassing Source and decorating it with @registry.sources.register("name"). Add a visualization by creating a widget with a set_data(dataset, mapping) method and decorating it with @registry.viz.register("name").

Known limitations

• No export to file yet (CSV/JSON/Markdown/PNG).
• No Excel, SQLite, clipboard, or stdin source yet.
• Live refresh applies to dashboards only (not single-source views).

License

MIT. See LICENSE.

Changelog

See CHANGELOG.md.