rekhly is a native Python API — it never exposes JS objects. Charts (ChartXY, Heatmap, FinancialChart, PolarChart, SmithChart, Spectrogram, DensityMap, Oscilloscope, EyeDiagram, VectorField, ParallelCoordinate, DataGrid, LinearGauge, BarChart, Chart3D, and more) build up data, then reach the runtime through a transport Bridge.
pip install -e .
pytest python/testsimport rekhly
chart = rekhly.ChartXY(title="Temperature")
series = chart.add_line_series()
series.add(x=[1, 2, 3], y=[10, 20, 15])
chart.open() # defaults to InProcessBridge — no networkChartXY carries the same chrome as the JS engine — markers (normal or delta), threshold limit lines, and highlighted detection-bar bands — all serialized alongside the series data:
chart.add_marker("peak", x=12, y=98, label="Peak")
chart.add_limit_line("max", "horizontal", 100, direction="above")
chart.add_detection_bar("spike", x0=2, x1=4, severity="critical")append_samples streams through whatever bridge open() was given — each call sends an {"type": "append"} message, not a full re-serialize:
series = chart.add_line_series()
chart.open(live=True)
series.append_samples([1], [2])
chart.close() # closes the bridge; safe to call twicerekhly.bridge defines a Bridge protocol (open/send/close) with two implementations:
InProcessBridge — the default. No network; records every message sent, for tests and offline chart construction.WebSocketBridge(url) — a real RFC 6455 client, standard-library only, for driving the local Rekhly runtime (e.g. apps/api's /ws/v1/streams/{id}):chart.open(live=True, bridge=rekhly.WebSocketBridge("ws://localhost:8000/ws/v1/streams/demo"))chart.on("click", lambda payload: ...)
chart.emit("click", {"x": 1, "y": 2})rekhly.set_license(value) / rekhly.get_license() read/write REKHLY_LICENSE from the environment — never hardcode secrets.
Got any questions? I'm happy to help with products, licensing, and examples.