Summary

Create a separate fluxopt-plot package that provides interactive plotly-based visualization for Result objects. Depends on fluxopt ~= X.Y, keeps heavy plotting dependencies out of core.

Design

The package monkey-patches a .plot accessor onto Result at import time. All plot methods return a PlotResult(data, figure) for both inspection and customization.

Plot methods

balance(node, *, select, lines, threshold, show, **plotly_kwargs)
carrier_balance(carrier, *, select, unit, threshold, show, **plotly_kwargs)
storage(storage, *, select, unit, charge_state_color, show, **plotly_kwargs)
charge_states(storages, *, select, show, **plotly_kwargs)
flows(*, start, end, component, select, unit, show, **plotly_kwargs)
sizes(*, select, max_size, show, **plotly_kwargs)
duration_curve(variables, *, select, normalize, show, **plotly_kwargs)
heatmap(variables, *, select, reshape, show, **plotly_kwargs)
effects(aspect, *, effect, by, select, show, **plotly_kwargs)
sankey.flows(*, aggregate, select, show, **plotly_kwargs)
sankey.sizes(*, select, show, **plotly_kwargs)
sankey.peak_flow(*, select, show, **plotly_kwargs)
sankey.effects(*, select, show, **plotly_kwargs)

Key features

• Smart defaults: flows with fixed_profile shown as lines, dispatch as bars
• lines= override: explicit control over which flows render as lines vs bars
• Period/scenario handling: periods faceted, scenarios aggregated (mean + band for lines)
• select= dict: xarray-style filtering (select={"period": "2030"})
• PlotResult: .data (xr.Dataset), .figure (go.Figure), .show(), .to_html()

Dependencies

[project]
dependencies = ["fluxopt~=X.Y", "plotly>=6.0", "xarray_plotly>=0.0.13"]

Acceptance Criteria

• Separate PyPI-installable package
• Monkey-patches .plot onto Result at import time
• All listed plot methods implemented
• Every method tested with data_only=True (assert on Dataset)
• Smoke tests verify figure creation does not raise

Refs

Part of #47 — Architecture: Package Separation, SolutionResult Design, and Stability Tiers