Crate charcoal 

A declarative, DataFrame-native chart library for Polars.

charcoal turns a Polars polars::frame::DataFrame into a publication-quality SVG, standalone HTML, or PNG with a single builder chain — no browser, no Python, no C FFI.

Feature Flags

Feature	Enables	Extra dependency
(default)	SVG and HTML output	—
static	Chart::save_png via pure-Rust resvg	resvg
notebook	[Chart::display] inline in evcxr Jupyter	evcxr_runtime

Quickstart

use charcoal::{Chart, Theme};
use polars::prelude::*;

fn main() -> Result<(), Box<dyn std::error::Error>> {
    let df = CsvReader::from_path("iris.csv")?.finish()?;

    let chart = Chart::scatter(&df)
        .x("sepal_length")
        .y("sepal_width")
        .color_by("species")
        .title("Iris Dataset")
        .theme(Theme::Default)
        .build()?;

    chart.save_html("iris.html")?;
    Ok(())
}

Examples

Runnable examples for every chart type live in the examples/ directory:

cargo run --example scatter_iris
cargo run --example line_timeseries
cargo run --example bar_categorical
cargo run --example histogram_distribution
cargo run --example heatmap_correlation
cargo run --example box_plot_groups
cargo run --example area_stacked

Notebook Usage

To display charts inline in an evcxr Jupyter notebook, enable the notebook feature:

[dependencies]
charcoal = { version = "0.1", features = ["notebook"] }

In your first notebook cell, load the dependency:

:dep charcoal = { version = "0.1", features = ["notebook"] }
use charcoal::Chart;

Then call .display() as the last expression in any subsequent cell:

let chart = Chart::scatter(&df).x("x").y("y").build()?;
chart.display();

Repository

https://github.com/your-handle/charcoal

Structs

Chart

The fully-rendered output of a .build() call.

Enums

BinMethod

Strategy used to choose the number of histogram bins.

CharcoalError

Errors that can be returned by charcoal’s public API.

CharcoalWarning

Non-fatal diagnostics attached to a successfully-built crate::Chart.

ColorScale

Continuous color gradient used to encode numeric values in heatmaps.

DashStyle

Dash style for line and area series.

FillMode

Controls the baseline of the filled area.

NullPolicy

How null y-values are handled in connected series (Line, Area).

Orientation

Bar chart orientation.

PointDisplay

Controls which individual data points are rendered on top of each box.

Theme

Overall visual style applied to a chart.