charton 0.4.1

A high-level, layered charting system for Rust, designed for Polars-first data workflows and multi-backend rendering.
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279

use crate::Precision;
use crate::chart::Chart;
use crate::core::context::PanelContext;
use crate::core::layer::{MarkRenderer, PolygonConfig, RenderBackend};
use crate::encode::y::StackMode;
use crate::error::ChartonError;
use crate::mark::bar::MarkBar;
use crate::visual::color::SingleColor;
use polars::prelude::*;

impl MarkRenderer for Chart<MarkBar> {
    fn render_marks(
        &self,
        backend: &mut dyn RenderBackend,
        context: &PanelContext,
    ) -> Result<(), ChartonError> {
        let df = &self.data.df;
        if df.height() == 0 {
            return Ok(());
        }

        let mark_config = self
            .mark
            .as_ref()
            .ok_or_else(|| ChartonError::Mark("MarkBar config missing".into()))?;

        // --- STEP 1: Encoding & Scales ---
        let x_enc = self
            .encoding
            .x
            .as_ref()
            .ok_or(ChartonError::Encoding("X missing".into()))?;
        let y_enc = self
            .encoding
            .y
            .as_ref()
            .ok_or(ChartonError::Encoding("Y missing".into()))?;
        let x_scale = context.coord.get_x_scale();
        let y_scale = context.coord.get_y_scale();

        let is_stacked = y_enc.stack != StackMode::None;
        let x_field = &x_enc.field;
        let color_field = self.encoding.color.as_ref().map(|c| c.field.as_str());

        // PIE MODE: An empty X field indicates a single-axis radial layout.
        let is_pie_mode = x_field.is_empty();

        // --- STEP 2: Coordinate & Mode Detection ---
        let hints = context.coord.layout_hints();
        let is_polar = hints.needs_interpolation;
        let needs_nightingale_sqrt = is_polar && !is_pie_mode;

        // --- STEP 3: Visual Parameter Resolution ---
        let eff_width = mark_config.width.unwrap_or(hints.default_bar_width);
        let eff_spacing = mark_config.spacing.unwrap_or(hints.default_bar_spacing);
        let eff_span = mark_config.span.unwrap_or(hints.default_bar_span);
        let eff_stroke = mark_config.stroke.unwrap_or(hints.default_bar_stroke);
        let eff_stroke_width = mark_config
            .stroke_width
            .unwrap_or(hints.default_bar_stroke_width);

        // --- STEP 4: X-Axis Step Calculation ---
        let n0 = x_scale.normalize(0.0);
        let n1 = x_scale.normalize(1.0);
        let unit_step_norm = (n1 - n0).abs();

        // --- STEP 5: Grouping & Layout Strategy ---
        let x_uniques_count = df.column(x_field)?.n_unique()?;
        let total_rows = df.height();

        let n_groups = if is_pie_mode {
            1.0
        } else {
            (total_rows as f64 / x_uniques_count as f64).max(1.0)
        };

        let bar_width_data = if is_stacked || n_groups <= 1.0 {
            eff_width.min(eff_span)
        } else {
            eff_span / (n_groups + (n_groups - 1.0) * eff_spacing)
        };

        let bar_width_norm = bar_width_data * unit_step_norm;
        let spacing_norm = bar_width_norm * eff_spacing;

        // --- STEP 6: Partitioning ---
        let is_self_mapping = color_field.is_some_and(|cf| cf == x_field);
        let groups = match color_field {
            Some(col) if !is_self_mapping => df.partition_by_stable([col], true)?,
            _ => vec![df.clone()],
        };

        // --- STEP 7.0: Global Total Calculation (CRITICAL FOR PIE PERCENTAGES) ---
        // We calculate this before the loop to ensure percentages are relative to the whole.
        let global_total = if is_pie_mode {
            df.column(&y_enc.field)?.f64()?.sum().unwrap_or(1.0)
        } else {
            1.0
        };

        let mut stack_acc = Vec::new();

        // --- STEP 7: Render Loop ---
        for (group_idx, group_df) in groups.iter().enumerate() {
            let group_color_fixed = if !is_self_mapping {
                Some(self.resolve_group_color(group_df, context, &mark_config.color)?)
            } else {
                None
            };

            let x_series = group_df.column(x_field)?.as_materialized_series();
            let y_series = group_df.column(&y_enc.field)?.as_materialized_series();

            let x_norms = x_scale.scale_type().normalize_series(x_scale, x_series)?;
            let y_vals: Vec<f64> = y_series.f64()?.into_no_null_iter().collect();

            for (i, (opt_x_n, y_val)) in x_norms.into_iter().zip(y_vals).enumerate() {
                if y_val == 0.0 && !is_stacked {
                    continue;
                }

                let final_color = if is_self_mapping {
                    let row_df = group_df.slice(i as i64, 1);
                    self.resolve_group_color(&row_df, context, &mark_config.color)?
                } else {
                    group_color_fixed.unwrap_or(mark_config.color)
                };

                let x_tick_n = opt_x_n.unwrap_or(0.0);

                // --- RESOLVE Y-BOUNDS (Radius Calculation) ---
                let (y_low_n, y_high_n) = if is_stacked {
                    if stack_acc.len() <= i {
                        stack_acc.push(0.0);
                    }
                    let start = stack_acc[i];
                    let end = start + y_val;
                    stack_acc[i] = end;

                    if needs_nightingale_sqrt {
                        (
                            y_scale.normalize(start).sqrt(),
                            y_scale.normalize(end).sqrt(),
                        )
                    } else {
                        (y_scale.normalize(start), y_scale.normalize(end))
                    }
                } else {
                    let n_val = y_scale.normalize(y_val);
                    let final_n = if needs_nightingale_sqrt {
                        n_val.sqrt()
                    } else {
                        n_val
                    };
                    (y_scale.normalize(0.0), final_n)
                };

                // --- POSITIONING & DODGING ---
                let offset_norm = if !is_stacked && n_groups > 1.0 {
                    (group_idx as f64 - (n_groups - 1.0) / 2.0) * (bar_width_norm + spacing_norm)
                } else {
                    0.0
                };

                let x_center_n = x_tick_n + offset_norm;
                let left_n = x_center_n - bar_width_norm / 2.0;
                let right_n = x_center_n + bar_width_norm / 2.0;

                // --- GEOMETRIC TRANSFORMATION ---
                let rect_path = if is_pie_mode {
                    vec![
                        (y_low_n, left_n),
                        (y_low_n, right_n),
                        (y_high_n, right_n),
                        (y_high_n, left_n),
                    ]
                } else {
                    vec![
                        (left_n, y_low_n),
                        (left_n, y_high_n),
                        (right_n, y_high_n),
                        (right_n, y_low_n),
                    ]
                };

                let pixel_points = if hints.needs_interpolation {
                    context.transform_path(&rect_path, true)
                } else {
                    rect_path
                        .iter()
                        .map(|(nx, ny)| context.coord.transform(*nx, *ny, &context.panel))
                        .collect::<Vec<_>>()
                };

                backend.draw_polygon(PolygonConfig {
                    points: pixel_points
                        .into_iter()
                        .map(|(x, y)| (x as Precision, y as Precision))
                        .collect(),
                    fill: final_color,
                    stroke: eff_stroke,
                    stroke_width: eff_stroke_width as Precision,
                    fill_opacity: mark_config.opacity as Precision,
                    stroke_opacity: mark_config.opacity as Precision,
                });

                // --- 8. PIE/DONUT CENTERED LABELING ---
                if is_pie_mode {
                    let theme = context.spec.theme;

                    // Calculate percentage using the pre-calculated global_total.
                    let percentage = if y_enc.normalize {
                        y_val * 100.0
                    } else {
                        (y_val / global_total) * 100.0
                    };

                    // Only render labels for sectors > 3% to maintain clarity.
                    if percentage > 3.0 {
                        let label_text = format!("{:.1}%", percentage);

                        // Centroid logic: Midpoint of angular and radial spans.
                        let mid_angle_n = (y_low_n + y_high_n) / 2.0;
                        let mid_radius_n = (left_n + right_n) / 2.0;

                        let (lx, ly) =
                            context
                                .coord
                                .transform(mid_angle_n, mid_radius_n, &context.panel);

                        backend.draw_text(crate::core::layer::TextConfig {
                            x: lx as Precision,
                            y: ly as Precision,
                            text: label_text,
                            font_size: (theme.tick_label_size - 1.0) as Precision,
                            font_family: theme.tick_label_family.clone(),
                            color: SingleColor::new("white"),
                            text_anchor: "middle".into(),
                            font_weight: "bold".into(),
                            opacity: mark_config.opacity as Precision,
                        });
                    }
                }
            }
        }
        Ok(())
    }
}

impl Chart<MarkBar> {
    /// Resolves the color for a specific data subset.
    /// In self-mapping mode (x == color), this is called per row.
    /// In grouping mode (x != color), this is called once per partition.
    fn resolve_group_color(
        &self,
        df: &DataFrame,
        context: &PanelContext,
        fallback: &SingleColor,
    ) -> Result<SingleColor, ChartonError> {
        if let Some(ref mapping) = context.spec.aesthetics.color {
            let s = df.column(&mapping.field)?.as_materialized_series();
            let s_trait = mapping.scale_impl.as_ref();

            // Normalize the first value of the provided DataFrame slice.
            let norms = s_trait
                .scale_type()
                .normalize_series(s_trait, &s.head(Some(1)))?;
            let norm = norms.get(0).unwrap_or(0.0);

            // Map normalized value to color via palette.
            Ok(s_trait
                .mapper()
                .map(|m| m.map_to_color(norm, s_trait.logical_max()))
                .unwrap_or_else(|| *fallback))
        } else {
            Ok(*fallback)
        }
    }
}