pathy-svg 0.1.3

Color arbitrary SVG paths by data values — turn any SVG into a heatmap

pathy-svg

Color arbitrary SVG paths by data values — turn any SVG into a heatmap.

Installation

pip install pathy-svg

Optional extras:

pip install pathy-svg[export]  # PNG, PDF, JPEG export (cairosvg + Pillow)
pip install pathy-svg[full]    # All features including Jupyter display

Quick Start

from pathy_svg import SVGDocument

doc = SVGDocument.from_file("examples/map.svg")

data = {
    "stomach": 0.5,
    "liver": 0.8,
    "heart": 0.3,
    "lung_l": 0.6,
    "lung_r": 0.7,
}

doc.heatmap(data, palette="YlOrRd").legend(title="Expression").save("output.svg")

Gradient and Pattern Fills

from pathy_svg import SVGDocument, GradientSpec

doc = SVGDocument.from_file("examples/map.svg")

# Gradient fill
doc.gradient_fill({
    "stomach": GradientSpec(start="#ff0000", end="#0000ff", direction="horizontal"),
}).save("gradient.svg")

# Pattern fill (string shorthand or PatternSpec)
doc.pattern_fill({
    "liver": "crosshatch",
    "heart": "dots",
}).save("patterned.svg")

Stroke Mapping and Highlighting

# Map data to stroke width and color
doc.stroke_map(data, width_range=(1, 5), palette="Reds").save("strokes.svg")

# Highlight specific elements, dim the rest
doc.highlight(["stomach", "liver"]).save("highlighted.svg")

Matching by Data Attributes

# Match elements by data-region instead of id
doc.heatmap({"north": 0.8, "south": 0.3}, key_attr="data-region").save("regions.svg")

# Works with all methods: recolor, stroke_map, highlight, annotate, etc.
doc.recolor({"north": "#ff0000"}, key_attr="data-region").save("recolored.svg")

Group Aggregation and Layers

# Color groups by the mean of their children's values
doc.heatmap_groups(data, agg="mean", palette="YlOrRd").save("groups.svg")

# Or use a custom aggregation function
doc.heatmap_groups(data, agg=lambda vals: max(vals) - min(vals)).save("range.svg")

# Compose multiple visualization layers
result = (
    doc.layers()
    .add("heat", lambda d: d.heatmap(data, palette="YlOrRd"))
    .add("borders", lambda d: d.stroke_map(data, palette="Greys"))
    .add("labels", lambda d: d.annotate({"stomach": "S", "liver": "L"}))
    .flatten()
)
result.save("layered.svg")

The source distribution includes a runnable examples/ directory with:

• examples/map.svg
• examples/data.csv
• examples/baseline.csv
• examples/treatment.csv

Features

• Heatmaps — data-driven coloring with any matplotlib colormap
• Categorical coloring — map categories to distinct colors
• Manual recolor — direct ID-to-color mapping
• Gradient fills — apply linear gradients (horizontal, vertical, diagonal) to elements
• Pattern fills — hatching, crosshatch, dots, and custom SVG patterns for accessibility
• Stroke mapping — map data to stroke width and/or color independently of fill
• Highlight/dim — emphasize specific elements while dimming others with desaturation
• Group aggregation — color <g> elements by aggregating children (mean, sum, min, max, median, or custom callable)
• Multi-layer system — compose named visualization layers with show/hide/reorder
• Diff visualization — compare datasets with delta, ratio, log2ratio, or percent change modes
• Side-by-side comparison — multiple datasets in a single SVG
• Legends — gradient, discrete, and categorical legend types
• Annotations — text labels at element centroids or custom positions
• Tooltips — hover text via SVG <title> or CSS popups
• Animations — CSS keyframe effects (pulse, fade_in, blink, sequential)
• Export — PNG, PDF, JPEG via cairosvg and Pillow
• Jupyter — inline SVG display with _repr_svg_ and _repr_mimebundle_
• CLI — heatmap, inspect, validate, guide, diff, and export commands
• Flexible element matching — match elements by id, data-* attributes, or class via key_attr
• Immutable API — method chaining with new instances returned on each call
• DataFrame support — load data directly from pandas
• Theme presets — medical, geographic, heatmap_classic

CLI Usage

# Create a heatmap
pathy-svg heatmap examples/map.svg examples/data.csv --id-col organ --value-col expression --palette YlOrRd --legend -o out.svg

# Inspect SVG structure
pathy-svg inspect examples/map.svg

# Validate data IDs against SVG
pathy-svg validate examples/map.svg examples/data.csv --id-col organ

# Compare two datasets
pathy-svg diff examples/map.svg examples/baseline.csv examples/treatment.csv --id-col organ --value-col expression --mode delta -o diff.svg

# Export to PNG
pathy-svg export examples/map.svg -o map.png --width 1200

API Overview

Loading

Method	Description
SVGDocument.from_file(path)	Load from file path
SVGDocument.from_string(svg)	Load from SVG string
SVGDocument.from_url(url)	Load from URL
SVGDocument.from_dataframe(df, ...)	Load SVG path from a DataFrame column

Coloring

Method	Description
.heatmap(data, palette=...)	Apply data-driven coloring
.heatmap_from_dataframe(df, ...)	Heatmap from pandas DataFrame
.recolor(color_map)	Manual ID-to-color mapping
.recolor_by_category(category_map)	Categorical coloring
.gradient_fill(gradients)	Apply linear gradients to elements
.pattern_fill(patterns)	Apply hatching, dots, or custom patterns
.stroke_map(data, width_range=..., palette=...)	Map data to stroke width/color
.highlight(ids)	Emphasize elements, dim the rest
.heatmap_groups(data, agg=...)	Color groups by aggregating children

Layers

Method	Description
.layers()	Create a LayerManager for composing layers
LayerManager.add(name, fn)	Add a named layer
LayerManager.hide(name) / .show(name)	Toggle layer visibility
LayerManager.reorder(names)	Change layer order
LayerManager.flatten()	Render all visible layers to an SVGDocument

Visualization

Method	Description
.legend(title=..., position=...)	Add a legend
.diff(baseline, treatment, mode=...)	Diff two datasets
.compare(datasets, layout=...)	Side-by-side comparison
.annotate(labels)	Add text labels
.add_tooltips(texts)	Add hover tooltips
.animate(effect=..., duration=..., loop=...)	CSS animations

Inspection

Method	Description
.path_ids	List of all path element IDs
.group_ids	List of all group element IDs
.element_ids	List of all element IDs
.viewbox	SVG viewBox as ViewBox namedtuple
.dimensions	(width, height) tuple
.inspect_paths()	Detailed metadata for all colorable elements
.validate_ids(ids)	Check which IDs match SVG elements

Export

Method	Description
.save(path)	Write SVG to file
.to_string()	SVG as string
.to_bytes()	SVG as bytes
.to_png(path)	Export to PNG
.to_pdf(path)	Export to PDF
.to_jpeg(path)	Export to JPEG
.show()	Display in Jupyter

License

This project is licensed under the GNU General Public License v3.0 — see the LICENSE file for details.