p5.plotsvg
v0.1.8
A Plotter-Oriented SVG Exporter for p5.js
0/weekUpdated 2 weeks agoMITUnpacked: 5.1 MB
Published by Golan Levin
npm install p5.plotsvgp5.plotSvg
A Plotter-Oriented SVG Exporter for p5.js
p5.plotSvg is a p5.js library for exporting SVG files tailored for pen plotting.
Version 0.1.8, January 22, 2026 • Initiated by Golan Levin (@golanlevin)
$3
* Download p5.plotSvg.js from this GitHub repo (raw)
* p5.plotSvg.js at npmjs.com: https://www.npmjs.com/package/p5.plotsvg
* p5.plotSvg.js at unpkg.com: https://unpkg.com/p5.plotsvg@latest/lib/p5.plotSvg.js
* p5.plotSvg.js at cdn.jsdelivr.net: https://cdn.jsdelivr.net/npm/p5.plotsvg@latest/lib/p5.plotSvg.js
* DOCUMENTATION
$3
* About p5.plotSvg
* Quickstart Installation
What the p5.plotSvg library IS*
What the p5.plotSvg library IS NOT*
* Example Programs
* Usage Notes
* Known Issues and Bugs
* Other Libraries and Related Work
* License and Code of Conduct
* Keywords
* Acknowledgments
---
About p5.plotSvg
The p5.plotSvg library allows the p5.js creative coding toolkit to generate SVG files specifically tailored for path-based vector output devices like the AxiDraw pen-plotter. Key advantages of p5.plotSvg are that it does not interfere with graphics performance during animation, and it is easy to add to projects.
Note that p5.plotSvg is not a general-purpose library for importing, exporting, optimizing, or rendering SVG files in p5.js: the p5.plotSvg library only considers the geometry of paths, and not how they are visually styled, with the expectation that the way a path eventually appears is a function of the type of physical tool you happen to have in your drawing machine. The p5.plotSvg library is known to be compatible with p5.js versions 1.4.2 through 1.11.11. Compatibility with p5.js version 2.x is forthcoming in 2026.
p5.plotSvg was initiated by Golan Levin in November 2024 as a resource for the Drawing with Machines course at CMU School of Art. It was created with encouragement and generous support from Bantam Tools, makers of the world's finest pen-plotting instruments.
---
Quickstart Installation
First, include p5.plotSvg.js in your project, alongside p5.js. You can do this by linking to an online copy of p5.plotSvg at unpkg.com or
cdn.jsdelivr.net in your project's index.html file. Alternatively, you can link to a local copy of p5.plotSvg.js, which you can download from this GitHub repo, here. The following example shows one way to include p5.plotSvg.js in your project's index.html file:
``html`
Next, create a p5.js file like the one below, called sketch.js, in the same directory as your index.html. When you run this example file in a web browser, you can press the s key to export an SVG file:
`js
// This is the sketch.js file.
// Press 's' to export the SVG.
// Note that p5.js is used in 'global mode'.
p5.disableFriendlyErrors = true; // hush
let bDoExportSvg = false;
function setup(){
// These canvas dimensions are 8.5"x11" at 96 dpi
createCanvas(816, 1056);
}
function keyPressed(){
if (key == 's'){
bDoExportSvg = true;
}
}
function draw(){
background(255);
if (bDoExportSvg){
beginRecordSvg(this, "myOutput.svg");
}
// Draw stuff here, such as:
line(0,0, mouseX, mouseY);
if (bDoExportSvg){
endRecordSvg();
bDoExportSvg = false;
}
}
`
$3
Alternatively, the p5.plotSvg library can also be used in p5's instance mode. Here's an alternative sketch.js file showing a minimal example of this:
`js
// This is a minimal sketch.js file for instance mode.
import p5plot from 'p5.plotsvg';
function sketch(context) {
context.setup = function() {
context.createCanvas(400, 400);
p5plot.beginRecordSvg(context, "output.svg");
context.circle(200, 200, 200);
p5plot.endRecordSvg();
};
};
new p5(sketch, document.getElementById("container"));
`
---
What the p5.plotSvg library IS:
* The p5.plotSvg library allows you to export a p5.js drawing as an SVG file that consists exclusively of scalable 2D vector paths, such as lines, arcs, shapes, polylines, and curves. We anticipate that you'll use the SVG files generated with this library to execute your drawings on a vector output device, such as a laser cutter, AxiDraw, or NextDraw pen-plotter.
* The p5.plotSvg library is intended for use with p5.js, and is modeled after the way in which PDF exporting and SVG exporting are implemented in Processing (Java). To use p5.plotSvg, you are expected to manage the timing of a beginRecordSvg() and endRecordSvg() function in your code.line()
The p5.plotSvg library works by temporarily overriding the functionality of the p5.js drawing commands. At the precise moment when you export the SVG, p5 drawing commands like and ellipse() are redefined so that they not only draw onscreen, but also* add their data to the SVG file. When the SVG is finished saving, the regular definitions of these functions are restored.
---
What the p5.plotSvg library IS NOT:
The p5.plotSvg library is not a "general purpose" p5-to-SVG exporter; it is intended for the specific needs of plotter enthusiasts*. Large parts of both the p5 and SVG specifications have been purposefully omitted, even where they are common to both. To ensure plotter compatibility, this library provides no support for exporting SVG files with graphics features that have no analogue in pen-plotting — such as pixel-based images, transparency, filters, shaders, blend modes, gradients, animation, or (even) fills and stroke weights. You might be able to render such things onscreen with p5.js, but they will not appear in the SVG vector files made with this library.
p5.plotSvg is not an SVG-based alternative renderer for the web browser. What you see onscreen is a standard p5.js canvas, not an SVG. If you want an SVG runtime in the browser as a substitute* for p5.js graphics, consider using Zenozeng's p5.js-svg library instead.
* This is not a library for loading, parsing, or displaying SVG files in p5.js. Zenozeng's p5.js-svg can do that as well.
This is not a library for optimizing the geometry of vector graphics for plotting or cutting. For example: no utilities are provided for sorting the order/direction of exported lines (using a TSP-like algorithm) to minimize your plotting time; for merging line segments with common endpoints into polylines (to eliminate unnecessary pen-up/pen-down movements); for de-duplicating multiple lines in precisely the same location; or for reordering* closed shapes from innermost to outermost for optimal laser-cutting. For such functionality, consider optimizing your SVGs with Antoine Beyeler's vpype for plotting, and/or Deepnest for laser cutting.
p5.plotSvg is not a library for vectorizing* pixel-based canvases rendered by p5.js. In other words, no utilities are provided for hatching or dithering that would "convert" the pixels on the screen into vector strokes. The only marks that get exported to SVG are the geometric paths you specify with vector-based p5.js drawing commands like line(), ellipse(), etc.
* This is not a library for computational geometry in p5. For problems like computing offset curves or shape-shape intersections, consider using libraries like Paper.js or Shapely.
---
Example Programs
This collection of examples shows how to generate plotter-friendly SVGs from p5.js using p5.plotSvg. All examples are mirrored in collections at editor.p5js.org and openProcessing.org. A visual index of examples is here.
1. plotSvg_smorgasbord: ⭐ Full demo of all p5.js drawing primitives exported to SVG. @editor • @openProcessing
2. plotSvg_hello_static: Simplest possible demo; all art in setup() only. @editor • @openProcessingsetup()
3. plotSvg_hello_animating: Simple demo; uses & draw() and a keypress. @editor • @openProcessing
4. plotSvg_generative: Simple "generative artwork"; press button to export. @editor • @openProcessing
5. plotSvg_drawing_recorder: Records a series of marks drawn by the user. @editor • @openProcessing
6. plotSvg_particle_paths: Accumulates the traces of some particles over time. @editor • @openProcessing
7. plotSvg_hatched_shapes: A trick for exporting hatched ("filled") SVG shapes. @editor • @openProcessing
8. plotSvg_instancemode: Example of SVG export in p5's instance mode. @editor
9. plotSvg_post_grouping: Merge groups of paths ("layers") computed at different times. @editor • @openProcessing
10. plotSvg_svg_font_text: Use monoline SVG fonts in your exported designs. @editor • @openProcessing
11. plotSvg_face_flipbook: Exports a tiny flipbook recording from a face-tracker. @openProcessing
---
Usage Notes
$3
* The p5.plotSvg library ignores p5.js fill() commands, and does not export SVG shapes with filled colors. To export filled shapes for a pen-plotter, consider implementing your own hatching method, as in this example.black
* SVGs produced with p5.plotSvg have a default stroke color, . This can be altered with setSvgDefaultStrokeColor(), which takes valid CSS color strings (e.g., 'red', '#ff0000', 'rgb(255,0,0)').stroke()
* To create SVG paths with non-default colors, simply use the command as usual. Note, however, that any such strokes will have additional style information added to their SVG representation (e.g. style="stroke:#ff0000;"); this could lead to potentially large file sizes, depending on your design. To restore the default stroke color, you can use stroke(getDefaultStrokeColor());.'red'
If you use stroke colors, the working assumption of p5.plotSvg is that you are using so to label different logical entities* — such as different color pens in a multi-pen plotter, different tools in a CNC machine, or different intensity settings in a laser cutter. For this reason, alpha (transparency) values are stripped out. I strongly recommend using just a small number of colors, and selecting easy-to-remember CSS color keyword names such as , 'green', 'blue', etc. stroke()
* This library only accepts p5 commands with the following types of arguments: CSS named colors in the set of 147 official SVG colors, hex formatted color strings, or as RGB/gray colors whose values range from 0-255. The p5 colorMode() command is not supported by p5.plotSvg, and calls to colorMode() may produce unpredictable results in your SVG.setSvgGroupByStrokeColor(true)
* When making multi-color plots, the function may be helpful in grouping paths you want plotted with the same color pen; see this example.
$3
The p5.plotSvg library offers two different ways of encoding graphics transformation operations (such as rotate(), translate(), scale(), shearX() and shearY()) into your SVG file. You can select the option you prefer using one the following functions during setup():
setSvgFlattenTransforms (true): The current transformation matrix is encoded into each SVG element. This leads to potentially larger SVG files, but graphical elements will appear with exactly the same transformations as they do in the corresponding p5 sketch. You might want to use this mode if your design concatenates many* transforms.
setSvgFlattenTransforms (false) (default option*): Graphics transforms are encoded into a hierarchy of SVG groups, each containing an atomic transform operation. This may produce smaller SVG files, depending on your design, but there is a possibility that different SVG rendering engines may accumulate the transforms with slightly different results.
* If no graphics transforms are used in a p5 sketch, none are encoded into the SVG file.
$3
p5.plotSvg offers two convenience functions which control how many digits of decimal precision are exported to SVG files. These can have a significant impact on SVG file size:
* setSvgCoordinatePrecision() – The default is 4 digits of precision for path coordinates, e.g. 3.1416
* – The default is 6 digits of precision for matrix transform data, e.g. 3.141593
---
Known Issues and Bugs:
* p5.plotSvg does not yet convert TTF/OTF text to outlines (though, this is planned). Additionally, p5.plotSvg does not embed TTF/OTF font data into SVGs; it refers to fonts by name and assumes they are installed on your host computer. You may need to perform additional operations on your SVGs (e.g. in Inkscape) to achieve desired typography. As an alternative, for quick typography with plotters, consider using single stroke fonts.
* As of p5.plotSvg v.0.1.x, non-default vertical textAlign() settings are not yet supported; only BASELINE currently works correctly.beginContour()
As of p5.plotSvg v.0.1.x, multi-contour* shapes (made with / endContour() etc.) are not yet unsupported. For the time being, users should encode each contour in its own beginShape() / endShape() block instead. curveVertex()
* As of p5.plotSvg v.0.1.x, there is a small discrepancy in the SVG output of polylines rendered with . Specifically, there is an error with the starting orientation of the first point of the polyline. WEBGL` mode. There is currently no support for converting 3D graphics to 2D, though this may be added later.
* As of p5.plotSvg v.0.1.x, this library is not intended to be used in
* p5.plotSvg v.0.1.x works with versions of p5.js as old as v.1.4.2. The forthcoming p5.js vertex API, which is due to come out with p5.js v.2.0, will likely cause breaking changes to portions of p5.plotSvg v.0.1.x.
* Other bugs and issues as listed here.
---
Other Libraries and Related Work
The following projects may be of interest to creative coders working with SVG files:
* p5.js-svg by Zenozeng allows for direct SVG rendering in p5.js sketches, offering an alternative SVG-based renderer for the web browser. Replacing the HTML canvas entirely, it supports a wide range of SVG elements but also aims for full compatibility with p5.js drawing functions.
* Paper.js by Jürg Lehni & Jonathan Puckey is a powerful open-source vector graphics scripting framework that runs on top of the HTML5 Canvas. It supports SVG import and export and offers a wide range of vector graphics manipulation features, such as path simplification and shape-shape intersection.
* vpype by Antoine Beyeler is a powerful command-line tool and Python library for preprocessing and optimizing vector graphics for plotting. It provides utilities for sorting paths, simplifying curves, and optimizing plotting jobs for pen plotters.
* Rune.js by Rune Madsen is a JavaScript library for creative coding, similar to p5.js. While it is not strictly a p5.js SVG exporter, it includes capabilities for working with vector graphics, including SVG import/export.
* p5-svg-test by Rune Madsen is a simple test for SVG generation using p5.js. This repository provides a proof of concept for exporting p5.js graphics to SVG format but is not a fully-featured library.
* PEmbroider by the Frank-Ratchye STUDIO for Creative Inquiry at Carnegie Mellon University is a library for computational embroidery using Processing (Java). It allows users to generate embroidery stitch paths from their Processing sketches, with support for various embroidery machine formats. It also supports SVG export.
* stitch.js by Stitchables is a JavaScript library for rendering SVG-based embroidery patterns. It allows users to convert drawings made with HTML5 Canvas or SVG paths into embroidery stitch paths.
* canvas2svg by Gliffy provides a way to export HTML5 Canvas content to SVG using JavaScript. It works by implementing a virtual canvas that mimics the CanvasRenderingContext2D interface, capturing drawing commands as SVG elements.
* Bob Cook's SVG Example provides an example demonstrating how to convert canvas-based drawings to SVG using a custom library in a jsFiddle example.
* ln by Michael Fogleman is a vector-based 3D renderer written in Go. It is used to produce 2D vector graphics depicting 3D scenes.
---
License and Code of Conduct
p5.plotSvg is released under the MIT License. The p5.plotSvg project adheres to a Code of Conduct adapted from the Contributor Covenant, version 1.1.0.
---
Keywords
Pen plotters, vector output, plotter art, p5.js, SVG, #plotterTwitter, creative coding, generative art, drawing machines, JavaScript library, EMSL AxiDraw, NextDraw, open-source software tools for the arts, #OSSTA.
---
Acknowledgments
This project was initiated by Golan Levin and made possible by support from the CMU School of Art, the Frank-Ratchye STUDIO for Creative Inquiry at Carnegie Mellon University, and Bantam Tools. Special thanks to the Processing Foundation's p5.js project, the students in my Drawing with Machines course at CMU, @lewi0622, @mariuswatz, @R4chel, @thrly, @Ucodia, @v3ga, @webholics, and everyone else in the community who has generously contributed by filing issues, thoughtful comments, and pull requests.
