p5.plotSvg is a p5.js library for exporting SVG files tailored for pen plotting.
Version 0.1.2, December 6, 2024 • by Golan Levin (@golanlevin)
- Download p5.plotSvg.js from GitHub (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@0.1.2/lib/p5.plotSvg.js
- p5.plotSvg.js at cdn.jsdelivr.net: https://cdn.jsdelivr.net/npm/p5.plotsvg@latest/lib/p5.plotSvg.js
- Documentation
- 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
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. 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 is known to be compatible with p5.js versions 1.4.2 through 1.11.2.
Some advantages of p5.plotSvg are that: it is compatible with current versions of p5.js; it is easy to add to projects; and it does not interfere with graphics performance during animation (except at the precise moment of exporting a file).
p5.plotSvg was developed 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.
First, include p5.plotSvg.js in your project. You can do this by linking to an online copy of p5.plotSvg at unpkg.com or
cdn.jsdelivr.net in your 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:
<!-- This is the index.html file -->
<html>
<head>
<script src="https://cdnjs.cloudflare.com/ajax/libs/p5.js/1.11.2/p5.js"></script>
<script src="https://cdn.jsdelivr.net/npm/p5.plotsvg@latest/lib/p5.plotSvg.js"></script>
</head>
<body>
<script src="sketch.js"></script>
</body>
</html>
Next, make a p5.js file like the one below, called sketch.js, in the same directory as your index.html. When you run it in a web browser, you can press the s key to export an SVG file.
// This is the sketch.js file.
// Press 's' to export the SVG.
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;
}
}
- 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. I 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()andendRecordSVG()function. - 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
line()andellipse()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.
- 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 computational geometry in p5. For problems like computing offset curves or shape-shape intersections, consider using libraries like Paper.js or Shapely.
- This is not a library for optimizing 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 reduce your plotting time; for merging line segments with common endpoints; for de-duplicating multiple lines in the same location; or for reordering graphical elements 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 ones you specify with vector-based p5.js drawing commands like
line(),ellipse(), etc.
These examples show how to generate plotter-friendly SVGs from p5.js using p5.plotSvg. Examples are mirrored at editor.p5js.org and openProcessing.org. A visual index of examples is here.
- plotSvg_smorgasbord: ⭐ Full demo of all p5.js drawing primitives exported to SVG. @editor • @openProcessing
- plotSvg_hello_static: Simplest possible demo; all art in
setup()only. @editor • @openProcessing - plotSvg_hello_animating: Simple demo; uses
setup()&draw()and a keypress. @editor • @openProcessing - plotSvg_generative: Simple "generative artwork"; press button to export. @editor • @openProcessing
- plotSvg_drawing_recorder: Records a series of marks drawn by the user. @editor • @openProcessing
- plotSvg_particle_paths: Accumulates the traces of some particles over time. @editor • @openProcessing
- plotSvg_hatched_shapes: A trick for exporting hatched ("filled") SVG shapes. @editor • @openProcessing
- plotSvg_face_flipbook: Exports a tiny flipbook recording from a face-tracker. @openProcessing
Examples awaiting implementation (more soon!):
- plotSvg_hershey: Displays Hershey Font text using Lingdong Huang's p5-hershey-js
- 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. - SVGs produced with p5.plotSvg have a default stroke color,
black. This can be altered withsetSvgDefaultStrokeColor(), which takes valid CSS color strings (e.g.,'red','#ff0000','rgb(255,0,0)'). - To create SVG paths with non-default colors, simply use the
stroke()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 usestroke(getDefaultStrokeColor());. - 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
'red','green','blue', etc. - This library only accepts p5
stroke()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 p5colorMode()command is not supported by p5.plotSvg, and calls tocolorMode()may produce unpredictable results in your SVG.
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.
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.3.1416setSvgTransformPrecision()- The default is 6 digits of precision for matrix transform data, e.g.3.141593
- As of p5.plotSvg v.0.1.x, non-default vertical
textAlign()settings are not yet supported; onlyBASELINEcurrently works correctly. - As of p5.plotSvg v.0.1.x, multi-contour shapes (made with
beginContour()/endContour()etc.) are not yet unsupported. For the time being, users should encode each contour in its ownbeginShape()/endShape()block instead. - As of p5.plotSvg v.0.1.x, there is a small discrepancy in the SVG output of polylines rendered with
curveVertex(). Specifically, there is an error with the starting orientation of the first point of the polyline. - As of p5.plotSvg v.0.1.x, this library is not intended to be used in
WEBGLmode. There is currently no support for converting 3D graphics to 2D, though this may be added later. - 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.
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. It supports a wide range of SVG elements but also aims for full compatibility with p5.js drawing functions. Note: As of December 2024, p5.js-svg is only compatible with p5.js up to v.1.6.0.
- 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.
- 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.
- 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.
- vpype by Antoine Beyeler is a 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.
- 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.
- ln by Michael Fogleman is a vector-based 3D renderer written in Go. It is used to produce 2D vector graphics depicting 3D scenes.
- 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.
- 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.
- 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.
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.
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.
This project was made possible by support from the CMU School of Art, the Frank-Ratchye STUDIO for Creative Inquiry at Carnegie Mellon University, and Bantam Tools.
