Skip to main content

SVGraft

SVGraft splits a large SVG file into a multipage PDF, tiling it across standard print pages with configurable margins and alignment corner marks. It is designed for printing oversized vector graphics (such as posters, banners, or blueprints) on ordinary home or office printers.

I developed the first version at an FRC competition to print a 25x30 poster for Engineering Inspiration judging on letter paper.


Features

  • Splits any SVG (with a viewBox defined in mm or in) into a grid of tiles that each fit on a single page
  • Outputs a single merged PDF ready to print and assemble
  • Configurable page size, margins, and tile orientation
  • Automatic orientation optimization — chooses portrait or landscape to minimize page count
  • Corner alignment marks and a dashed cut border on every tile to aid physical assembly
  • Progress display and tile summary table via rich
  • Scale control — scale the output up or down before tiling, using percentage, target page count, or target physical dimensions
  • PDF input (optional) — accepts PDF files directly as input (first page is converted and tiled)

Requirements

  • Python 3.9+
  • CairoSVG and its native Cairo library
  • See requirements.txt for the full list of Python dependencies
  • Optional: PyMuPDF for PDF input (installed with the pymupdf extra, see below)
  • Tests only: the test extra (pdf2image) and Poppler, which it uses to render the output PDF

Install system dependencies (Debian/Ubuntu):

sudo apt-get install libcairo2-dev          # required
sudo apt-get install poppler-utils          # only needed to run the tests

Installation

pip install SVGraft

To also enable PDF input:

pip install "SVGraft[pymupdf]"

PyMuPDF is a large download and is licensed under the AGPL-3.0, so it is an opt-in extra rather than a required dependency. Without it, SVG input works normally and PDF input prints a message telling you to install the extra.

Or install it from source:

git clone <repository-url>
cd SVGraft
pip install .                 # or: pip install -e ".[test,pymupdf]" for development

Usage

Command Line

svgraft input_file [options]
Argument Default Description
input_file (required) Path to the input SVG or PDF file
-o, --output ./output_merged.pdf Path for the output PDF file
--page-width 8.5 Page width in inches
--page-height 11.0 Page height in inches
--margin 7.0 Margin on each page in millimeters
--scale — Scale factor as a multiplier (1.5) or percentage (150%)
--output-width — Target assembled width (e.g. 48in, 1200mm)
--output-height — Target assembled height (e.g. 36in, 900mm)
--pages-wide — Target number of tiles across
--pages-tall — Target number of tiles tall
--no-optimize-orientation — Disable automatic portrait/landscape optimization
--border-stroke 0.1 Border stroke width in millimeters
--corner-stroke 0.1 Corner triangle stroke width in millimeters

Running svgraft with no input file, or with a file that does not exist, prints the usage text and exits with status 2.

Examples:

Split a poster onto US Letter pages with default settings:

svgraft my_poster.svg

Use A4 pages with a 10 mm margin and a custom output path:

svgraft my_poster.svg --page-width 8.27 --page-height 11.69 --margin 10 -o poster_tiled.pdf

Scale the SVG to 150% of its native size before tiling:

svgraft my_poster.svg --scale 150%

Scale so the assembled output is exactly 4 feet wide:

svgraft my_poster.svg --output-width 48in

Tile the SVG to span exactly 3 pages wide:

svgraft my_poster.svg --pages-wide 3

Tile the first page of a PDF (requires SVGraft[pymupdf]):

svgraft document.pdf --scale 200%

Python API

from svgraft.cli import split_svg
from svgraft.config import ProcessConfig

config = ProcessConfig(
    input_file="my_poster.svg",
    page_width_in=8.5,
    page_height_in=11.0,
    margin_mm=6.35,
    output_file="output.pdf",
    optimize_tile_orientation=True,
    scale_factor=1.5,      # optional: scale to 150%
    total_width_in=48.0,   # optional: target assembled width
    pages_wide=3,          # optional: target tile columns
)

with open(config.input_file, "rb") as f:
    split_svg(f, config)

SVG Requirements

The input SVG must have:

  • A viewBox attribute (e.g. viewBox="0 0 1200 900")
  • A width attribute defined in inches (in) or millimeters (mm)

Example SVG header:

<svg xmlns="http://www.w3.org/2000/svg"
     viewBox="0 0 2400 1800"
     width="24in" height="18in"/>

Scale Control

SVGraft can scale your SVG before tiling using any of three modes. All modes account for per-page margins when computing the final tile grid.

Percentage / multiplier scaling

Scale the SVG up or down relative to its native size:

svgraft poster.svg --scale 150%
svgraft poster.svg --scale 0.75

Target physical dimensions

Specify the desired real-world width or height of the assembled output. SVGraft derives the required scale factor from the SVG's native physical size:

svgraft poster.svg --output-width 48in
svgraft poster.svg --output-height 1200mm

Supported units: in, mm, m.

Target page count

Specify how many tiles wide or tall the output should span. SVGraft back-calculates the scale factor to hit that grid exactly. Orientation optimization is disabled in this mode since you have expressed an explicit layout intent:

svgraft poster.svg --pages-wide 3
svgraft poster.svg --pages-tall 2

PDF Input

SVGraft accepts PDF files as input in addition to SVG, provided the optional PyMuPDF dependency is installed (pip install "SVGraft[pymupdf]"). The first page of the PDF is converted to SVG internally, with its size converted from PDF points to inches, and run through the standard tiling pipeline:

svgraft document.pdf
svgraft document.pdf --scale 200% -o tiled_output.pdf

Note: Only the first page of a multipage PDF is currently processed.


How It Works

  1. Parse the input file. SVG files are parsed directly; PDF inputs have their first page converted to SVG via PyMuPDF before continuing.
  2. Apply scaling — if a scale mode is specified (--scale, --output-width, --output-height, --pages-wide, or --pages-tall), the units-per-inch value is adjusted so the scaled content maps correctly onto the tile grid.
  3. Compute tiles — calculate the grid of columns and rows needed so each tile fits within the usable page area (page size minus margins).
  4. Optimize orientation — if rotating tiles from portrait to landscape (or vice versa) produces fewer total pages, SVGraft automatically swaps the dimensions and re-tiles. Disabled when --pages-wide or --pages-tall is used.
  5. Render tiles — for each grid cell, clip and translate the SVG, add corner alignment triangles and a dashed border, then render to PDF via CairoSVG.
  6. Merge — all per-tile PDFs are merged into a single output PDF using pypdf.

Assembly

Each page includes:

  • Corner triangles at the four content corners for precise cut-and-align assembly
  • A dashed border showing the exact content boundary

Pages are ordered left to right, then top to bottom. To assemble a print: trim half of the pages to their dashed border, then align them to the corner marks of the adjacent tiles.


Configuration Defaults

Defaults are defined in the DEFAULTS dictionary in src/svgraft/config.py:

Setting Default Description
page_width_in 8.5 Page width (inches)
page_height_in 11 Page height (inches)
margin_mm 7.0 Per-page margin (mm)
corner_triangle_size_mm 2.0 Size of corner alignment triangles (mm)
corner_triangle_stroke 0.1 Stroke width of corner marks (mm)
border_stroke_mm 0.1 Stroke width of the dashed border (mm)
optimize_tile_orientation True Auto-flip orientation for fewer pages
output_file ./output_merged.pdf Where the merged PDF is written

Running Tests

pip install -e ".[test,pymupdf]"
python -m unittest discover -s tests -p "test*.py" -v

Install the package in editable mode first. If an older copy of SVGraft is installed in your environment, the tests will import that instead of your working tree, so run pip uninstall -y svgraft before the editable install.

  • tests/test_core.py covers unit parsing, tiling math, config defaults, page layout, rendering helpers, PDF merging, end-to-end splits (page counts, page sizes, scaling, orientation), the sample posters, and the command line.
  • tests/testSVGraft.py splits tests/data/test_poster_inches.svg onto US Letter pages and checks the output. It opens a preview of the first page on machines with a display and skips the preview on headless machines and in CI.
  • The PDF input tests are skipped automatically if PyMuPDF is not installed.

A few tests are marked as expected failures. They document a known bug in count_not_none (src/svgraft/util.py), which stops the "only one size option allowed" check from firing.

Continuous integration

Gitea Actions workflows live in .gitea/workflows/:

Workflow Trigger What it does
test.yml every push / pull request Installs the package and runs the unit tests
publish.yml push of a v* tag Builds and uploads to PyPI
publish-test.yml manual (workflow_dispatch) Builds and uploads to TestPyPI

The runner needs libcairo2 and poppler-utils available for the test workflow.


Releasing

  1. Bump __version__ in src/svgraft/__init__.py.

  2. Commit and push.

  3. Tag the release with the same version, prefixed by v, and push the tag:

    git tag v0.0.6
    git push origin v0.0.6
    

publish.yml checks that the tag matches __version__, builds the sdist and wheel, and uploads them with twine. PyPI never accepts the same version twice, so always bump the version for a new release.

Repository secrets (Settings → Actions → Secrets):

Secret Used by Where to get it
PYPI_API_TOKEN publish.yml pypi.org → Account settings → API tokens
TEST_PYPI_API_TOKEN publish-test.yml test.pypi.org → Account settings → API tokens

To try a TestPyPI release, run the publish-test.yml workflow from the repository's Actions tab, then install it with:

pip install --index-url https://test.pypi.org/simple/ --extra-index-url https://pypi.org/simple/ SVGraft

Project Structure

SVGraft/
├── .gitea/workflows/
│   ├── test.yml          # Unit tests on every push
│   ├── publish.yml       # Publish to PyPI on v* tags
│   └── publish-test.yml  # Manual publish to TestPyPI
├── src/svgraft/
│   ├── cli.py          # Entry point: argument parsing and split_svg()
│   ├── config.py       # Default settings and ProcessConfig
│   ├── tiling.py       # Tile grid computation
│   ├── rendering.py    # Per-tile SVG manipulation and PDF rendering
│   ├── pdf.py          # In-memory PDF merging and PDF-to-SVG conversion
│   ├── svg_io.py       # SVG unit parsing and physical dimension conversion
│   ├── geometry.py     # Unit conversion utilities
│   ├── util.py         # Small helpers
│   └── debug.py        # Debug/console helpers
├── tests/
│   ├── data/           # Sample posters (inches, millimeters, US Letter)
│   ├── test_core.py    # Unit and end-to-end tests
│   └── testSVGraft.py  # Split test with optional preview
├── gui.py              # Experimental GUI (requires dearpygui)
├── pyproject.toml
└── requirements.txt

License

This project is licensed under the GNU General Public License v3.0 or later. See LICENSE for details.


Author

Derek Baier — derek.m.baier@gmail.com

Metadata

Release files for SVGraft 0.0.7

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for SVGraft 0.0.7
File Size Uploaded Provenance
svgraft-0.0.7.tar.gz 33.7 kB Without provenance Details

Built distribution (wheel)

Table of built distributions (wheels) for SVGraft 0.0.7
File Interpreter ABI Platform
svgraft-0.0.7-py3-none-any.whl Python 3 none any Details

Total release size: 61.6 kB

Release files / svgraft-0.0.7.tar.gz

Download URL svgraft-0.0.7.tar.gz
Size 33.7 kB
Tags Source
SHA-256 checksum
How to use checksums
916eb5a3bfaf253466132691e3494a049195bfe8e282526758b48f5cff30c298
BLAKE2b-256 checksum
How to use checksums
80733f9ff15d7cf5c1c9e559d05b32d0795977ccac781687e1c55b1d48a9ed14
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/7.0.0 CPython/3.12.15

Release files / svgraft-0.0.7-py3-none-any.whl

Download URL svgraft-0.0.7-py3-none-any.whl
Size 28.0 kB
Tags Python 3
SHA-256 checksum
How to use checksums
91b62757165045847069a70d53ddc09408b140e25f7c9e438efced067738e385
BLAKE2b-256 checksum
How to use checksums
938d6513fc09ff9bccec0c7da71901fb0316996b5b94016ca24bf912de16f299
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/7.0.0 CPython/3.12.15

Security for SVGraft 0.0.7

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page