svgutils

You can install svgutils from the conda-forge channel. You can either add the channel to your global configuration first, or specify it directly during the installation command.

# Option 1: Add channel to config then install
conda config --add channels conda-forge
conda install svgutils

# Option 2: Install specifying the channel directly
conda install svgutils -c conda-forge

If you are installing from the source repository, you can perform a system-wide installation (requires administrator privileges) or a local user installation.

Note: If installing locally with the --user flag, ensure you add your local site-packages directory (e.g., $HOME/python/lib/python3.x/site-packages/) to your PYTHONPATH.

# System-wide installation (requires sudo/admin)
python3 setup.py install

# Local user installation
python3 setup.py install --user

Instead of manually arranging plots in vector graphics editors like Inkscape or Adobe Illustrator, you can use svgutils to automate the composition of multiple plots into a single figure. This workflow is particularly useful when you need to frequently update your data or analysis, as you can re-run your scripts to regenerate the entire composite figure without manual repositioning or annotation.

Recommended Workflow

1. Generate Plots: Use matplotlib to create individual plots.
2. Export to SVG: Save your matplotlib figures as .svg files.
3. Compose with svgutils: Use a Python script with svgutils to import the SVG files, arrange them into a layout, and add annotations (e.g., labels like 'A', 'B').
4. Convert to Final Format: Use a tool like inkscape to convert the final composite SVG into PDF or PNG for publication.

Once you have generated your final composite figure using svgutils, you can use the Inkscape command line interface to convert the .svg file into high-quality PDF or PNG formats without opening the Inkscape GUI.

To export to PDF:

inkscape --export-pdf=fig_final.pdf fig_final.svg

To export to PNG:

inkscape --export-png=fig_final.png fig_final.svg

The svgutils.compose module provides a high-level wrapper around svgutils.transform to simplify the process of layering and arranging different SVG elements into a single figure. You can use the Figure class to combine various elements like SVG files and Image files, applying transformations such as .scale() and .move() to each element before saving the final composition.

from svgutils.compose import Figure, SVG, Image

Figure(
    "10cm", "5cm",
    SVG('svg_logo.svg').scale(0.2),
    Image(120, 120, 'lion.jpeg').move(120, 0)
).save('test.svg')

You can install svgutils from the Python Package Index (PyPI) using pip3.

Prerequisites: pip3 will attempt to install the lxml library. For a successful installation, you must have the development libraries libxml2 and libxslt1 installed on your system.

On Ubuntu and Debian-derived Linux distributions, install these dependencies using:

sudo apt-get install libxml2-dev libxslt-dev

pip3 install svgutils --user

A Panel() object allows you to group multiple elements (like a graph and its label) so they can be treated as a single unit.

Key characteristics:

• Transformations: Applying .move() or .scale() to a Panel affects all elements inside it simultaneously.
• Tiling: The Figure.tile() method treats each Panel as a single unit in the grid, rather than tiling the individual elements within the panel.
• Limitations: Unlike Figure(), a Panel() cannot have a defined size and does not have a .save() method.

Figure("16cm", "6.5cm", 
       Panel(
          Text("A", 25, 20),
          SVG("sigmoid_fit.svg")
       ),
       Panel(
          Text("B", 25, 20),
          SVG("anscombe.svg").scale(0.5)
       ).move(280, 0)
       )

The Grid(spacing_x, spacing_y) object generates a grid of horizontal and vertical lines with numerical labels in pixel units. Adding Grid() as an element in your Figure helps you visually estimate the required move() and scale() values for your elements.

Figure("16cm", "6.5cm", 
       SVG("sigmoid_fit.svg"),
       SVG("anscombe.svg").scale(0.5).move(280, 0),
       Grid(20, 20)
       )

The Figure() object is the top-most container in the compose module. To create a figure, specify its width and height (supporting units like "px" and "cm"; defaults to "px" if no unit is provided) and its contents. Use the .save("filename.svg") method to export the result.

from svgutils.compose import *

Figure("16cm", "6.5cm", 
       SVG("sigmoid_fit.svg")
       ).save("fig1.svg")

Every element in a Figure exposes transformation methods that can be chained:

• scale(factor): Scales the element by the provided factor.
• move(dx, dy): Moves the element by dx (horizontal) and dy (vertical) pixel units.

# Chaining scale and move
SVG("anscombe.svg").scale(0.5).move(280, 0)