A (reverse-engineered) library intended for deciphering and manipulating the .pxd files used by the image editor Pixelmator Pro.

Grab Python 3.6 or above and pip install pxdlib!

Documentation exists for the API, and a much longer set of documentation exists for the reverse-engineering of the .pxd format.

pxdlib can be used for a variety of purposes. For example, its first use case was to automagically manipulate coordinates, a little like:

from csv import reader
from pxdlib import PXDFile

x0, y0 = 50, 100

COORDS = dict()
with open('data.csv') as f:
    for name, x, y in reader(f):
        COORDS[name] = x0 + float(x), y0 + float(y)

with PXDFile('graph.pxd') as pxd:
    for l in pxd.all_layers():
        if l.name in COORDS:
            l.position = COORDS[l.name]

# The file is now saved and modified!

As pxdlib is available on PyPI, it will be updated in production only when a new version is available.

The reverse-engineering document, however, will be kept up-to-date in production as behaviour is confirmed.

The following are future goals:

• a much better file API, which doesn't just operate in-place,
• raster layers are accessible in a format which is compatible with a good raster-manipulating API;
• vector layers are fully modifiable;
• text layers can be modified, at least with the common formatting capabilities;
• all effects and filters are documented and available;
• layers can be created and destroyed;
• the above is all formally tested;
• actually being kept 100% up-to-date;
• backwards compatibility

Thankfully Pixelmator is pretty backwards-compatible, so the basic automation tools – toggling visibility, moving items around – should probably work forever. Maybe.

A changelog is available here.