abeldan / qpy3

A simpler way to use the PyQGIS libraries with standalone scripting

qpy3

Standalone open source python module for spatial analysis and GIS processing.

What is it?

I've found that standalone GIS processing to be quite a chore in the open source world, coming from being a heavy ESRI user in university, and wanting to transition away from closed source GIS processing to a fully open stack. This is intended to be a drop-in replacement for ESRI's closed source, Windows only arcpy python library. The goal of this project is to reach as close to feature parity as possible, enabling a user to replace import arcpy with from qpy3 import Qpy as arcpy at the top of their script, and have identical outputs to their ArcPy script. This module works with Python 3 and QGIS 3.10. QGIS 3.4 works, but spews a lot of deprecation warnings.

Setup

Ubuntu

On Ubuntu, simply run setup.sh. It will install the necessary dependencies. xvfb and xvfbwrapper create a dummy graphical interface for running on a headless environment.

Windows

Currently, this library does not support "true" standalone processing in Windows, but rather must be used within the OSGeo4W Python shell. You can invoke your qpy script by calling python-qgis my_qpy_script.py from within the OSGeo4W shell. We simply do not have the resources to make true standalone qpy happen on Windows, but we will gladly accept a pull request that enables this!

MacOS

Nobody on the development team currently owns a mac or hackintosh, so we are unable to develop for this operating system. We assume that it would only be a minor change to how it works on Linux. A pull request enabling this feature would be readily accepted.

Using the library

Example: Buffer your data

from qpy3 import Qpy 

distance = 2 # decimal degrees in a geographic CRS, meters in a projected CRS

in_shp = "/path/to/shapefile.shp"

out_shp = "/path/to/output/shapefile.shp"

Qpy.Buffer(in_shp, out_shp, distance)

# Close the dummy QGIS instance
Qpy.close()

Testing

To ensure that it works, run the test.py script available in the tests subfolder. It will run some basic geoprocessing tools on the sample data provided in the repo.

Contributing

Want to contribute? Simply create a pull request to add new functionality.