QGIS layers from pandas DataFrame.
Script to load, group and style raster layers based on a passed DataFrame in QGIS using the load_layers() function.
This aims for a (processing) plugin in the future!
Create "pyqgis" conda environment using the provided YAML file:
mamba env create -f mamba_env.yml
conda activate pyqgis
Install pyqgis-raster-dataframe from source:
pip install -e .
When the installed conda environment is activated, you can start QGIS by running:
qgis
Now, one can load the input raster files into a pandas DataFrame (filepaths column containing the input path as
Python Path object is required!). This can be done by the filepaths2dataframe or get_yeoda_dataframe function for now.
The load_layers() function will load the raster layers to QGIS with the additional option to group the layers based on selected columns of the DataFrame.
Optionally, a column can be defined containing style details as well as a time column can be passed to connect to the QGIS temporal control
(support developed by marxT) can be used.
from yeoda_light_qgis.yeoda_light_qgis import filepaths2dataframe, load_layers
list_of_files = [...] # list of files following the Yeoda filenaming convention (available in geopathfinder)
df = filepaths2dataframe(list_of_files, filenaming_class=YeodaFilenaming)
load_layers(raster_df=df, group_columns=['var_name', 'day'])
This package was create using cookiecutter and the following skeleton: https://github.com/TUW-GEO/cookiecutter-tuw-package