sertit 1.7.4

SERTIT python library for generic tools

Library gathering functions for all SERTIT's projects.

Find the API documentation here.

Installing

Pip

For installing this library to your environment, please type this: pip install sertit[full]

[full] will allow you to use the whole library, but you will need to install alsorioxarray and geopandas (with GDAL installation issues on Windows, so please install them from wheels that you can find here).

However, if you do not need everything, you can type instead:

• nothing, and you won't need rasterio, rioxarray: pip install sertit --extra-index-url ...
• [rasters], and you won't need rioxarray: pip install sertit[rasters] --extra-index-url ...
• [rasters_rio]: pip install sertit[rasters_rio] --extra-index-url ...
• [colorlog]: pip install sertit[colorlog] --extra-index-url ... to have colorlog installed

Conda

You can install it via conda (but you will automatically have the full version):

conda config --env --set channel_priority strict

conda install -c conda-forge sertit

What is in it ?

Files

File gathering file-related functions:

• paths
• Create archive
• Add folder to zip file
• file extraction
• file name
• copy/remove
• find files
• JSON/pickles
• hash

Logs

• Init simple logger
• Create complex logger (file and stream + color)
• Shutdown logger

Misc

• Function on lists: convert a list to a dict, remove empty values...
• Function on dicts: nested set, check mandatory keys, find by key
• Run a command line
• Get a function name
• Test if in docker
• Change current directory (cd) as a context manager

Strings

• Conversion from string to bool, logging level, list, list of dates...
• Convert the string to be usable in command line
• Case conversion (snake_case to/from CamelCase)

Vectors

• Load an AOI as WKT
• Get UTM projection from lat/lon
• Manage bounds and polygons
• Get geopandas.Geodataframe from polygon and CRS

Rasters and rasters_rio

Basically, these functions are overloads of rasterio's functions:

• Get extent and footprint of a raster
• Read/write overload of rasterio functions
• Masking and cropping with masked array
• Collocation (superimpose)
• Sieving
• Vectorization and get nodata vector
• Merge rasters (as GTiff and VRT)
• Get the path of the BEAM-DIMAP image that can be read by rasterio
• Manage bit arrays

The main difference between the two is that rasters outputs one xarray variable when rasters_rio outputs numpy.ma.masked_arrays + dict for the raster array and its metadata.

Network

• Standard Exponential Backoff algorithm

SNAP

• Function converting bytes to SNAP understanding
• Function creating a SNAP optimized commande line

Documentation

An HTML documentation is provided to document the code. It can be found:

• online (here),
• on git, in docs. To consult it, just open the index.html file in a web browser (you need first to clone this project) To generate the HTML documentation, just type pdoc sertit -o docs\html -f --html -c sort_identifiers=False