use gdal>=3.3, rasterio>=1.2 for rms

umr-lops / xsar

Synthetic Aperture Radar (SAR) Level-1 GRD python mapper for efficient xarray/dask based processing

https://cyclobs.ifremer.fr/static/sarwing_datarmor/xsar/

MIT License

28 stars 8 forks source link

use gdal>=3.3, rasterio>=1.2 for rms #8

Closed oarcher closed 2 years ago

oarcher commented 3 years ago

Once required deps will be available, we will switch from 'average' to 'rms' by default. (ie change all occurrences from rasterio.enums.Resampling.average to asterio.enums.Resampling.rms

But for now, we have to use 'average' to get most of CD/CI tests to pass (see #7)

oarcher commented 3 years ago

Here is the gdal 3.3 workflow test : https://github.com/umr-lops/xsar/actions/workflows/check-gdal3.3-available.yml While it fail, we can't do anything here

oarcher commented 3 years ago

@alevieux , from https://github.com/umr-lops/xsar/issues/7#issuecomment-889893237 , about forcing rasterio>=1.2.6 :

Let's have the same installation procedure for every OS. It will fail for py3.6 on windows, but that not a big problem. Can you submit a PR for that ?

alevieux commented 3 years ago

Conda install Rasterio 1.1.0 on python 3.7 and Rasterio 1.2.6 on python 3.6, 3.8 and 3.9 and I think it's better to have Rasterio 1.2.6 ?

For windows when you force the Rasterio 1.2.6 is not exist just for python 3.6, I think just inform user the rms is not available in python 3.6 for windows For Mac the installation in python 3.7 and 3.8 must install tbb package and for windows the installation in python 3.7 and 3.8 must install Fiona package. It's specific, we can add in doc a warning for installation ?