Hi all, We've got a new version of rasterio! Rasterio 1.3.0 has new features, deprecations, and many bug fixes. It supports Python versions 3.8+ and GDAL versions 3.1+. Please see https://github.com/r

Hi all, Here is the GitHub release summary: https://github.com/rasterio/rasterio/releases/tag/1.3.0. Here is the full list of changes since 1.2.10:https://github.com/rasterio/rasterio/blob/master/CHAN

Hi all, This is a belated announcement that rasterio 1.3b3 is on PyPI. For the first time we have a small set of Windows wheels. This same set of wheels is what we intend to provide for the final 1.3.

Hi all, Rasterio 1.3b1 is on PyPI now. We're not accepting new features for 1.3.0 and are only working on major runtime and installation bugs. If you would be willing to try out this pre-release and p

Hi, Rasterio has no option to scale on opening or on reading. It's my understanding that gdal_translate's -a_scale sets a scaling metadata value but that I/O methods like https://gdal.org/api/gdaldata

Dear Rasterio Contributors and Users, Since 2013 I have been the sole arbiter of the Rasterio project. I consulted the project’s contributors on changes but had the final say over the scope, schedule,

Thanks for following up! You can also unset AWS_END_POINT since s3.amazonaws.com is the default.

Hi all, The third 1.3 pre-release is on the package index now. We have bug fixes, new features, and some major upgrades in wheel library versions. See https://github.com/rasterio/rasterio/blob/master/

Hi, Is it possible that you downloaded a source distribution for 1.2.10? Are there any .so files in it at all? I don't recommend making your application depend on the wheel library locations. These ar

Hi Ian, That would be a good approach. Rasterio's polygonizer requires the entire raster to be read into memory and the result polygons are also stored in memory as they are computed. It doesn't scale

Hi, If I read your code correctly, you are misusing the WarpedVRT constructor. You are passing a dict as the second positional argument to WarpedVRT(). This ends up being bound to the name src_crs in

Hi, The sample method is not optimized. It reads data from GDAL's block cache or disk for each coordinate. Unless your coordinates are specially sorted, it is very likely that you will get block cache

Hi Ian, GDAL doesn't support creating a GeoTIFF on S3 except via use of a temporary file. Configuration for that is explained in https://gdal.org/user/virtual_file_systems.html#vsis3-aws-s3-files. I t

Hi, reshape_as_image is intended for use with data read using rasterio. If you're not using rasterio to read your data, you probably don't need it.

Hi, I'm in favor of making filling the default for a masked array. In the same vein, I wonder if write should also call write_mask if a masked array is passed?

Hi Matt, No, we don't have a plugin system for rio-calc. If this is for your personal workflows, maybe a sitecustomize.py or usercustomize.py file that imported and patched snuggs would be worth a try

The transfer is complete.

Hi, Yes, we're following GDAL, which returns an identity matrix instead of NULL. https://gdal.org/api/gdaldataset_cpp.html#_CPPv4N11GDALDataset15GetGeoTransformEPd. GDAL's design decision has complica

Hi Shaunak, A virtual mosaic using GDAL's VRT format https://gdal.org/drivers/raster/vrt.html is the best-tested approach for doing this. Rasterio can open and operate on such VRT XML files.

Hello, I don't know about MSAVI2, but once you have a numpy array of 0 and 1, you can mask another array using https://numpy.org/doc/stable/reference/generated/numpy.where.html. Rasterio only reads an