Sebastien Tremblay-Johnston underchemist

Install GDAL

Install GDAL and its dependencies as a release version using triplet. This will install to %VCPKG_ROOT%\installed\<triplet>

vcpkg install --triplet=x64-windows-release gdal[tools,webp,zstd,sqlite3,postgresql,openjpeg,png,geos,libspatialite,curl]

Install Rasterio

Virtual Environment

Geopandas

Moved from trial to adopt
You're probably already familiar, but recent under-the-hood upgrades to IO (pyogrio over fiona) and geometry functionality (shapely 2.0), both of which are aimed at better vectorized operations.
In practice, things are more streamlined and potentially you spend less of your life trying to load a 20 GB GeoJSON into a DataFrame.

pyogrio and shapely

pyogrio is optimized for bulk reading/writing of spatial vector data

Pyogrio is fast because it uses pre-compiled bindings for GDAL/OGR to read and write the data records in bulk. This approach avoids multiple steps of converting to and from Python data types within Python, so performance becomes primarily limited by the underlying I/O speed of data source drivers in GDAL/OGR. We have seen >5-10x speedups reading files and >5-20x speedups writing files compared to using row-per-row approaches (e.g. Fiona).

	#!/usr/bin/env python

	# This tool has been copied from https://github.com/vinayak-mehta/pdftopng/blob/main/scripts/wheel_repair.py
	# and extended to supported hierarchical folder architecture with mulitple .pyd
	# to update, and to move all the DLL in a common folder package.lib installed
	# jointly with the package itself, similarly to auditwheel on Linux platform.
	#(see also https://discuss.python.org/t/delocate-auditwheel-but-for-windows/2589/9).

	# Additionally adapted from https://github.com/Wandercraft/jiminy/blob/648c0ec918ca2c2f734a32bada1dbfaf6226de48/build_tools/wheel_repair.py
	# added sorting of pyd and dll files such that repaired dlls pointing to their own

	"""
	RPC_DEM_SRS only used in GDAL >= 3.2.0, otherwise ignored
	"""

	import rasterio
	import rasterio.warp
	from rasterio import logging

	logging.basicConfig(filename='rasterio.log', level=logging.DEBUG)

	from osgeo import gdal, osr

	driver = gdal.GetDriverByName("GTiff")
	src = driver.Create('/vsimem/test.tif', 32, 32, 1)
	srs = osr.SpatialReference()
	srs.ImportFromEPSG(4326)
	src.SetGCPs((gdal.GCP(0, 0, 0, 0, 0),), srs)

	dst = gdal.Translate('/vsimem/test.vrt', src)
	print(dst.GetMetadata('xml:vrt'))

	from osgeo import gdal
	import joblib
	from itertools import cycle, product, chain
	from more_itertools import ichunked
	import time
	import os

	SRC = 'RS2_OK76385_PK678064_DK606753_F2N_20080419_142127_HH_HV_SLC/RS2_OK76385_PK678064_DK606753_F2N_20080419_142127_HH_HV_SLC'

	def block_windows(src, xsize, ysize):

	from itertools import product, chain

	def block_windows(src, xsize, ysize):
	height, width = src.shape
	nrows, rmod = divmod(height, ysize)
	ncols, cmod = divmod(width, xsize)
	offsets = product(range(0, width, xsize), range(0, height, ysize))
	sizes = product(chain([xsize]ncols, [cmod]), chain([ysize]nrows, [rmod]))
	for (xoff, yoff), (xsize, ysize) in zip(offsets, sizes):
	yield xoff, yoff, xsize, ysize

	If you are working with python module that depends on a custom sqlite3.dll, IPython on startup will load sqlite3.dll from the installation's vendored DLLs. This can cause a conflict resulting in a failed DLL load since the module that needs the custom sqlite3.dll cannot override the currently loaded DLL (DLL HELL).

	WORKAROUND: copy the custom sqlite3.dll to the installation location

	>>> import os
	>>> os.add_dll_directory('C:\path\to\dll\dir')
	>>> import rasterio

	rem https://github.com/OSGeo/PROJ/blob/475e7fd9f11af184e7e1e1618b3e1bb110a79714/src/networkfilemanager.cpp#L1613
	set PROJ_UNSAFE_SSL=YES
	projsync --list-files