Saving gdal_rasterize output to memory in Python script

Question

I have the following Python line to execute GDAL's rasterize function on a shapefile

subprocess.call("gdal_rasterize -te {txextent} {tyextent} -tr {resolution} -burn 1 {shapefile} {output_file}.tif")

The {output_file} is the filename of the originally ingested input file.

I'd like to save the output_file to memory, so that I can call it down in another statement later on, for example, to upload to an S3 bucket.

How can I do this?

• bugmenot123: Can you do the same in actual Python code without calling a subprocess?
• fuzzy_raster9873: I haven't figured out how to. The only thing which actually works for me at the moment is subprocess

Answer 1

Here is some code which should give you an idea of how to do it. For gdal.Rasterize only the first two parameters are required. The rest of the parameters are **kwargs from RasterizeOptions . The **kwargs will be user/project specific.

import gdal
ulx, xsize, rotx, uly, roty, ysize = geotransform # from gdal's GetGeoTransform()
XSize = 100 # Raster size in pixels (columns)
YSize = 100 # Raster size in pixels (rows)
lrx = ulx + (XSize*abs(xsize)) # Calculate the lower right x coordinate
lry = uly - (YSize*abs(ysize)) # Calculate the lower right y coordinate
 
outMemRas = '/vsimem/raster_name.tif'
shpFilePath = 'shapefile_path.shp'
gdal.Rasterize(outMemRas, shpFilePath,
 outputType=gdal.GDT_Byte,
 outputSRS='EPSG:32632',
 width=XSize, height=YSize,
 outputBounds=[ulx, lry, lrx, uly],
 attribute='column_name',
 allTouched=True)
shapeRasDS = gdal.Open(outMemRas) # Open raster file in memory with gdal
shape_arr = shapeRasDS.ReadAsArray() # Read into numpy array
gdal.Unlink(outMemRas) # remove file from memory