Rasterio

Warnings

• breaking Rasterio 1.5.0 introduced significant minimum version requirements: Python 3.12+, GDAL 3.8+, and NumPy 2+. Older Rasterio versions have different requirements (e.g., 1.4.x required Python 3.10+, GDAL 3.6+).
  Fix: Ensure your environment meets these new minimums before upgrading to 1.5.0. If unable to upgrade dependencies, use an older Rasterio version (e.g., 1.4.x) that aligns with your environment.
• gotcha Rasterio relies on an external GDAL C/C++ library. Installing GDAL (and its dependencies like PROJ and GEOS) can be complex and platform-dependent, often requiring specific system packages or conda environments.
  Fix: Consult the official Rasterio and GDAL documentation for platform-specific installation instructions (e.g., `conda install -c conda-forge rasterio`). Pre-built wheels for Rasterio often bundle GDAL, but direct system-level GDAL installation might be required for certain features or environments.
• gotcha Behavioral changes in masking and `reproject()`: Prior to 1.4.3, boundless, masked reads could erroneously mask 0-valued data. Also, `reproject()` in versions before 1.4.2 might not consistently return 2-D arrays.
  Fix: Upgrade to Rasterio 1.4.3 or newer to resolve erroneous masking and ensure consistent `reproject()` output. Always test masking and reprojection carefully when using older versions.
• gotcha Passing an open dataset object to `rasterio.open()`: Prior to Rasterio 1.4.3, this common mistake could lead to unexpected crashes. It now raises a `TypeError` for safer error handling.
  Fix: Upgrade to Rasterio 1.4.3+ to receive a `TypeError` instead of a crash. Always ensure you are passing a file path or URL string, not an already open `rasterio.DatasetReader` object, to `rasterio.open()`.

Install

• pip install rasterio Install Rasterio (may require GDAL pre-installation)

Imports

• rasterio

  import rasterio

• open

  import rasterio
  src = rasterio.open('path/to/file.tif')

• CRS

  from rasterio.crs import CRS

Quickstart

Demonstrates how to open a raster dataset, access its metadata (profile, CRS, bounds), and read a specific band into a NumPy array. A dummy GeoTIFF is created for standalone execution.

import rasterio
import numpy as np
import os

# Create a dummy GeoTIFF for demonstration purposes
# In a real application, you would open an existing file
dummy_filepath = '/tmp/example_rasterio.tif'

# Ensure the directory exists
if not os.path.exists(os.path.dirname(dummy_filepath)):
    os.makedirs(os.path.dirname(dummy_filepath))

with rasterio.open(
    dummy_filepath,
    'w',
    driver='GTiff',
    height=10,
    width=10,
    count=1,
    dtype=rasterio.uint8,
    crs='EPSG:4326',
    transform=rasterio.transform.from_origin(0, 0, 1, 1),
) as dst:
    dst.write(np.zeros((10, 10), dtype=rasterio.uint8), 1)

# --- Quickstart: Open and inspect a raster --- 
try:
    with rasterio.open(dummy_filepath) as src:
        print(f"Dataset profile: {src.profile}")
        print(f"Number of bands: {src.count}")
        print(f"Coordinate Reference System: {src.crs}")
        print(f"Bounds: {src.bounds}")

        # Read the first band as a NumPy array
        band1 = src.read(1)
        print(f"Shape of band 1: {band1.shape}")
        print(f"Data type of band 1: {band1.dtype}")

except rasterio.errors.RasterioIOError as e:
    print(f"Error opening raster: {e}. Make sure '{dummy_filepath}' exists and is a valid raster.")
except Exception as e:
    print(f"An unexpected error occurred: {e}")

# Clean up the dummy file
if os.path.exists(dummy_filepath):
    os.remove(dummy_filepath)