fauxgeo 0.2.0

A python library that generates fake geospatial data

A python library that generates simple OSGeo-supported rasters and vectors. The primary purpose for this library is to help test geoprocessing functions.

• Free software: BSD license

• Documentation: https://fauxgeo.readthedocs.org.

Requirements

fauxgeo 0.2.0 requires

• NumPy

• Matplotlib

• GDAL

• affine == 1.0

Installation

pip install fauxgeo

Features

• Raster Class

• RasterFactory Class

• Vector Class

• VectorFactory Class

Usage

The Raster Class

import numpy as np
from affine import Affine
import gdal

# set arguments
array = np.ones((3, 3))
affine = Affine.identity()
proj = 4326
datatype = gdal.GDT_Float64
nodata_val = -9999.0

# initialize raster
raster = Raster.from_array(array, affine, proj, datatype, nodata_val)
raster = Raster.from_file('path/to/geotiff')

raster.uri  # equals '/path/to/geotiff'
raster.driver  # e.g. 'GTiff'
raster.dataset

raster.get_band(1)  # returns 2d numpy masked array
raster.get_bands()  # returns 3d numpy masked array
raster.get_nodata()  # returns nodata value
raster.get_shape()  # returns 2-tuple (rows, cols)
raster.get_projection()
raster.get_affine()
raster.get_bounding_box()
raster.get_cell_area()

raster2 = raster1.set_datatype(gdal.GDT_Int32)
raster2 = raster1.set_nodata(-9999)
raster2 = raster1.set_datatype_and_nodata(gdal.GDT_Int32, -9999)

# Operations
raster3 = raster1.align(raster2)
raster3 = raster1.align_to(raster2)
assert(raster3.is_aligned(raster2))

raster4 = raster3 * raster2
raster4 = raster3 / raster2
raster4 = raster3 + raster2
raster4 = raster3 - raster2
raster4 = raster3 ** raster2

raster4 = raster3 * 4.5
raster4 = 4.5 / raster3
raster4 = raster3 + 4.5
raster4 = 4.5 - raster3
raster4 = raster3 ** 4.5

raster4 = raster3.minimum(raster2)

# returns base rasters with same nodata and datatype
zeros_raster = raster3.zeros()
ones_raster = raster3.ones()

raster4 = raster3.clip('/path/to/aoi_shapefile')
raster4 = raster3.reproject(epsg_code)

reclass_table = {
    1: 2,
    2: 1
}
raster4 = raster3.reclass(reclass_table)

raster4 = raster3.resize_pixels(pixel_size, resample_method)

# visualization
image = raster4.get_grayscale_image()  # returns PIL Image object

raster.save_raster('/path/to/dst.tif')
del raster  # cleans up temporary file on object deletion or program exit

The RasterFactory Class

from affine import Affine
import gdal

# set arguments
shape = (3, 3)
affine = Affine.identity()
proj = 4326
datatype = gdal.GDT_Float64
nodata_val = -9999

# initialize factory
factory = RasterFactory(proj, datatype, nodata_val, shape[0], shape[1], affine=affine)

# create test rasters
test_raster_1 = factory.uniform(5)  # returns raster with 1 band filled with 5's
test_raster_2 = factory.alternating(0, 1)
test_raster_3 = factory.random()
test_raster_4 = factory.horizontal_ramp(1, 10)  # interpolated from 1 to 10 across columns
test_raster_5 = factory.vertical_ramp(1, 10)  # interpolated from 1 to 10 across rows

The Vector Class

from shapely.geometry import *

# set arguments
shapely_object = Polygon([(0, 0), (0, 1), (1, 1)])
proj = 4326

# initialize vector
vector = Vector.from_shapely(shapely_object, proj)
vector = Vector.from_file('/path/to/shapefile')

shapely_object = vector.get_geometry()

vector.save_vector('/path/to/dst.shp')
del vector

Tests

python setup.py test

Planning

• Add basic visualization functionality

• Raster Operations

  • Reclass

  • Overlay - intersection, union, clip

  • Dissolve

  • Buffer

  • Raster_to_Vector

  • Slope

  • Aspect

History

0.2.0 (2015-04-21)

• Local math operations added to Raster class, intergrated with Pygeoprocessing

0.1.1 (2015-03-15)

• Raster, TestRaster, and RasterFactory classes added

0.1.0 (2015-01-21)

• First release on PyPI.