A pure Python implementation of geodesy
tools for various ellipsoidal
and spherical earth models using precision trigonometric, vector-based,
exact, elliptic, iterative and approximate methods for geodetic
(lat-/longitude) and geocentric (ECEF cartesian) coordinates.
Transcoded from JavaScript originals by Chris Veness (C) 2005-2022 and from several C++ classes by Charles F.F. Karney (C) 2008-2022 and published under the same MIT License.
There are four modules for ellipsoidal earth models, ellipsoidalExact, -Karney, -Vincenty and -Nvector and two for spherical ones, sphericalTrigonometry and -Nvector. Each module provides a geodetic LatLon and a geocentric Cartesian class with methods and functions to compute distance, surface area, perimeter, initial and final bearing, intermediate and nearest points, circle intersections and secants, path intersections, 3-point resections, rhumb and rhumb lines, trilateration (by intersection, by overlap and in 3d), conversions and unrolling, among other things. For more information and further details see the documentation, the descriptions of Latitude/Longitude, Vincenty and Vector-based geodesy, the original JavaScript source or docs and Karney's Python geographiclib and C++ GeographicLib.
Also included are modules for conversions to and from Cassini-Soldner, ECEF (Earth-Centered, Earth-Fixed cartesian), UTM (Universal Transverse Mercator and Exact), UPS (Universal Polar Stereographic) and Web Mercator (Pseudo-Mercator) coordinates, MGRS (Military Grid Reference System, UTM and UPS) and OSGR (British Ordinance Survery Grid Reference) grid references, TRF (Terrestrial Reference Frames) and modules to encode and decode EPSG, Geohashes, Georefs (WGRS) and Garefs (GARS) .
Other modules provide Albers equal-area projections, equidistant and other azimuthal projections, Lambert conformal conic projections and positions, functions to clip paths or polygons of LatLon points using the Cohen-Sutherland, Forster-Hormann-Popa, Greiner-Hormann, Liang-Barsky and Sutherland-Hodgman methods, functions to simplify or linearize a path of LatLon points (or a numpy array), including implementations of the Ramer-Douglas-Peucker, Visvalingam-Whyatt and Reumann-Witkam algorithms and modified versions of the former. Other classes provide boolean operations between (composite) polygons or interpolate the Height of LatLon points and Geoid models or compute various Frechet or Hausdorff distances.
To install PyGeodesy, type python[3] -m pip install PyGeodesy
or
python[3] -m easy_install PyGeodesy
in a terminal or command window.
Alternatively, download PyGeodesy-yy.m.d.zip
from PyPI or GitHub,
unzip
the downloaded file, cd
to directory PyGeodesy-yy.m.d
and type python[3] setup.py install
.
To run all PyGeodesy tests, type python[3] setup.py test
or type
python[3] test/run.py
or type python[3] test/unitTestSuite.py
before or after installation.
Installation of Karney's Python package geographiclib is optional,
but required to use modules ellipsoidalKarney
and css
, azimuthal
classes EquidistantKarney
and GnomonicKarney
and the
HeightIDWkarney
interpolator.
Both numpy and scipy must be installed for most Geoid and Height
interpolators, except GeoidKarney
and the HeigthIDW...
ones.
Functions and LatLon
methods circin6
, circum3
, circum4_
,
soddy4
, trilaterate3d2
and trilaterate5
require numpy.
Modules ellipsoidalGeodSolve
and geodsolve
and azimuthal
classes EquidistantGeodSolve
and GnomonicGeodSolve
depend
on Karney's C++ utility GeodSolve to be executable and set with
env variable PYGEODESY_GEODSOLVE
.
To compare MGRS
results from modules mgrs
and testMgrs
with
Karney's C++ utility GeoConvert, the latter must be executable and
set with env variable PYGEODESY_GEOCONVERT
.
Module rhumbsolve
needs Karney's C++ utility RhumbSolve to
be executable and set with env variable PYGEODESY_RHUMBSOLVE
.
In addition to the pygeodesy
package, the PyGeodesy distribution
files contain the tests, the test results (on macOS only) and the
complete documentation generated by Epydoc using command line:
epydoc --html --no-private --no-source --name=PyGeodesy --url=... -v
pygeodesy
.
The tests ran with Python 3.11.1 (with geographiclib 2.0), Python 3.10.8 (with geographiclib 2.0, numpy 1.23.3, scipy 1.9.1, GeoConvert 1.51, GeodSolve 1.51 and RhumbSolve 1.51), Python 3.9.6, Python 3.8.10 (with geographiclib 1.52, GeodSolve 1.51, numpy 1.19.2 and scipy 1.5.2) and Python 2.7.18 (with geographiclib 1.50, numpy 1.16.6, scipy 1.2.2, GeoConvert 1.51, GeodSolve 1.51 and RhumbSolve 1.51), all on macOS 13.2.1 Ventura and in 64-bit only.
All tests ran with and without lazy import
for Python 3 and with command
line option -W default
and env variable PYGEODESY_WARNINGS=on
for all
Python versions. The results of those tests are included in the distribution
files.
Python 3.11.1, 3.10.8 and 3.9.6 run on Apple M1 Silicon (arm64
),
natively. Python 3.8.10 and 2.7.18 run on Intel (x86_64
) or Intel
emulation ("arm64_x86_64
", see function pygeodesy.machine).
Test coverage has been measured with coverage 4.5.4 using Python 3.10.8, 3.9.6 and 2.7.18. The complete coverage report in HTML and a PDF summary are included in the distribution files.
The tests also ran with Python 3.10.8 (and geographiclib 2.0) on Debian 11 in 64-bit only and with Python 3.9.6, 3.8.0 and 2.7.17 (all with geographiclib 1.52) on Windows Server 2012R2 in 64- and/or 32-bit.
A single-File and single-Directory application with pygeodesy
has
been bundled using PyInstaller 3.4 and 64-bit Python 3.7.4 and 3.7.3
on macOS 10.13.6 High Sierra.
Previously, the tests were run with Python 3.10.1-7, 3.9.1, 3.8.7, 3.7.1, 2.7.15, PyPy 7.3.1 (Python 3.6.9) and PyPy 7.1.1 (Python 2.7.13) (and geographiclib 1.52, numpy 1.16.3, 1.16.4, 1.16.6, 1.19.0, 1.19.4, 1.19.5 or 1.22.4 and scipy 1.2.1, 1.4.1, 1.5.2 or 1.8.1) on Ubuntu 16.04, with Python 3.10.0-1, 3.9.0-5, 3.8.0-6, 3.7.2-6, 3.7.0, 3.6.2-5, 3.5.3, 2.7.13-17, 2.7.10 and 2.6.9 (and numpy 1.19.0, 1.16.5, 1.16.2, 1.15.2, 1.14.0, 1.13.1, 1.8.0rc1 or 1.6.2 and scipy 1.5.0), PyPy 7.3.0 (Python 2.7.13 and 3.6.9), PyPy 6.0.0 (Python 2.7.13 and 3.5.3) and Intel-Python 3.5.3 (and numpy 1.11.3) on macOS 13.0.0-1 Ventura, 12.1-6 Monterey, 11.0-5.2-6.1 Big Sur (aka 10.16), 10.15.3, 10.15.5-7 Catalina, 10.14 Mojave, 10.13.6 High Sierra and 10.12 Sierra, MacOS X 10.11 El Capitan and/or MacOS X 10.10 Yosemite, with Pythonista 3.2 (with geographiclib 1.50 or 1.49 and numpy 1.8.0) on iOS 14.4.2, 11.4.1, 12.0-3 on iPad4, iPhone6, iPhone10 and/or iPhone12, with Pythonista 3.1 on iOS 10.3.3, 11.0.3, 11.1.2 and 11.3 on iPad4, all in 64-bit only and with 32-bit Python 2.7.14 on Windows 10 Pro and 32-bit Python 2.6.6 on Windows XP SP3.
All Python source code has been statically checked with PyChecker, PyFlakes, PyCodeStyle (formerly Pep8) and McCabe using Python 2.7.18 and with Flake8 using Python 3.10.8, both in 64-bit on macOS 13.2.1 Ventura.
For a summary of all Karney-based functionality in pygeodesy
, see
module karney.
Last updated: Mar 14, 2023.
Copyright (C) 2016-2023 -- mrJean1 at Gmail -- All Rights Reserved.
Permission is hereby granted, free of charge, to any person obtaining a
copy of this software and associated documentation files (the "Software"),
to deal in the Software without restriction, including without limitation
the rights to use, copy, modify, merge, publish, distribute, sublicense,
and/or sell copies of the Software, and to permit persons to whom the
Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included
in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR
OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
OTHER DEALINGS IN THE SOFTWARE.