Skip to content

Latest commit

 

History

History
192 lines (122 loc) · 5.23 KB

installation.rst

File metadata and controls

192 lines (122 loc) · 5.23 KB
Raw

shell

Installation

The easiest methods for installing pyproj are:

  1. Use pip to install the binary wheels on PyPI:
pip install pyproj

Note

Linux (manylinux2014) wheels require pip 19.3+

Note

pyproj 3+ wheels do not include transformation grids. For migration assistance see: transformation_grids

  1. Use conda with the conda-forge channel:
conda config --prepend channels conda-forge
conda config --set channel_priority strict
conda create -n pyproj_env pyproj
conda activate pyproj_env

Note

"... we recommend always installing your packages inside a new environment instead of the base environment from anaconda/miniconda. Using envs make it easier to debug problems with packages and ensure the stability of your root env." -- https://conda-forge.org/docs/user/tipsandtricks.html

Warning

Avoid using pip install with a conda environment. If you encounter a python package that isn't in conda-forge, consider submitting a recipe: https://github.com/conda-forge/staged-recipes/

If these installation methods do not meet your needs, the section below provides further instructions for getting setup.

Transformation Grids

See: transformation_grids

Installing from source

Version compatibility matrix:

pyproj PROJ
<= 1.9.6 <= 5.2
2.0-2.1 6.0-7
2.2-2.3 6.1-7
2.4-2.6 6.2-7
3.0.0 7.2
3.0.1+ 7.2-9
3.3.0+ 8.0-9
3.4.0+ 8.1+

Setup PROJ

PROJ is required when building from source.

PROJ Installation Instructions <install>

You can also download PROJ from:

pyproj Build Environment Variables

PROJ_VERSION

3.0

This sets the version of PROJ when building pyproj. This enables installing pyproj when the PROJ executables are not present but the header files exist.

PROJ_DIR

This is the path to the base directory for PROJ. Examples of how to set the PROJ_DIR environment variable:

Windows:

set PROJ_DIR=C:\OSGeo4W\

Linux:

export PROJ_DIR=/usr/local

PROJ_LIBDIR

This is the path to the directory containing the PROJ libraries. If not set, it searches the lib and lib64 directories inside the PROJ directory.

PROJ_INCDIR

This is the path to the PROJ include directory. If not set, it assumes it is the includes directory inside the PROJ directory.

PROJ_WHEEL

This is a boolean value used when building a wheel. When true it includes the contents of the pyproj/proj_dir/proj/share directory if present.

PYPROJ_FULL_COVERAGE

Boolean that sets the compiler directive for cython to include the test coverage.

Setup pyproj

In the setup.py, the order for searching for PROJ is:

  1. The PROJ_DIR environment variable
  2. The internal PROJ directory (pyproj/proj_dir)
  3. The proj executable in sys.prefix
  4. The proj executable on the PATH

For best results, set the PROJ_DIR environment variable to point to location of PROJ installation before running setup.py.

If you have a previous version of PROJ installed alongside the current version of PROJ, the best way to avoid conflicts is to:

  1. Remove the previous PROJ from PATH & unset the PROJ_DATA (PROJ 9.1+) | `PROJ_LIB (PROJ<9.1) environment variables (temporarily)
  2. Install PROJ to the internal PROJ directory (pyproj/proj_dir)
  3. Set the environment variable PROJ_DIR to point to the internal PROJ directory
  4. Set the environment variable PROJ_WHEEL to true
  5. Build pyproj

Install pyproj

Note

Cython or pip>=10.0.1 is required for the installation.

Note

You may need to run pip with administrative privileges (e.g. sudo pip) or perform a user only installation (e.g. pip install --user).

From pypi:
pip install pyproj --no-binary pyproj
From GitHub with `pip`:
pip install git+https://github.com/pyproj4/pyproj.git
From cloned GitHub repo for development:
pip install -e .