Skip to content

Latest commit

 

History

History
104 lines (85 loc) · 3.46 KB

LinuxSourceBuild.md

File metadata and controls

104 lines (85 loc) · 3.46 KB
Raw

Linux

Here we provide guidelines on how to build GDAL 3.0+ and PROJ 4 (6.X.X) from source for Linux machines.

For Mac see the respective installation instructions provided on the ARIA-tools page.

Contents

  1. Anaconda3
  2. PROJ 4 SETUP
  3. GDAL SETUP
  4. Jupyter Notebooks SETUP
  5. Setting of environment variables
  6. Return to back to ARIA-tools page

Anaconda3

First install python3 using either Anaconda3 or Miniconda3.

Below we use a clean installation of Miniconda3. First we will download Miniconda3:

wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh

Next execute the installer script and follow the instructions as provided by the installer.

The miniconda installation does not contain all the python module we need. Use the conda excecutable to install the compiler tools that are needed for PROJ 4 installation and ARIA-tools. We can add the conda-forge channel to our installation so the packages in conda-forge are indexed.

conda config --add channels conda-forge

After adding conda-forge we can make the installation of required packages

conda install autoconf automake libtool numpy netcdf4 matplotlib pandas sqlite pkg-config shapely postgresql libcxx lapack --yes

As soon as conda is done with the installation you can run update command below and finish conda installation

conda update --all

PROJ 4 SETUP

Clone the PROJ 4 repository from github and install at least the version 6 release (i.e. main branch).

git clone https://github.com/OSGeo/proj.4 proj

Build the PROJ package.

cd proj
./autogen.sh
setenv CXXFLAGS "-DPROJ_RENAME_SYMBOLS -O2"
setenv CFLAGS "-DPROJ_RENAME_SYMBOLS -O2"
./configure --prefix=/my/proj/install/directory --disable-static
make -j4
make install

GDAL SETUP

Clone the GDAL repository from github with a version of at least 3.0 (i.e. main branch).

git clone https://github.com/OSGeo/gdal

Build the GDAL package with the python bindings:

cd gdal/gdal/
./configure --without-libtool --with-proj=/my/proj/install/directory --prefix=/my/gdal/install/directory --with-python
make -j4
make install

Jupyter Notebooks Setup

Instructions to install jupyter notebooks in a conda environment

conda install -c conda-forge jupyterlab jupyter_contrib_nbextensions rise --yes

Conda will install all required jupyter packages plus contributed notebook extensions, configurator, hide code and RISE to turn notebooks into slideshow.

Setting of environment variables

Edit your private module or start-up shell and add the PROJ and GDAL environment variables.

For example for csh do:

vi ~/.cshrc

Add the following and update my to the location where you installed the packages.

setenv LD_LIBRARY_PATH $LD_LIBRARY_PATH:/my/proj/install/directory/lib:/my/gdal/install/directory/lib
setenv PROJ_LIB /my/proj/install/directory/share/proj
setenv PYTHONPATH $PYTHONPATH:/my/gdal/install/directory/lib/python3.7/site-packages
set path = ('/my/gdal/install/directory/bin' $path)

Return to back to ARIA-tools page