This package supports the look-up of FCCS fuelbed information by lat/lng or vector geo spatial data.
This software is provided for research purposes only. It's output may not accurately reflect observed data due to numerous reasons. Data are provisional; use at own risk.
This package was originally developed to support python 2.7, followed by refactoring to support various python 3.* versions. It currently supports python 3.12.
Whether cloning the repo or installing with pip, you'll first need to manually install gdal, proj, and netcdf, which fccsmap depends on.
These instructions assume you already have python and pip installed, as well as C and C++ compilers, etc.
On a mac, using Homebrew:
brew install homebrew/science/netcdf
brew install proj
brew install gdal --with-netcdf --enable-unsupported
Note that the '--with-netcdf' option is required to build gdal with the netCDF driver. See http://trac.osgeo.org/gdal/wiki/NetCDF for more information.
Note that another source for Open Source GIS packages is http://www.kyngchaos.com/software/unixport.
These instructions assume you already have C and C++ compilers, etc. installed
sudo apt-get update && sudo apt-get install -y \
python3 python3-dev python3-pip \
libnetcdf-dev libproj-dev \
libgdal-dev \
libxml2-dev libxslt1-dev
Via ssh:
git clone git@github.com:pnwairfire/fccsmap.git
or http:
git clone https://github.com/pnwairfire/fccsmap.git
After installing the non-python dependencies (mentioned above), run the following to install required python packages:
pip3 install -r requirements.txt
pip3 install -r requirements-test.txt
pip3 install -r requirements-dev.txt
If you get an error like AttributeError: 'NoneType' object has no attribute 'skip_requirements_regex, it means you need in upgrade
pip. One way to do so is with the following:
pip install --upgrade pip
If, when you use fccsmap, you get the following error:
*** Error: No module named _gdal_array
it's because your osgeo package (/path/to/site-packages/osgeo/) is
missing _gdal_array.so. This happens when gdal is built on a
machine that lacks numpy. The --no-binary :all: in the pip
install command, above, is meant to fix this issue. If it doesn't work,
try uninstalling the gdal package and then re-installing it individually
with the --no-binary option to pip:
pip uninstall -y GDAL
pip install --no-binary :all: gdal==<VERSION>
If this doesn't work, uninstall gdal, and then install it manually:
pip uninstall -y GDAL
wget https://pypi.python.org/packages/source/G/GDAL/GDAL-<VERSION>.tar.gz
tar xzf GDAL-<VERSION>.tar.gz
cd GDAL-<VERSION>
python setup.py install
Links:
- http://trac.osgeo.org/gdal/wiki/DownloadSource
- http://gis.stackexchange.com/questions/28966/python-gdal-package-missing-header-file-when-installing-via-pip
- pygdal
To import fccsmap in development, you'll have to add the repo root directory to the search path.
Use pytest:
py.test
py.test test/unit/fccsmap/test_lookup.py
You can also use the --collect-only option to see a list of all tests.
py.test --collect-only
See pytest for more information about
First install the non-python dependencies (mentioned above).
First, install pip (with sudo if necessary):
apt-get install python-pip
Then, to install, for example, version v5.0.2, use the following (with sudo if necessary):
pip install --no-binary gdal --extra-index https://pypi.airfire.org/simple fccsmap==5.0.2
See the Development > Install Dependencies > Notes section, above, for notes on resolving pip and gdal issues.
TODO: fill this in
fccsmap returns fuelbed information by geographical location or region.
To see it's options and examples, use the -h option:
$ fccsmap -h
fccscreatetiles generates tiled raster data. To see is't options and
examples, use the -h option:
$ fccscreatetiles -h