This project closes the gap between geospatial data formats and repositories and similarity measurements. Or group extended an existing Free and Open Source Software (FOSS) - Zenodo - with the functionality to retrieve and view similar records. This comprises both the API and UI, namely providing an HTTP endpoint to retrieve an ordered list of records based on a provided record and displaying similar records.
For all installationsteps we presuppose pip and python ❗
Here you can find our Zenodo-Repository:
https://github.com/corneliazy/zenodo
For Zenodo installation, please follow the instructions below:
https://github.com/corneliazy/zenodo/blob/master/INSTALL.rst
(Who wants to save time, should not try it under Windows. On Linux its much faster and easier 😉 )
After installation execute:
pip install -eThis installation was previously tested only with Linux, but should also work under Windows.
https://pypi.org/project/extractTool/
pip install extractTool(you can write this command in your console and follow the installation description on the webside OR you can download the file and install the CLI tool local: )
pip for pip install is required.
After downloading our Tool from PyPi To run our CLI tool, the following file must be executed in the project folder:
pip install -r requirements.txt --user (or sudo pip install -r requirements.txt) In this file all required plugins are listed, which we use in our tool.
Because sometimes there are some problems with the installation you must follow these next steps:
First, ensure you have gdal installed. I just run the following:
sudo apt-get install libgdal1i libgdal1-dev libgdal-devTo get the version that apt-get provided you please run:
gdal-config --versionFor example I get 2.2.3 so my pygdal version will be 2.2.3.3. (but replace the version with your version)
pip install pygdal==2.2.3.3If you get some error you may look here
Next step:
pip install pytest Then you can navigate in any common console in the folder of the tool ("extractTool") and there, the following command must be executed
python extractTool.py --path='[filepath]' --detail=[bbox|convexHull] --timefor filepath you must insert a filepath to your testdata
--bbox ← for the bounding box of the file/folder (set as default)
--convexHull ← to get all the covexHull of the file/folder
--time ← (optionally) You can add this parameter to get additionally the timeextend
python extractTool.py --path='/home/maxmusterman/test1.geojson' --detail=bbox --time
python extractTool.py --path='/home/maxmusterman/test2.nc' --detail=convexHull
python extractTool.py --path='/home/maxmusterman/testdict' --timeOur similarity calculation code can be found in the similar.pyfile.
The associated tests are in the test_similar.py file.
Our tests can be executed with the command
pytestAfter Zenodo-Installation At last switch to the distro branch with
git checkout distroIt should automatically switch to the changed zenodo version If not reboot the zenodo development installation with:
docker-compose up db es cache mqin a new terminal
workon zenodo
celery worker -A zenodo.celery -l INFO --purge --loglevel=DEBUGin a new terminal
workon zenodo
FLASK_DEBUG=True zenodo run --reload --with-threadsNow you can open http://localhost:5000/. There you can create your own profile and upload files.
If you have installed our version of zenodo you can find the integrated extractTool in your virtual enviroment in /Envs/zenodo/lib/python2.7/site-packages/extractTool
If problems or questions arise during installation, create an issue directly so we can help you and correct mistakes 😊