ARC2 Server

REST server for ARC2 rainfall data

Run the server

She server may be run using the command line or docker

Command line

Run server on command line

python3 server.py

Build Docker Image

Build the docker image

docker build -t arc2_server .

NOTE regarding docker base image. numpy and other things such as pyproj come with non-trivial dependencies. alpine as a lightweight base image did not work well. The (larger) Ubuntu base image did work.

Run Server in Docker Container

Run server using docker

docker run --name arc2-server -v $PWD/data:/data/arc2 -d -p 5000:5000 arc2_server

Run a test server in parallel on port 5001

docker run -v $PWD/data:/data/arc2 -d -p 5001:5000 arc2_server

Test the Server

curl -X GET "http://localhost:5000/arc2/rainfall?lat=-0.9&long=37.7&date=20200201&days=7"

This should produce the following output

20200201 19.597993850708008
20200202 0.0
20200203 1.1990208625793457
20200204 0.31660813093185425
20200205 0.40143030881881714
20200206 1.095880150794983
20200207 0.0

To check the cache content of the server you may use

curl -X GET "http://localhost:5000/arc2/cache?date=20200205&days=4"

Which then produces something as follows

20200205 /data/arc2/africa_arc.20200205.tif.zip
20200206 /data/arc2/africa_arc.20200206.tif.zip
20200207 /data/arc2/africa_arc.20200207.tif.zip
20200208 initialized

Hint: The result will depend on which rainfall data has been accessed already

Geotiff dependency

geotiff code is by KipCrossing provided with LGPL 2.1 licence https://github.com/KipCrossing/geotiff