The Clean Air Project is a collaboration between scientists and software engineers to create a website which will allow users to upload, access, process and download air quality data. We will be using cutting edge technology and software to engineer a fully-functional, easy-to-use one-stop-shop for air quality data, including resources for researchers and decision-makers such as analysis pipelines and health impacts.
Software dependencies are intended to be installed using conda:
conda env create -f environment.ymlThen, install the clean_air package to this environment using pip:
conda activate cap_env
pip install .Remember to use the -e option to pip install for development work.
We've decided to temporarily reuse the metadata implementation from edr_server as a time-saving measure because our metadata storage requirements are currently being driven by the EDR standard.
For now, it's difficult to justify separately maintaining two different copies of the same functionality.
We'll keep this under review to identify if/when the clean air project's metadata requirements diverge from
edr_server
Some code accesses resources held in an AWS S3 compatible object store. Credentials must be provided by (TODO... we use boto3, so probably need the credentials stored in a way that boto3 will access)
Fully setting up the environment requires some steps beyond installing the code:
Tests are run using pytest.
They also require the sample data located in the cap-sample-data repository.
This repo should be checked out to the same directory as this repo, so that the 2 repositories are peers.
The custom --sampledir option to the pytest command can also be used to change the default location the tests look
for the same data. E.g. pytest --sampledir /path/to/sample/data/repo/root