Data for this project was provided in CSV format. This repo includes a preprocessing script to convert the CSV files into JSON files. This is necessary to:
- Build the many-to-one relationships of CSV rows into a JSON tree structure
- Compute means when an AOI has multiple RCAs, each with their own set of values
- Omit blank and infinite values
- Optimize data by removing unused fields and rounding to sensible digits
To generate the required JSON files from the CSV data files, run:
cd scripts
pipenv install
pipenv run ./csv2json.py
The JSON files will appear in the assets/data subdirectory of this repo, where they will be imported by the Pinia store.
An assets/hash_map.json file will be generated at the same time. This contains the mappings of 6-digit hashes used in permalinks with their corresponding AOI names.
The map component of this webapp uses a boundary GeoJSON file to determine the clickable area and adjust the mouse cursor accordingly. A shadow mask of the boundary is also loaded as a layer into the map to make the unclickable areas darker.
cd scripts
pipenv install
pipenv run ./boundary.py
The boundary GeoJSON file will appear in assets/boundary.json.
Clone the geospatial-vector-veracity repo, then run:
cd geospatial-vector-veracity/utilities
pipenv install
pipenv run python symmetric_difference.py ../serdp-fish-and-fire/scripts/input/AOI_v2_1.shp \
--bounds -180 -90 180 90 'SERDP Fish and Fire AOI shadow mask' 'AOI_v2_1_shadowmask' \
AOI_v2_1_shadowmask.shp preview.png
The AOI_v2_1_shadowmask shapefile then needs to be uploaded to our GeoServer and added as a store and layer.
nvm use lts/hydrogen
npm install
npm run dev
export NUXT_APP_BASE_URL=/tools/fish-and-fire
nvm use lts/hydrogen
npm run build
aws s3 cp .output/public s3://fish-and-fire/ --acl public-read --recursive