Collects data from ReDATA that can be used to create reports. Data is output in CSV format.
- Items report: Information about public and private items and their versions
- Users report: Information about user accounts
The project is structured as a Digital Ocean Function but can be executed independently as well. The code should work with any Figshare for Institutions instance.
Install
pip install -r ./packages/redata_reports/run/requirements.txt
Show the command line options by running
python ./packages/redata_reports/run/main.py -h
Prior to generating any reports, configure API endpoints and keys. Copy ./packages/redata_reports/run/secrets.example.py to ./packages/redata_reports/run/secrets.py and edit the fields with the appropriate values. See the comments in that file for specific instructions. To generate reports locally, only the Figshare API credentials are needed.
Generate a report. E.g the users report. A CSV will be output in the current working directory in this example.
python ./packages/redata_reports/run/main.py -r users -o
The --sync-to-dashboard option uploads data to the Google dashboard (data is stored in a Google Sheet and displayed in a Looker Studio dashboard). Setting this flag includes -u B -r items -r users. The destination sheet is configured in secrets.py.
Prior to using --sync-to-dashboard, the target Google Sheet must be set up. See the readme in gsheet_webapp. Once that is done, run:
python packages/redata_reports/main.py --sync-to-dashboard
A data.json file is auto-generated in the current working directory. This file can be deleted once the sync completes.
Call the webapp in the sheet, replacing XXXXX and YYYYY with the appropriate values
Simple test (no data uploaded):
curl -L "https://script.google.com/macros/s/XXXXXXXX/exec" -H "Content-Type: application/json" --data '{"action":"insertupdate","accesskey":"YYYYYY"}'
With data upload:
curl -L "https://script.google.com/macros/s/XXXXXXXX/exec" -H "Content-Type: application/json" --data-binary @data.json
An example of how the data can be presented/summarized in a dashboard using LookerStudio.
Screenshot
Authenticate to DO with doctl and connect to the desired functions namespace. The connect command will let you pick from a list if there is more than one functions namespace defined in the DO Functions console.
doctl serverless connect
Then deploy with
./deploy.sh
Once deployed, the function can be manually run which will call the equivalent of main.py --sync-to-dashboard. The URL of the function can be obtained from the Settings tab for the function in the DO console.
curl -X GET "https://<do url>/redata_reports/run?t=<token>" -H "Content-Type: application/json"
If automatic triggering is desired, the trigger should be configured manually in the DO console.