Peppermint Runner

Source of truth integrator for Peppermint.

Architecture

See Peppermint.

The runner will crawl a directory and send CATALOG.json files to Peppermint. The script has been updated to:

• ensure one process runs: the script will create a run flag file, that will delete after it executes
• remembers which entry has been processed: it will create a empty file in the tracking directory that indicates the entry had been processed. To reprocess a record, delete the corresponding tracking file.
• send records in batches. See config file.

Running

Quick set up

• Install docker-compose
• Clone Peppermint and this repo under the same directory.
• Run using Peppermint's docker-compose.yml: docker-compose up

The runner will start harvesting 'crawling' the Datacrate directory specified in config.docker.json.

Running using Docker without docker-compose

Run docker run --rm -v <path to datacrate directory>/:<path specified as 'source' in config file> -v <host config file>:/opt/peppermint-runner/config.json -e "RUN_AS_CRON=true" [ if peppermint target is a docker instance: --network=<network name>] qcifengineering/peppermint-runner:latest.

E.g. docker run --rm -v /mnt/data/publication/public/:/mnt/data/publication/public/ -v /home/vagrant/source/peppermint-runner/config.docker.json:/opt/peppermint-runner/config.json -e "RUN_AS_CRON=true" --network=peppermint_default qcifengineering/peppermint-runner:latest

Running as standalone, without Peppermint, Docker, etc.

• Perform 'Developing' quick set up below
• Run node index.js or node index.js --config <path to config>

Customising

• Edit config.json
• Edit crontab

Developing

Quick set up

• Clone branch you are developing against (e.g. master)
• Install Node version 10.15
• Install Yarn
• Run yarn install
• After making changes, run npm run-script compile if you don't have editor that supports tsconfig.json.
• Follow 'Running as standalone' above