Start of ETL for all events for progressive applications.
The following are steps to setup the ETL for BSD.
$ heroku create <name>
$ git push heroku master
| key | description |
|---|---|
AWS_ACCESS_KEY_ID |
Access Key for S3 bucket |
AWS_SECRET_ACCESS_KEY |
Secret AccessKey for Indivisible bucket |
AWS_HOST |
AWS Host for S3 |
S3_BUCKET |
Name of the bucket we're going to put the data in |
CLOUDFRONT_ID |
Cloudfront instance dedicated to the S3 instance |
BSD_ENDPOINT |
URL for the BSD search endpoint. Usually ends with /page/event/search_results |
REMOTE_FILENAME |
Name of the file to be taken. |
SUPERGROUP_NAME |
This is the name of the campaign. It's included in the json as supergroup |
DISABLE_COLLECTSTATIC |
Set this to 1 . This is for when we activate redis |
Most of the work are already done in the etl. The only thing you need now is to setup redis. Run these in your setup
heroku addons:create redistogo
heroku scale worker=1
heroku scale clock=1
You should be all set! You can monitor logs via:
$ heroku logs --tail
$ heroku run python test.py
This should output the following:
If the AWS and cloudfront endpoints are properly setup, the following files will be available:
- //
cloudfront_id.cloudfront.net/output/remote_filename.js.gz - //
cloudfront_id.cloudfront.net/raw/remote_filename.json
You should be able to use these for your event-map application or any event interface.