Usage
Webapp
git clone http://github.com/neo4j-examples/slack-neo4j pip install -r requirements.txt export NEO4J_URL=http://localhost:7474 export SLACK_TOKEN=xop-34987294-23423-324 export TEAM_TOKEN=fVowo4UQVpajf394802fakj python app.py 8001 curl http://localhost:8001 curl -XPOST http://localhost:8001/slack
Push to Heroku
The application already contains a Procfile for Heroku to run with the port provided. You can use a Neo4j instance on GrapheneDB or GraphStory for your experiments.
git clone http://github.com/neo4j-examples/slack-neo4j heroku apps:create my-slack-neo4j git remote add heroku https://git.heroku.com/my-slack-neo4j.git heroku config:set NEO4J_URL=http://server:port heroku config:set NEO4J_URL=http://localhost:7474 heroku config:set TEAM_TOKEN=fVowo4UQVpajf394802fakj git push heroku master
Integration with Slack
Configure a /graph slash command that interacts with your server, e.g. http://my-slack-neo4j.herokuapp.com/slack
Setup a Slack-API token and note it down and configure it as SLACK_TOKEN to retrieve data from slack.
Note down the team token and configure it with your app as TEAM_TOKEN
Add infos for these commands
-
list overview
-
[import] … Import Users and Channels into Neo4j
-
[cypher] query … Run Cypher query
Implementation Notes
Howto Integrate Slash commants https://api.slack.com/slash-commands They POST a payload to our /slack endpoint like this
token=gIkuvaNzQIHg97ATvDxqgjtO team_id=T0001 team_domain=neo4j channel_id=C2147483705 channel_name=test user_id=U2147483697 user_name=Steve command=/graph text=cypher match (u:User) return u.name
Idea: dispatch on first word as command:
-
nothing → list statistics
-
cypher query→ run cypher query and return results (todo only allow read-only queries) -
import→ import users and channels idempotently