Skip to content

Latest commit

 

History

History
114 lines (70 loc) · 3.75 KB

README.md

File metadata and controls

114 lines (70 loc) · 3.75 KB
Raw

slack-filter-function

This is a Serverless function for OpenFaaS that filters messages and sends them to separate slack channels based on a sentiment analysis.

Create Docker ID

If you don't have a Docker ID, you can register for it here.

Create a Docker host

You can provision an instance using play-with-docker by signing in with your Docker ID and starting a new session.

By clicking ADD NEW INSTANCE you will create a single Docker host.

Then you can install OpenFaaS on your instance with:

# docker swarm init --advertise-addr eth0 && \
  git clone https://github.com/openfaas/faas && \
  cd faas && \
  git checkout 0.7.0 && \
  ./deploy_stack.sh && \
  docker service ls

Update gateway in filter.yml with the link to OpenFaaS UI (The 8080 port link from the Docker host UI, e.g. http://ip172-18-0-27-babuu59lukr000as9tcg-8080.direct.labs.play-with-docker.com).

Deploy Sentiment Analysis function

In order to use this filter function, you will need to deploy the Sentiment Analysis function first. This is a python function that provides a rating on sentiment positive/negative (polarity -1.0-1.0) and subjectivity provided to each of the sentences sent in via the TextBlob project.

Download the function with

curl https://codeload.github.com/openfaas/faas/tar.gz/master | \
> tar -xz --strip=2 faas-master/sample-functions/SentimentAnalysis & cd SentimentAnalysis/

Instead of <URL> use the same URL you use for the filter function (The gateway value from filter.yml).

Deploy with:

curl -s <URL>/system/functions --data-binary \
'{ 
   "service": "sentimentanalysis",
   "image": "functions/sentimentanalysis",
   "envProcess": "python ./handler.py",
   "network": "func_functions"
   }'

and test

# curl <URL>/function/sentimentanalysis -d "I am really excited to participate in the OpenFaaS workshop."
Polarity: 0.375 Subjectivity: 0.75

# curl <URL>/function/sentimentanalysis -d "The hotel was clean, but the area was terrible"; echo
Polarity: -0.316666666667 Subjectivity: 0.85

Update image with your Docker ID

You need to update image value in filter.yml and replace docwareiy with your Docker ID. You can check your images in Docker Hub.

Create Slack account

If you have never used Slack before, download it from this link and create Slack account.

Then you should create two separate slack chanels for positive and negative statements.

In order to get the Slack channel tokens, you need to add the Incoming WebHooks App to each of the channels. Copy the contents of slack.examples.yml to a slack.yml file and update slack_hook_positive and slack_hook_negative with the channels tokens.

Create an Applet

Visit If This Then That and create a new Applet. Choose Tweeter and New tweet from search.

In the Search for field you should enter the query you'd like to search for.

For URL you should use the gateway value from filter.yml, followed by /function/filter.

Select POST for a Method as we would like to submit the data to our filter function. As a Content Type select application/json and a Body in json format:

{ "text": "<<<{{Text}}>>>", "username": "<<<{{UserName}}>>>", "link": "<<<{{LinkToTweet}}>>>" }

Then save your Applet and it is ready for use from the filter function.

Build and Deploy:

Use the CLI to build and deploy the function:

faas build -f filter.yml & faas push -f filter.yml & faas deploy -f filter.yml

View the logs by executing docker service logs -f filter on the Docker instance.

The filter function is ready to go. You can check you Slack channels for inputs.