Deploying Updates to BostonInfo

The Preferred Way

If you followed the Installation wiki page, you are have set up to use the ASK CLI tools for deployment. Just run ask deploy in the project repo.

The Old Way

Before updating to ASK CLI v2, we developed our own scripts for deployment. We keep these around since the automatic deployment systems developed with the City of Boston still rely on them.

deploy_tools.py does all the work for us as long as we have a few things taken care of first:

• Install and configure the two command line tools it uses:
  • AWS CLI (installation instructions)
  • ASK CLI (installation instructions)
• Set up an environment variable called BOSTON_INFO_SKILL_ID, and assign to it the ID of our skill
  • To see the skill ID, click View Skill ID under the skill's name in the Alexa developer console

Once the above is complete, you should be able to upload your current project code to Lambda and upload/build the interaction model with the following command:

python deploy_tools.py -f [your Lambda function name] -i [your Alexa skill ID]

You can find your Lambda function name in the Function Name column on the Lambda main page.

This will:

• Create the zip file with all necessary project code and dependencies and upload it to Lambda.
• Upload the interaction model and build it.

Note: You must either provide your skill ID after the -i flag like -i amzn1.ask.skill.1234abc-fake-skill-hash or define an environment variable called BOSTON_INFO_SKILL_ID whose value is the skill ID. If you pass the -i flag without a skill ID and the environment variable does not exist, the upload/build of your interaction model will fail.

If you just want to upload/build the interaction model, use:

python deploy_tools.py -i

If you just want to create the .zip file without uploading it, use:

python deploy_tools.py -p

(the .zip file will be saved in the base directory of the project)

For help:

python deploy_tools.py -h

Back   |   Next