alexa-logger

Utility functionality that manages request/response for Alexa utterances. This module generates a report that lets you see the interaction of your sessions grouped by user.

Usage

The exposed functions from this library are saveLog which your Alexa skill should call to save a request and response in a supplied AWS S3 bucket and processLog which you can call offline to process a directory of log files for easy consumption.

saveLog(event, response, options, callback)

The arguments to this function are:

• event - The event as passed into your skill's exported function
• response - The response from your skill
• options - an options structure defined below
• callback - an optional callback function

The options structure is composed of the following fields with the following default values:

{
  bucket,     // Required - the name of the S3 bucket to write to
  region,     // AWS region hosting the S3 bucket; default is 'us-east-1'
  keyPrefix,  // Prefix for the key that will be generated in this bucket
  fullLog,    // If true, the full request is logged - by default
              // only elements reported on in processLogs are retained
}

For example, if you want to write to a 'logs' directory off a 'mydata' S3 bucket, you could call this function as follows:

saveLog(event, 'Welcome to my skill', {bucket: 'mydata', keyPrefix: 'logs/'});

processLogs(options, resultFile, callback)

This function will process the logs from a directory and organize them by user and by session for easy processing. The content will be saved into a comma-separated file that you can easily view to get a sense of the intents and slot data being passed into your skill, as well as the responses that you are providing to your end customers.

The arguments to this function are:

• options - an options structure defined below
• resultFile - the file to write the final results to
• callback - an optional callback function

The options structure is composed of the following fields:

{
  directory,   // If provided, logs will be read from local filesystem
  s3: {        // If provided, this structure contains details about
               // the S3 bucket containing the logs to process
    bucket,    // Required - the name of the S3 bucket
    region,    // AWS region hosting the S3 bucket; default is 'us-east-1'
    keyPrefix, // The prefix for the keys to read
  },
  daterange: { // v1.1 or higher - date range of logs to retrieve
    start,     // Starting date as a primative value - logs that have
               // a timestamp strictly greater than this will be returned
               // If absent, all logs prior to end date will be returned
    end,       // Ending date as a primative value - logs that have
               // a timestamp strictly less than this will be returned
               // If absent, all logs after start date will be returned
  },
}

The callback function, if specified will be called with the following parameters:

• error - an error string, if any
• info - v1.1 or higher, a structure with additional information, provided when there is no error

The info structure is composed of the following fields:

{
  last,   // The last timestamp for a logfile. Passing this
          // in as daterange.start will cause only new logs to
          // be reviewed on a subsequent call
}

I hope you find this utility module useful; contributions are welcome!