Robot processes speech from user.

1. User speaks to the robot.
2. Robot processes the speech.
3. Robot displays text that was spoken.

Expected Outcome

Users speech is displayed to them.

Management UI
Sub-tasks:

• Login-logout flow (#56)
• Login authentication to database
• Other things and pull requests

• CSS values across time (daily, weekly, monthly) entry vs exit
• Dropdown for specific timing per filter
• Average CSS for specified time period
• Average CSS per staff member
• Graph historic CSS values for staff member and show on graph.
• Trend of selected meals from group (nice to have)

Make the tests faster!

https://dba.stackexchange.com/questions/53198/can-postgresql-support-integration-test-with-some-kind-of-throwaway-overlay

Could add it as a decorator

Implement the database connection pool classes, as defined by the logical class diagram.

The suite of unit tests will need to check the simple and advanced constraints of our schema - to ensure ACID-compliant transactions.

Tested by #83 and #84

Sub Tasks

• Task... (PR)
• Designs
• HTML
• Add to flask <accessible from https://ict-irs.herokuapp.com/ >
• Connect with data from db

Interfaces from robot are also available on tablet.

1. Tablet is opened to show same interface as robot.

Expected Outcome

Interface should be identical and have same functionality.

Robot captures image of user

1. User stands in front of the robot.
2. Robot captures image of the user.
3. Robot displays image captured.

Endpoint in robot.py

UI Design: 9.1 of https://docs.google.com/document/d/19sFZmp3gpnUUBAwQmX9ADK9wBfQw1Miklpnxs_YiCr8/edit#heading=h.6ol07zt2rka5

Interaction: 6.1.1 of https://docs.google.com/document/d/105dhwgfuZhn7RUYRUEhM0qxlRo4OvLpqXy5B-BIrT9I/edit#heading=h.jx6ibiv1q6kz

Pages:
GET /robot Figure 9.1.1 of UI Design
GET /robot/table?people={number of people to seat} Figure 9.1.2&9.1.3 of UI Design
POST /robot/table to support GET /robot/table
GET /robot/proceed{:resid} to support successful POST /robot/table
GET /robot/full to support negative POST /robot/table

Process:

• implement all the GET endpoints (#99)
• implement the POST endpoint

Thinking about putting all the scripts in a makefile of sorts. Either an actual makefile, or to fit with the theme, a python script. This would be a central point which implements and also documents:

• Running tests e.g. make test
• Setup database e.g. make apply-schema
• Linter e.g. make lint

This would be expandable as we add more features and 'dev-ops' related features.

Running the unit testing suite with the environment variable FLASK_DEBUG=1 will cause the unit tests to fail intest_db.py, with:

AssertionError: A setup function was called after the first request was handled.  This usually indicates a bug in the application where a module was not imported and decorators or other functionality was called too late. To fix this make sure to import all your view modules, database models and everything related at a central place before the application starts serving requests.

This is a known issue as per here: ga4gh/ga4gh-server#791.

Let it be known, that people running unit tests locally should do so with export FLASK_DEBUG=0. (I don't think there is any other quick fix to this)

Tested by #81

CSS score is calculated.

1. Image is taken of the user.
2. Image is proccessed and passed through the function.
3. CSS score is assigned to the user.

Expected Outcome

CSS score is created and is between 0 and 100.

Sql stuff etc.

Image capture of user.

1. User stands in front of the tablet.
2. Image is captured.
3. Image is displayed to concierge.

Expected Outcome

Concierge is able to see the photo that has been taken.

• Take inputs: s3 filename and csss (int from 0 to 100) and store in db. (#108)
• Need to design schema for creating the filename.
• connect lambda - upload to s3 (#108)
• trigger - get css - populate entry (#107).

Lambda Functions:

• Flask App - HTTP Trigger
• Azure Face Recognition - S3 Trigger

S3:

• Static Assets Bucket
• Face Recognition Bucket

Route53:

• setup domain

RDS:

• Host the database

CloudFront:

• [] Make pretty url like this.

Tested in #84 and #85

The current approach for obtaining the APP secret is bad because it will force logging in over and over.

CSS score is stored with table data.

1. CSS is calculated.
2. CSS is stored with the table reservation in the database.

Expected Outcome

Searching for a reservation should show CSS score for table.

Robot speaks to user to prompt them for an answer.

1. User is detected by the robot.
2. Robot speaks to the user.

Expected Outcome

Robot is able to speak to the user to prompt them.

Exit kiosk utilising Chrome as the web interface

Tested by #79 and #80

Need a test fixture to inject the db connection into unit tests as they require it.

Need to update README.md to reflect changes

Table state changes as users are assigned.

1. User selectes a table
2. State of the table is changed.

Expected Outcome

Searching the table in the database should show that the table has been allocated.

Tables are shown to the user.

1. User enters the size of their party.
2. List of tables are shown to the user.
3. User selects a table from the list.
4. Table is assigned to that group.

Expected Outcome

Interface should show available tables for the user based on their party size.

Build 0 works

A small issue - not a top priority - but a menu item (in the db schema) should have a field for its price.

• Bootstrap modal for pop up where you can select new table state

• Reading from and posting (updated status) to db

• Applying css (already created) to a table if it is now unavailable

As defined by the ERD and persistent data dictionary in the report.

• Functions to list/add menu items for frontend views
• Functions to list/add staff records for authentication

• An interface for managing/interacting with the restaurant table records in the database.

Travis:

• Add deploy scripts lambda
• Add deploy scripts s3 static assets

Lambda:

• Flask App, HTTP Trigger
• Sentiment Analysis, S3 Trigger

S3:

• Static Assets
• Images pending recognition

API Gateway:

• Lambda triggers

CloudFront:

• Amalgamate S3, Lambda
• Make the url pretty

Route53:

• Pretty URL