Skip to content

Latest commit

 

History

History
110 lines (75 loc) · 3.64 KB

README.md

File metadata and controls

110 lines (75 loc) · 3.64 KB
Raw

Python Api Demo: Restaurant Opening Hours

Purpose

An api that returns what restaurants are open per sample data of restaurants previously provided.

Design

Requirements

By Version

Initial POC, and, if time, follow-on versions

POC/MVP/V0.1
Use Case What restaurants are open at this time?
SLA/KPI N/A
Auth N/A
CI-CD N/A
Deploy Constraints Deployed locally only
Host https://localhost:{TBD}
GET Endpoint /restaurant/open/{datetime}
V0.2

Containerize the above via Docker

V1.0
Use Case What restaurants are open at this time?
SLA/KPI N/A
Auth N/A
CI-CD GitHub Actions
Deploy Constraints None
Host https://www.codeonward.com
GET Endpoint /restaurant/open/{datetime}

Technical Requirements

Endpoint

  1. The endpoint must accept a datetime string
  2. The endpoint must return a list of open restaurants by name
  3. The endpoint will return an error to the client if the restaurant name is unrecognized
  4. The endpoint will return an error to the client on all other unrecoverable errors

Assumptions and Business Rules

  1. A restaurant is considered closed if the day-of-week is not included in the data for that restaurant
  2. If the given time is at the closing time of a restaurant, it is not included
  3. When parsing the given hours of operation, any parsing error will cause the restaurant to be unrecognized
  4. Unrecognized restaurants will return an error to the client
  5. The data given in CSV format will not have "broken" ranges, e.g. "Mon-Wed 11:00 am - 1pm" with "Mon-Wed 5pm-8pm"
  6. The data will never have hours of operation that span midnight

Technical Constraints

  1. Use python and the python standard library
  2. This solution must include testing (of course! Here to be checked off)

Screenshots of Results

  • Screenshot of successful query
  • Screenshot of default response
  • Screenshot of no results
  • Screenshot of invalid datestamp

Architecture/Design/Approach

This example solution is using, loosely, the IDesign software methodology. For more information, see Righting Software.

The api router is very thin, and routes to the IDesign managers directly. Managers only communicate with engines and accessors. Managers can talk with other managers, but engines and accessors do not talk to other engines.

TODO: update the following with a diagram specific to this solution

One example of IDesign:

IDesign Example

Volatility

The main area of volatility is in the data: the hours a particular restaurant is open.

  1. If a set csv of data was a fixed technical requirement, parsing the hours-open into a list of start-end times would happen outside of the main GET request, perhaps with an internal endpoint

  2. Even better, an internal endpoint that takes the name of the restaurant with its new open hours would be saved into a database meeting/supporting our non-existent SLAs or KPIs.

Expediting for the MVP

  1. A python library for APIs will be used: Flask
  2. The given data in csv format will be parsed on startup into a json format such that each restaurant will have a list of open and closed day-of-week with time
{[
  "Caffe Luna": [
    { "Mon": { "Open": "11:00", "Closed": "13:00" }},
    { "Tue": { "Open": "11:00", "Closed": "13:00" }}
  ],
  "Neo": [
    { "Fri": { "Open": "18:00", "Closed": "24:00" }},
    { "Sat": { "Open": "14:00", "Closed": "24:00" }}
  ],
]}