Data

Tools

Indexes

Applications

Experiments

Awesome

An open API service indexing awesome lists of open source software.

https://github.com/davila23/javascript-valence-test

Get 5-day forecast from Metaweather API
https://github.com/davila23/javascript-valence-test

fetch javascript json requests weather-app

Last synced: about 2 months ago
JSON representation

Get 5-day forecast from Metaweather API

Awesome Lists containing this project

README 

          
# javascript-valence-test

Get 5-day forecast from Metaweather API



REF : https://drive.google.com/file/d/1rcfnwS4EebaJWQqbmDz441yZ1Po3AGiR/view



AWS Cloud Developer (Javascript) CodingAssessment

  

The goal of this assessment is for us to take a look at how you write and organize your code. 

This is the most technical part of the Valence interview process, and we want to make sure that you are not under any stress and have plenty of time to complete this assessment.

It should take you approximately 2-3 hours to complete, and we promise we'll never ask you any white- boarding pop-quizzes during the culture fit interview that follows

Guidelines

Take as much time as you need

Feel free to use an IDE of your choice

Use Google as you normally would

Use of third party frameworks is encouraged

Clearly define dependencies within package.json No need to write any unit tests

To remove any potential bias we will scrub any personally identifiable information from your submission before handing it over to an engineer for a code review.

Evaluation Criteria

We are looking at the following things:

Does the app compile and run? Are all the base features implemented?

How well is your code organized? (Hint: Break your code into meaningful functions and compose them in the endpoint hook)

Are your variable and class names descriptive? Do they make sense? (Tip: don't name things with a single character)

Are you using proper encapsulation?

  

Is your memory management sound? Are there any memory leaks?

Are you adhering to the Javascript code style guidelines and naming conventions?

Are all public methods and classes appropriately documented?

Did you implement any bonus features? Points for appropriate asynchronous operations (network/disk access), and parallelization as necessary

Less code = more . Brevity is the soul of wit

Note that we will test your function by running it from the command line and comparing to some expected output, so make sure your formatting is correct (order of entries doesn't matter).



AWS Lambda Function



We are using Serverless as the framework to enable running of this Lambda function locally without the need to deploy to AWS.

What to Do

Your mission, should you choose to accept it, is to develop an AWS Lambda function (using the serverless framework setup described below), to act as an API endpoint that reads a CSV file from S3, queries weather data from an external API for each of the locations, and returns a single JSON document describing the weather across all locations.

The API you will be querying is the MetaWeather service - it provides an aggregated forecast across many different weather services into a single 5-day forecast for each location. You will parse the CSV passed in to your API endpoint, and for each location name, you need to devise a scalable and reliable strategy to acquire the 5-day forecast data for each location name, and return it in a single, useful JSON document format intended to be consumed by a client application, or data processing service.



Your output should be a simplified version of the weather data, across multiple locations as defined in the CSV file:

      

{

"location_1_name" : [



{



"date" : "some_date", "temp" : "some_temp", "weather" : "some_weather"



}, ...



],

"location_2_name" : [



...



], ...



}



Example:



Notes

You must fetch the CSV file from S3, it can't be locally embedded into your function.

Not all data within the CSV needs to be used. You should decide which data is relevant, and which isn't. 

Make sure to read the API documentation extensively - querying the required forecast data for this assignment is not completely straightforward and may require multiple queries to get the data you need. Proper parallelization and usage of asynchronous operations is extremely important in contexts such as this - as you know, network operations are blocking, and an efficient program doesn't spendunnecessary time waiting.



Setup Installation

1. Install Serverless

2. Download Serverless Template

Using an IDE (VSCode)

You are welcome to use an IDE to develop the Lambda function. One IDE that works really well with the Serverless framework (allowing for local debugging) is VSCode. You will want to install the serverless-vscode extension if you want to use it. This should work equally well on a Mac and on Windows (most of us use Macs).

To open the template and create a debug configuration, do the following:

1. In VSCode go to File > Open and point to the template folder.

2. Open the handler.js file by double-clicking it. This step allows VSCode to infer that you'll be working

with JS/Node.

3. Click the Debug tab (leftmost menu).

4. You will see a drop-down near the Run symbol that says "No Configuration". Click on that dropdown and

select Add Configuration. Select Node.js as your environment.

5. A launch.json file will be automatically generated in the hidden .vscode folder within your

project's root.

6. Replace the contents of launch.json with the following.

        

  {

"version": "0.2.0", "configurations": [



{

"type": "node",

"request": "launch",

"name": "Launch Lambda",

"program": "${workspaceFolder}/node_modules/serverless/bin/serverless", "args": [

"invoke", "local", "-f", "myFunction"

],



// Required, otherwise Serverless thinks stdin is a stream and tries to r from it.

"console": "integratedTerminal"

ead options

} ]

}



Usage Local -->  serverless invoke local -f myFunction -l



Deploy  --> serverless deploy



Usage Remote  --> serverless invoke -f myFunction -l