HN Explorer

Ask HN: Simplest way to create a very basic data driven website these days?

10 points by Terry_B ↗ HN
Hey guys

I'm not up to speed on the latest cloud offerings out there.

For very simple side project ideas that are essentially static websites with a search engine component that needs to query some data, what is the quickest/simplest options out there these days to get something running in a few hours?

Here's what I was hoping to achieve:

1. Upload a CSV spreadsheet of data to some cloud storage solution

2. Be able to quickly define some simple API endpoints hosted in the cloud to query said data.

3. Write the website as static files with a js client querying the API endpoints as needed.

4. Host the website as static files in S3 or similar for low cost and ease.

What is out there that is similar to this and do people recommend as the easiest to use?

Thanks!

T

6 comments

[ 3.1 ms ] story [ 27.2 ms ] thread
noodlio
Firebase + Firebase Hosting for the backend API. Filestack to upload csv files and or to S3.
ryanfitz
I would look into AWS Lambda. Lambda host backend functions written in javascript, python or java. You can then use the AWS javascript sdk in your frontend code to invoke these lambda functions. This will be very low cost and cuts out a lot of boilerplate work. You pay on a per request basis versus an hourly or monthly basis. So if no one is using your website, then you pay nothing.
joeclark77
Python with the Flask framework is great, very easy to do the API endpoints. The whole app can fit in one .py file of about four lines plus 3 per endpoint, see the sample at http://flask.pocoo.org/.

You can host it on Heroku in their free tier of service if it's just for testing. That includes a PostgreSQL instance if you want one.

I don't know much Javascript but with the Flask framework, all HTML/JS/CSS files are static templates that sit in a folder with your app's code. You can deploy with Git or even with Dropbox if you want it on super easy mode.

tmaly
if the data is small, I would convert it to JSON first, it would be one less thing to have to do in your code.