caffeinate/ProgImage

a coding challenge, not real, just for fun.

ProgImage

a coding challenge, not real, just for fun.

It's a Flask app for RESTFul API image storage transformation. It's just a demo, the decision log details the differences between real world and demo as well as why it's structured as it is.

Overview

• No database
• Meta data is stored alongside images in an object store.
• Filesystem can be used as the object store (this is the current store implementation)
• Image transforms (make thumbnail etc.) are cached in the object store

Example Deployment on AWS

App running in gunicorn on EC2 instance within auto scaling group behind load balancer (either type). Shared filesystem for object store is EFS.

Running it locally

Setup

Virtual environment is your choice. Instructions here are with one.

python3 -m venv .venv
. .venv/bin/activate
pip3 install -r requirements.txt

Unit tests

get venv shell as above

export PYTHONPATH=`pwd`
python3 prog_image/test/test_all.py 

Running it locally

create your own config by copying the template and editting anything there that needs changing.

cd prog_image/settings/
cp local_config_template.py local_config_yourname.py
ln -s local_config_yourname.py local_config.py
cd ../..

i.e. local_config.py is used in the next bit.

export PYTHONPATH=`pwd`
python3 prog_image/app.py

You should now see a message saying the port it's running on.

Using it

For the examples below I'll assume it's running as http://127.0.0.1:5070/. And I'm using curl but you might want to use httpie or wget etc...

Send an image

curl -v -F "image=@IMG_1780.JPG" localhost:5070

You should see 'Location' in the returned http header. This is the URL for the stored image to be retrieved.

Transforms of the image can retrieved with a url of the format http://localhost:5070/images/<ident>/<transform>/.

Worked example, some lines have been removed for readability.

curl -v -F "image=@14590987878890.jpg" localhost:5070
> POST / HTTP/1.1
> Host: localhost:5070
> User-Agent: curl/7.54.0
> Accept: */*
> Content-Length: 280866
> Expect: 100-continue
> Content-Type: multipart/form-data; boundary=------------------------9050b98ac18a3548
> 
< HTTP/1.1 100 Continue
* HTTP 1.0, assume close after body
< HTTP/1.0 201 CREATED
< Content-Type: application/json
< Content-Length: 71
< Location: http://localhost:5070/images/89387845b87c8aa70ab5f433b84148c193286d22/
< Server: Werkzeug/0.14.1 Python/3.6.5
< Date: Fri, 09 Nov 2018 15:32:32 GMT
< 
{
  "id": "89387845b87c8aa70ab5f433b84148c193286d22", 
  "msg": "ok"
}

curl http://localhost:5070/images/89387845b87c8aa70ab5f433b84148c193286d22/ > image_copy.jpeg

curl http://localhost:5070/images/89387845b87c8aa70ab5f433b84148c193286d22/thumbnail/ > image_copy_thumbnail.jpeg

On demand images

The source image is POSTed to a transform. In this case a thumbnail. The response returns the thumbnail and it's being saved to file.jpeg.

curl -v -F "image=@14590987878890.jpg" localhost:5070/on_demand/thumbnail/ > file.jpeg