This project is an example of a layered set of APIs, which expose a single public API and subsequently make many calls to sub-apis to accomplish a goal. This API is then packaged up as a Docker container, which can be spooled up for handling requests.

This example is written in Python and uses Flask for HTTP operations. The underlying data-structures utilize my library QuinnStruct, to provide BinarySearchTree, HashTable and Link List (hashtable collisions).

On GET /, the API will request X random integer values via GET /generate, it will then POST /sort the list of integers to be inserted into a binary-search-tree and then perform a IN_ORDER traversal of the tree to return the sorted results. In order for the binary-search-tree to compare each nearly inserted value, it will make a GET /comparison to compare the two values.

./api/__init__.py
./tests/test_index.py

python -m pytest .\tests\test_index.py

NOTE: Requires a copy of the API running on http://127.0.0.1:5000 @TODO: Use fixture to remove this dependancy

docker-compose build

docker-compose up

Windows

cd $REPO_DIR

set FLASK_APP=api
set FLASK_ENV=development

flask run

Linux

cd $REPO_DIR
export FLASK_APP=api; flask run

cd $REPO_DIR
python ./src/index.py --port 5000

Main public facing API, GET / will return a JSON payload containing a numerically sorted list of integer values.

Performs GET /generate to obtain a random list of integers, validates the payload, then sends this JSON payload to POST /sort

http GET :5000/

{
    "count": 5,
    "values": [
        "2",
        "14",
        "61",
        "907",
        "994"
    ]
}

http GET :5000/generate

{
    "count": 5,
    "values": [
        {
            "name": "915",
            "value": 1
        },
        {
            "name": "283",
            "value": 1
        },
        ...
        {
            "name": "350",
            "value": 1
        },
        {
            "name": "69",
            "value": 1
        },
        {
            "name": "78",
            "value": 1
        }
    ]
}

cat sort.json | http POST :5000/sort

{
    "count": 5,
    "values": [
        "69",
        "78",
        "283",
        "350",
        "915"
    ]
}

sort.json

{
  "count": 5,
  "values": [
    {
      "name": "915",
      "value": 1
    },
    {
      "name": "283",
      "value": 1
    },
    {
      "name": "350",
      "value": 1
    },
    {
      "name": "69",
      "value": 1
    },
    {
      "name": "78",
      "value": 1
    }
  ]
}

http GET :5000/compare val1==1 val2==10 cs==1

{
    "result": true
}