Music Metadata Processing System with MusicBrainz

Public API

Base endpoint

http://34.123.189.132:60080

Endpoints

GET /search-track

Searches a track by track title and returns track metadata if found.

Request parameters:

Name	Type	Mandatory	Description
q	STRING	yes	search query that is tested against the track titles in the database. If a track with similar name is present in the database, returns the best match.

The only required response parameter is status, other response parameters depend on its value.

Response parameters for status code 200:

Name	Type	Description
status	INTEGER	HTTP status code (200)
artist_name	STRING	name of the artist
duration	STRING	time length of the track in the format 'mm:ss'
mbid	STRING	MusicBrainz UUID (a.k.a. mbid) of the track
release_title	STRING	name of the release that includes the track. The database only assigns one release per a track with a particular track_title
track_title	STRING	name of the track

Response parameters for status codes 4xx&5xx:

Name	Type	Description
status	INTEGER	HTTP status code (4xx/5xx)
error	STRING	explanation

---

Request example:

curl http://34.123.189.132:60080/search-track?q=radeoactive

Response example:

{
  "artist_name": "Imagine Dragons",
  "duration": "3:06",
  "mbid": "4d28f3d7-6e11-317e-9085-407c931211b9",
  "release_title": "Night Visions",
  "track_title": "Radioactive",
  "status": 200
}

Request example:

curl http://34.123.189.132:60080/search-track?q=radeeaeaoactive

Response example:

{
  "error": "No match found",
  "status": 404
}

Build from source and run

1. create set-env.sh file with env variables

Use set-env.sh.example as an example

(optionally) 2. test

The tests do not rely on the k8s environment. You can use your python installed on your system.

For the first run, I recommend creating a python venv and install the dependencies for both python pods:

python -m venv venv; # python 3.8 and 3.12 work
. venv/bin/activate

pip install -r populate_task/requirements.txt
pip install -r rest_api_service/requirements.txt

Then, run tests:

. venv/bin/activate
. set-env.sh

 #1 
python populate_task/test.py
 #2 
python rest_api_service/test.py 

3. build and run for your k8s environment

❕ Requires installed docker and kubectl

A) Locally (minikube)

❕ Requires installed minikube

minikube requires non-root docker access. You may need to execute sudo usermod -aG docker $USER && newgrp docker to add your user to docker group [R]

Run:

bash build.sh

if pods started successfully and services are running, check ip address of the service by running minikube ip. The address of the REST API service will be the ip address from its output with the 30080 port.

B) on Google Kubernetes Engine (GKE)

❕ Requires installed gcloud and logged in to gcloud [R], and a k8s cluster created in Google Kubernetes Engine console with corresponding credentials specified in set-env.sh.

Run:

bash build-gpc-gke.sh

if pods started successfully and services are running, check the external ip address of the REST API service by running kubectl get services. GCP GKE will automatically assign a EXTERNAL-IP field to rest-api-service shortly after pods start.

Use that EXTERNAL-IP with the port 60080 as an endpoint base for the REST API service. (In my case, EXTERNAL-IP is 34.123.189.132)