Side Hustle Chronicles: How I Built and Deployed an EMA Drug Data API on RapidAPI

Luben Sirakov — Tue, 04 Mar 2025 14:06:45 +0000

Introduction

Have you ever wanted easy access to structured European Medicines Agency (EMA) drug data? 💊Well, unless you are building a project that requires constant up to date medicine information I doubt that you’d even have any reason to think about it. 😁

However, I recently built an API that automates the extraction, transformation, and deployment of EMA drug data, making it accessible to developers, researchers, and businesses.

My name is Lyuben Sirakov, a Master of Pharmacy turned Software Engineer and in this post, I'll walk you through how I built, deployed my API on RapidAPI. Let’s get started! 🚀

Why Build an EMA API?

While looking for my next side project to build, I rediscovered RapidAPI—I'd forgotten I even had an account there—and decided to explore what APIs were available. When I checked the medicines/drugs section, I was surprised to find no API for data delivered by the EMA. The inspiration came quickly: build a proper API for medicines data. 💡

While the EMA provides a data hub, its downside is that it’s in Excel format and not readily consumable by applications.

My goal was to:

Automate data retrieval from EMA’s public dataset. 🔁
Convert it into JSON for easy use in applications. 📜
Host the data in a MongoDB database for efficient querying. 📁
Expose an API via Express.js to serve developers. 🖥️
Deploy it on a cost-effective cloud provider (Railway). 🚅
List it in RapidAPI and create subscription tiers based on usage. 💲
Use the API as a side hustle to hopefully generate some revenue. 😁

Step 1: Automating EMA Data Retrieval

EMA provides drug-related data in Excel format. To keep the API up-to-date, I automated the process of:

Downloading the latest EMA dataset from their public portal.

const EXCEL_URL = 'https://www.ema.europa.eu/en/documents/report/medicines-output-medicines-report_en.xlsx';
const EXCEL_DIR = path.join(__dirname, 'ema-data-excel');
const EXCEL_PATH = path.join(EXCEL_DIR, 'medicines_output_medicines_en.xlsx');

async function downloadExcelFile() {
    try {
        if (!fs.existsSync(EXCEL_DIR)) {
            fs.mkdirSync(EXCEL_DIR, { recursive: true });
        }

        console.log('Downloading Excel file from EMA website...');
        const response = await axios({
            method: 'get',
            url: EXCEL_URL,
            responseType: 'arraybuffer'
        });

        fs.writeFileSync(EXCEL_PATH, response.data);
        console.log('Excel file downloaded successfully to:', EXCEL_PATH);
    } catch (error) {
        console.error('Error downloading Excel file:', error.message);
        throw error;
    }
}

2.Parsing the Excel file into JSON using xlsx in Node.js.

3.Cleaning and structuring the data into a consistent format.

Here’s a simplified script for parsing the Excel file:

const xlsx = require("xlsx");
const fs = require("fs");

const filePath = "medicines_output_medicines_en.xlsx";
const workbook = xlsx.readFile(filePath);
const sheet = workbook.Sheets[workbook.SheetNames[0]];
const jsonData = xlsx.utils.sheet_to_json(sheet);

fs.writeFileSync("ema_medicines.json", JSON.stringify(jsonData, null, 2));
console.log("Data extracted successfully!");

Step 2: Storing the Data in MongoDB

To efficiently serve the data, I set up a MongoDB Atlas cluster and wrote a script to upload the extracted JSON data:

const mongoose = require("mongoose");
const fs = require("fs");

mongoose.connect("mongodb+srv://mongodb_url", {
  useNewUrlParser: true,
  useUnifiedTopology: true,
});

const MedicineSchema = new mongoose.Schema({
  category: {
        type: String,
        default: '',
        required: true,
    },
    medicineName: {
        type: String,
        default: '',
        required: true,
    },
    emaProductNumber: {
        type: String,
        default: '',
        required: true,
    },
    medicineStatus: {
        type: String,
        default: '',
        required: true,
    },
    inn: {
        type: String,
        default: '',
        required: true,
    },
    //...ect.
});

const Medicine = mongoose.model("Medicine", MedicineSchema );

const jsonData = JSON.parse(fs.readFileSync("ema_medicines.json"));

Medicine.insertMany(jsonData)
  .then(() => console.log("Data uploaded to MongoDB!"))
  .catch((err) => console.error(err));

Step 3: Building an Express.js API

Here’s the simplified version of the Express.js server I built to expose endpoints for querying drug data:

const express = require("express");
const mongoose = require("mongoose");

const app = express();
const PORT = process.env.PORT || 3000;

mongoose.connect("mongodb+srv://mongodb_url", {
  useNewUrlParser: true,
  useUnifiedTopology: true,
});

const Medicine = mongoose.model("Medicine", new mongoose.Schema({ name: String, activeSubstance: String }));

app.get("/api/medicines", async (req, res) => {
  const medicines = await Medicine.find();
  res.json(medicines);
});

app.listen(PORT, () => console.log(`Server running on port ${PORT}`));

Step 4: Deploying the API on Railway

I chose Railway for hosting due to its free tier and easy integration with MongoDB. Deployment steps:

I pushed my project to GitHub.
Connected GitHub to Railway.
Added environment variables (MongoDB URL, API keys).
Deployed the app

Step 5: Publishing the API on RapidAPI

I listed the API on RapidAPI with different pricing tiers.

Basic Tier: Limited requests for testing (10K requests).
Pro Plan ($15/month): 500K requests.
Ultra Plan ($30/month): 1M requests.
Mega Plan ($135/month): 5M requests.

👉 Check out the API here: https://rapidapi.com/LubenSirakov/api/ema-medicines-data-api

Final Thoughts

In my day-to-day job as a front-end dev, I found it quite refreshing to do some back-end programming. This was a fun project that combined data automation, backend and API development. If you’re looking for an easy way to access EMA drug data, feel free to try my API!

📌 Interested? Try it here: https://rapidapi.com/LubenSirakov/api/ema-medicines-data-api

📌 Let’s connect! Lyuben Sirakov

I'd love to hear your feedback—please share any features you'd like to see, ideas you have, or constructive criticism. What would you like to see next? 🚀

#api #nodejs #expressjs #mongodb #webdev #backend #railway #rapidapi #ema #medicine

DEV Community: Luben Sirakov