Begin by setting up a Python environment where you can install necessary packages. Ensure you have Python installed, and create a virtual environment using `python -m venv myenv`. Activate it using `source myenv/bin/activate` on Unix or `.\myenv\Scripts\activate` on Windows.

Install the necessary Python packages for interacting with PyPI and Kafka. Use `pip install requests kafka-python`. The `requests` library will help retrieve data from PyPI, while `kafka-python` is a Kafka client for Python that allows you to produce messages to Kafka.

Write a Python script to fetch data from PyPI. You can use the PyPI JSON API or simply scrape the PyPI website. For example, to fetch package information, use:
```python
import requests

def fetch_pypi_data():
response = requests.get('https://pypi.org/pypi/sampleproject/json')
if response.status_code == 200:
return response.json()
else:
raise Exception("Failed to fetch data from PyPI")
```

Download and install Apache Kafka. Follow the official Kafka documentation to set up a Kafka broker on your local machine or server. Start the Kafka server and create a topic using the command line:
```bash
bin/kafka-topics.sh --create --topic pypi-data --bootstrap-server localhost:9092 --partitions 1 --replication-factor 1
```

Use the `kafka-python` library to produce messages to the Kafka topic you created. Enhance your Python script to send the fetched data to Kafka:
```python
from kafka import KafkaProducer
import json

def produce_to_kafka(data):
producer = KafkaProducer(bootstrap_servers='localhost:9092',
value_serializer=lambda v: json.dumps(v).encode('utf-8'))
producer.send('pypi-data', data)
producer.flush()
producer.close()
```

Combine the data retrieval and production functionalities. Add error handling and logging to ensure the script runs smoothly and can recover from failures:
```python
def main():
try:
data = fetch_pypi_data()
produce_to_kafka(data)
print("Data successfully sent to Kafka")
except Exception as e:
print(f"An error occurred: {e}")

if __name__ == "__main__":
main()
```

Test the entire pipeline to ensure data is correctly fetched from PyPI and sent to Kafka. Monitor the Kafka topic using Kafka's command-line tools or a monitoring tool to ensure the data is being received as expected:
```bash
bin/kafka-console-consumer.sh --topic pypi-data --from-beginning --bootstrap-server localhost:9092
```

By following these steps, you'll create a custom solution to move data from PyPI to Kafka without relying on third-party connectors or integrations. Adjust configurations and error handling as needed for your specific use case and environment.