ETL in Python: Extract CSV, Transform Dicts, Load JSON

Build a simple ETL pipeline that reads a CSV, normalizes keys and converts price to float, then writes structured JSON.

Easy Python 3.9+ Aug 9, 2026 Data pipelines & processing 12 views 0 copies

etl csv json data-pipeline dictreader

Python code

27 lines

Python 3.9+

import csv
import json
from pathlib import Path

def etl_csv_to_json(csv_path: str, json_path: str) -> None:
    """Extract CSV, transform rows to dicts, load to JSON."""
    with open(csv_path, mode='r', newline='', encoding='utf-8') as f:
        reader = csv.DictReader(f)
        records = list(reader)

    # Transform: normalize keys and convert price to float
    transformed = []
    for rec in records:
        cleaned = {k.strip().lower(): v.strip() for k, v in rec.items()}
        cleaned['price'] = float(cleaned['price'])
        transformed.append(cleaned)

    # Load: write JSON
    with open(json_path, 'w', encoding='utf-8') as f:
        json.dump(transformed, f, indent=2)

if __name__ == "__main__":
    src = Path("data.csv")
    dest = Path("output.json")
    src.write_text("Name,Price\nApple,1.25\nBanana,0.75\n")
    etl_csv_to_json(str(src), str(dest))
    print(dest.read_text())

Output

stdout

[
  {
    "name": "Apple",
    "price": 1.25
  },
  {
    "name": "Banana",
    "price": 0.75
  }
]

How it works

The csv.DictReader reads each row into a native dictionary, using the header row as keys. The transform step strips whitespace, lowercases keys, and casts the price field to float to keep data types clean. json.dump writes the list of dicts to disk with indentation for readability. Using with open handles file lifecycle safely, and __main__ guard makes this runnable as a script.

Common mistakes

Forgetting `newline=''` when opening CSV files in Python 3
Hardcoding paths instead of using `Path` or checking file existence
Assuming all CSV fields are strings without converting numeric types

Variations

Use `pandas.read_csv` then `.to_json` for larger datasets
Add filtering or deduplication logic in the transform step

Real-world use cases

Moving exported CRM reports into a JSON API-consumable format for dashboards.
Normalizing vendor CSV drops before loading them into a document database.
Converting legacy flat-file exports into structured JSON for warehouse ingestion.

ETL in Python: Extract CSV, Transform Dicts, Load JSON

Python code

Output

How it works

Common mistakes

Variations

Real-world use cases

More from Data pipelines & processing

Tutorials

Quizzes

Python code

Output

How it works

Common mistakes

Variations

Real-world use cases

More from Data pipelines & processing

Keep learning

Tutorials

Quizzes