Python Data Cleaning

I’m sitting on a growing mix of CSV tables, SQL exports, free-form text files, and JSON logs, and the mess is slowing down the rest of our analytics work. I need a concise Python workflow that reads these varied sources, applies consistent cleaning rules, and outputs tidy, analysis-ready data.

Core scope
• Ingest both structured (CSV, SQL) and unstructured (text, JSON) inputs.
• Detect and repair common issues—missing values, inconsistent encodings, rogue delimiters, duplicate records, noisy text.
• Normalise field names and data types so downstream BI tools see a single schema.
• Deliver the cleaned data back as neatly partitioned CSV or Parquet plus a reusable .py script (pandas, numpy, regex, or other sensible libraries).

Acceptance criteria
– Running one command processes all sample files I provide and produces a consolidated “clean” dataset with zero schema drift.
– The script is clearly commented so my team can tweak rules later.
– A brief README explains setup and usage (virtualenv, pip install, run example).

That’s it—straightforward, efficient, and built in idiomatic Python so we can drop it into our existing automation stack immediately.