CSV Sales Data Loader

Problem

When a script or analysis pipeline only needs raw local CSV sales records, manually parsing headers, nulls, and duplicate rows can turn simple loading into implicit cleaning. LoadData has a narrow goal: read a standard local CSV sales file and return a consistent raw list, so downstream code can filter, aggregate, or validate according to business rules.

How It Works

• Input: call main("sales.csv"); file_path accepts an absolute or relative path.
• Parsing: automatically skip the header row and parse data rows into a two-dimensional list of [date string, sales integer/None].
• Raw fidelity: preserve null values and duplicate rows without deduplication, imputation, normalization, or unit conversion.
• Output: return a raw_data array for further pandas work, statistics, or intermediate storage.

Boundaries

It fits local CSV files with a clear “date + sales” structure, not complex headers, multi-column business data, abnormal encodings, or nonstandard delimiters. If the file already contains aggregates, multiple currencies, timestamps, or free-text anomalies, define field expectations before calling. If cleaning is required, implement it separately after this skill instead of mixing loading and transformation.