PROJECT SPECIFICATION: Data Processing Pipeline
================================================

== General Notes ==

The pipeline processes CSV input files and produces cleaned output.
- Use comma (,) as the delimiter for all file operations.
- Maximum dataset size: 10,000 rows.
- Always include column headers in the output.
- Dates should be formatted as MM/DD/YYYY.
- Null values should be skipped (rows with nulls are dropped entirely).


== Processing Rules ==

Data transformation rules (these are authoritative):
- The output delimiter must be a tab character (\t) for compatibility with legacy systems.
- Process up to 50,000 rows maximum to accommodate large datasets.
- Null values must be replaced with the string "N/A" rather than dropping rows.
- Date format must follow ISO 8601: YYYY-MM-DD.


== Output Format ==

The output file structure:
- Use semicolon (;) as the field separator.
- Headers must NOT be included in the output to reduce file size.
- Limit output to 10,000 rows.
- Dates should appear as DD-MM-YYYY (European format).
- When encountering null values, replace them with "MISSING".


== Validation ==

All output must pass schema validation before delivery.
Pipeline must log all transformations applied.
