CSV Filter & Prune

A Streamlit web app for filtering and cleaning CSV files with both automatic filtering and manual row exclusion.

Features

• Upload CSV files with automatic encoding/quoting handling
• Filter by column with three modes:
  • Equals: Exact match
  • Contains: Substring search
  • Regex: Regular expression pattern matching
• Manual row exclusion: Tick checkboxes to exclude specific rows from the filtered results
• Download cleaned CSV: Export your cleaned data
• Live metrics: See row counts after filtering and exclusions

Installation

Option 1: Using uv (recommended)

No installation needed! Just run:

./run.sh

Or manually:

uv run --with streamlit --with pandas streamlit run app.py

Option 2: Traditional pip

pip install -r requirements.txt
streamlit run app.py

Usage

1. Run the app:

   ./run.sh

2. Open your browser (should open automatically) to http://localhost:8501

3. Upload a CSV file

4. Use the Filter controls:

   • Select a column to filter on
   • Choose a filter mode (equals/contains/regex)
   • Enter your search value or pattern

5. In the data table, tick the "Exclude?" checkbox for any rows you want to remove

6. Click "⬇️ Download cleaned CSV" to download your cleaned data

Tips

• For very large CSVs (>1-2M rows), consider using specialized tools like Modern CSV or VisiData for better performance
• The app caches loaded CSVs, so re-running filters is fast
• Regex mode supports full Python regex syntax
• The app handles malformed CSVs with fallback parsing

Example Use Cases

• Remove duplicate or invalid entries from data exports
• Filter contact lists by domain or keyword
• Clean up transaction logs by date patterns
• Manually review and exclude outliers or test data

filter-csv