A minimal command line application for managing LanceDB databases. This CLI tool provides an easy interface to query, update, and manage data stored in LanceDB.

• Query Tables: Retrieve data from LanceDB tables with filtering and column selection
• Interactive Mode: Interactive SQL shell with command history and completion
• Direct SQL Execution: Run complex SQL queries against LanceDB databases via DuckDB integration
• Data Management: Insert, update, delete, and drop records
• Schema Inspection: View the schema of tables
• Import/Export: Load data from CSV files and dump tables to CSV format
• Multiple Output Formats: Display results as formatted tables or JSON
• Auto-completion: Interactive mode includes SQL keywords and table name completion
• Safe Operations: Confirmation prompts for destructive operations

Install the package using pip:

pip install lancedb-cli

List all tables in a database:

lsql list-tables /path/to/database

View table schema:

lsql schema /path/to/database my_table

Query a table with filtering and limits:

lsql query /path/to/database my_table --limit 10
lsql query /path/to/database my_table --where "age > 25" --limit 20
lsql query /path/to/database my_table --select "name,email" --where "active=true"

Execute SQL queries directly (using DuckDB):

lsql sql /path/to/database "SELECT * FROM my_table WHERE id > 5"
lsql sql /path/to/database "SELECT name, COUNT(*) FROM my_table GROUP BY name"

Update records:

lsql update /path/to/database my_table \
  --set-clause "name='John',age=30" \
  --where "id=1"

Delete specific records:

lsql delete /path/to/database my_table --where "id=1"
lsql delete /path/to/database my_table --where "age < 18"

Empty a table (delete all rows):

lsql empty /path/to/database my_table

Drop a table permanently:

lsql drop /path/to/database my_table --confirm

Load data from a CSV file:

lsql load /path/to/database my_file.csv my_table
lsql load /path/to/database my_file.csv my_table --overwrite

Export a table to CSV:

lsql dump /path/to/database my_table output.csv
lsql dump /path/to/database employees export.csv --overwrite

Start an interactive session:

lsql interactive /path/to/database

Inside the interactive shell, you can:

• Type SQL queries directly
• Use special commands:
  • .tables - List all tables
  • .schema <table> - Show table schema
  • .refresh - Refresh all table views
  • .update <table> <set> <where> - Update rows
  • .delete <table> <where> - Delete rows
  • .empty <table> - Empty a table
  • .drop <table> - Drop a table
  • .exit - Exit the interactive shell

• --create: Create the database if it doesn't exist (available on most commands)
• --output <format>: Output format - table (default) or json (for query and sql commands)

• --limit N: Limit results to N rows (default: 10)
• --where <condition>: Filter rows with SQL WHERE clause
• --select <columns>: Select specific columns (comma-separated)

• --set-clause <values>: Column=value pairs for updates (e.g., "name='John',age=30")
• --where <condition>: SQL WHERE clause to specify which rows to modify

• --overwrite: Overwrite existing table (load) or file (dump) if it exists

• --confirm: Skip confirmation prompts for destructive operations (drop command)

To set up a development environment:

git clone https://github.com/yourusername/lancedb-cli.git
cd lancedb-cli
pip install -e ".[dev]"

Run tests:

pytest

• Python 3.8 or higher
• lancedb
• duckdb
• typer
• rich
• prompt-toolkit
• pygments

# Create a new database and load data
lsql load ./my_db employees.csv employees --create

# Query the data
lsql query ./my_db employees --limit 5

# Run SQL analytics
lsql sql ./my_db "SELECT department, AVG(salary) FROM employees GROUP BY department"

# Update specific records
lsql update ./my_db employees --set-clause "salary=50000" --where "id=10"

# Export results
lsql dump ./my_db employees employees_backup.csv

# Interactive exploration
lsql interactive ./my_db

Apache License 2.0 - see LICENSE file for details

Contributions are welcome! Please feel free to submit a Pull Request.

If you encounter any issues or have suggestions, please open an issue on GitHub.