DBCrust

The modern database CLI that speaks your language
PostgreSQL • MySQL • SQLite with zero compromises

[PyPI : pypi.org] [GitHub : github.com] [MIT License : opensource.org] [Rust 2024 : blog.rust-lang.org]

🚀 Why DBCrust?¶

DBCrust revolutionizes database interaction by combining the speed of Rust with the simplicity of modern CLIs. Whether you're debugging production issues, analyzing data, or automating workflows, DBCrust provides an unmatched developer experience.

Stop juggling between psql, mysql, and sqlite3. DBCrust speaks all three languages fluently.

✨ Features That Matter¶

SELECT id, na[TAB] → name, email, created_at
FROM us[TAB] → users, user_sessions, user_preferences
WHERE st[TAB] → status, state, start_date

dbc pos[TAB] → postgres://
dbc docker://my[TAB] → docker://my-postgres-container
dbc session://prod[TAB] → session://production_db

○ Execution Time: 1.23 ms • Planning Time: 0.15 ms
Index Scan
│ Optimized lookup using email_idx
│ ○ Duration: 0.96 ms • Cost: 4 • Rows: 1
│   on users using email_idx
│   filter (email = 'user@example.com')
└► id • name • email • created_at

# Automatic tunnel detection
dbcrust postgres://user@db.internal.company.com/prod
# → Automatically routes through configured jumphost

# Dynamic credentials from Vault
dbcrust vault://app-role@database/postgres-prod
# → Fetches credentials automatically

# Interactive container selection
dbcrust docker://
# → 1. postgres-dev (postgres:15)
#   2. mysql-test (mysql:8.0)
#   3. redis-cache (redis:7)

# Direct container access with autocompletion
dbcrust docker://post[TAB] → docker://postgres-dev

# Works with OrbStack DNS
dbcrust postgres://user@postgres.myproject.orb.local/db

import dbcrust

# Execute queries directly
result = dbcrust.run_command(
    "postgres://user@localhost/db", 
    "SELECT * FROM users LIMIT 10"
)

# Launch interactive CLI
dbcrust.run_cli("postgres://user@localhost/db")

from dbcrust.django import analyzer

# Detect N+1 queries and optimization opportunities
with analyzer.analyze() as analysis:
    books = Book.objects.all()
    for book in books:
        print(book.author.name)  # Will detect N+1

results = analysis.get_results()
print(results.summary)  # Performance insights

from dbcrust import PostgresClient

client = PostgresClient(host="localhost", user="postgres")
results = client.execute("SELECT COUNT(*) FROM orders")
tables = client.list_tables()

🏃‍♂️ Quick Start¶

curl -fsSL https://clement-tourriere.github.io/dbcrust/install.sh | sh

irm https://clement-tourriere.github.io/dbcrust/install.ps1 | iex

dbcrust postgres://user:pass@localhost/mydb
dbc postgres://user:pass@localhost/mydb  # Short alias

# Run immediately without installation
uvx dbcrust postgres://user:pass@localhost/mydb

# Or install globally
uv tool install dbcrust
dbcrust --help

pip install dbcrust
dbcrust postgres://user:pass@localhost/mydb

git clone https://github.com/clement-tourriere/pgcrust.git
cd pgcrust
cargo install --path .
dbcrust --help

🎯 Real-World Examples¶

Database Administration¶

-- List all databases
\l

-- Show table sizes
SELECT schemaname,
       tablename,
       pg_size_pretty(pg_total_relation_size(schemaname || '.' || tablename)) as size
FROM pg_tables
WHERE schemaname = 'public'
ORDER BY pg_total_relation_size(schemaname||'.'||tablename) DESC;

Performance Analysis¶

-- Toggle EXPLAIN mode
\e

-- Now all queries show execution plans
SELECT u.email, COUNT(o.id) as order_count
FROM users u
         LEFT JOIN orders o ON u.id = o.user_id
WHERE u.created_at > '2024-01-01'
GROUP BY u.email
HAVING COUNT(o.id) > 5;

DevOps Workflows¶

# Production database health check
dbcrust vault://readonly@prod/postgres-main \
  --query "SELECT version(), current_database(), current_user"

# Backup verification
dbcrust postgres://backup-user@replica.db.internal/app \
  --ssh-tunnel jumphost.company.com \
  --query "SELECT MAX(created_at) FROM critical_table"

🎨 What Makes DBCrust Special?¶

Modern CLI Experience¶

• Syntax highlighting for SQL and output
• History search with fuzzy matching
• External editor support for complex queries
• Session management with saved sessions and connection history
• Smart URL schemes with docker://, session://, recent://, vault://
• Shell autocompletion for connection URLs and contextual suggestions

Intelligent Features¶

• Named queries with parameter substitution
• Automatic paging for large result sets
• Copy-paste friendly output formats
• Error highlighting with helpful suggestions

Built for Speed¶

• Rust performance - start in milliseconds
• Efficient rendering - handle millions of rows
• Smart caching - autocompletion data persists
• Minimal memory footprint

🛡️ Security First¶

• ✅ TLS/SSL encryption by default
• ✅ SSH key authentication support
• ✅ Password-free workflows via Vault
• ✅ No plaintext storage of credentials
• ✅ Audit logging for compliance

🌟 Community & Support¶

Built with ❤️ using Rust, SQLx, and reedline