PG-dbHydrate

01Select
Choose the objects to review.
02Inspect
Read the generated SQL side by side.
03Validate
Check warnings and rehearse the plan.

THE CHANGE, IN CONTEXT

See exactly what will change.

Keep the source, target and decisions together. Select the differences you want to include before applying them.

Target · currentUAT

CREATE TABLE public.customers (
  id bigint PRIMARY KEY,
  first_name varchar(80),
  email varchar(255)
);

Source · desiredDEV

CREATE TABLE public.customers (
  id bigint PRIMARY KEY,
  first_name varchar(80),
  email varchar(255),
  phone varchar(32)
);

+ One reviewed column

Selected objectsFocus on the tables, indexes and views that matter.
Generated SQLInspect current and desired definitions.
Reviewed warningsCheck impact, dependencies and the target.

Illustrative definitions. Review the actual plan for your database.

PostgreSQL transaction phases

Transactional DDL is grouped into phases. Operations such as CREATE INDEX CONCURRENTLY require separate execution and review.

1
CompareIdentify differences.
2
Dry runRehearse the plan.
3
ConfirmReview and approve.
4
ApplyExecute on the target.

Refresh selected data

Choose tables and rows, include the relations you need, and inspect masking and overrides.

Explore the refresh workflow

Manage users & access

Create users, update accounts, and add or remove database access. Review engine-specific roles and permissions before applying.

View the access workspace

Engine details & considerations

BUILT FOR POSTGRESQL ARCHITECTURE

Confidence in every DDL change.

From ad-hoc hotfixes to coordinated environment migrations, dbhydrate inspects live PostgreSQL catalogs without guessing.

Transactional DDL Rollback

PostgreSQL supports transactional schema updates. dbhydrate groups statements into atomic phases. If a constraint or type alteration fails, the entire transaction rolls back automatically, undoing the statements in that phase. Non-transactional operations need separate review.

BEGIN;
ALTER TABLE app.orders ADD COLUMN status varchar(32) NOT NULL DEFAULT 'pending';
CREATE INDEX idx_orders_status ON app.orders (status);
COMMIT; -- Rolls back entirely if any statement fails

Selective Object Sync

Cherry-pick exact tables, columns, indexes, enum types, sequences, views, triggers, and functions. dbhydrate automatically determines creation and alteration order based on foreign key and view dependencies.

-- Objects staged with dependency graph
1. custom enum: order_status_type
2. parent table: app.customers
3. child table: app.orders (FK → customers)

Referential Data Subsetting

Refresh development or staging databases with realistic production slices. Apply a WHERE filter (e.g., specific tenant or date range) and dbhydrate copies referenced parent and dependent child rows in valid foreign-key sequence.

SELECT * FROM app.orders WHERE created_at >= '2026-01-01';
→ Choose to include related customers and order_items

In-Memory PII & PHI Masking

Protect customer identities when refreshing lower environments. Apply deterministic masking to email addresses, names, and phone numbers in workstation memory before writing to target tables, preserving relationship joins across tables.

customer.email: [email protected]
masked value: [email protected] (Deterministic seed)

LIVE SCHEMA DIFF & DDL BUILDER

Interactive Schema Diff Playground

Simulate schema divergences in real time. Add or modify columns, switch database engines, and inspect the verified, dependency-ordered DDL synchronization script.

app.orders

Column	Data Type	Status
`id`PK	`bigint`
`customer_id`FK	`bigint`
`created_at`	`timestamp with time zone`
`status`	`varchar(32)`
`metadata`	`jsonb`

Add Schema Column

Generate matching secondary index for added columns

Generated Synchronization SQL2 Added · 0 Modified

-- dbhydrate generated PostgreSQL synchronization plan
-- Protection tier: Safe Transactional DDL
BEGIN;

ALTER TABLE app.orders
  ADD COLUMN status varchar(32) DEFAULT 'pending' NOT NULL;
ALTER TABLE app.orders
  ADD COLUMN metadata jsonb;

-- Create index on added column
CREATE INDEX idx_app_orders_status
  ON app.orders (status);

COMMIT;
-- End of PostgreSQL transactional execution phase

POSTGRESQL DETAILS

Frequently asked questions.

Which PostgreSQL versions does dbhydrate support?

dbhydrate supports PostgreSQL versions 13, 14, 15, 16, and 17. This includes self-hosted PostgreSQL installations on Linux, Docker, or bare metal, as well as managed cloud instances including AWS RDS, Amazon Aurora PostgreSQL, Supabase, Neon, Google Cloud SQL, and Azure Database for PostgreSQL.

How does transactional DDL rollback protect PostgreSQL deployments?

PostgreSQL supports transactional Data Definition Language (DDL). dbhydrate organizes migration operations into transactional phases wrapped in BEGIN and COMMIT blocks. If any statement or constraint fails during execution, PostgreSQL automatically rolls back the entire phase to prevent partial schema drift. Operations requiring non-transactional execution (such as CREATE INDEX CONCURRENTLY) are staged and reviewed in dedicated execution phases.

Can dbhydrate copy selected rows while maintaining foreign key integrity?

Yes. dbhydrate analyzes foreign key relationships between tables. When you copy rows filtered by a WHERE clause (e.g., WHERE created_at >= NOW() - INTERVAL '30 days'), it lets you include referenced parent rows and dependent child rows in correct dependency order, avoiding foreign key constraint violations.

How does in-memory PII masking work for PostgreSQL data transfers?

When copying sensitive data from a production or pre-production database to lower environments like Dev or Sandbox, dbhydrate transforms values in local workstation memory before sending them over the wire. Masking rules support deterministic email and name generation, salted hashes, and custom regex replacements without transmitting data to any cloud AI service.

Bring clarity to your next
PostgreSQL change.

14-day trial · Starts when you register in the desktop app. Each purchased seat allows 1 device(s).

Try PostgreSQL View pricing