FOR TECHNICAL USERS
Check SQL order and target state.
- Discover ordered SQL and run offline validation before reviewing a dry-run plan.
- Inspect PostgreSQL catalog drift, SQL-file checksums, and recorded runs through the CLI and operator interfaces.
DATABASE CHANGE REVIEW
Review SQL changes before you run them.
Discover ordered SQL and validate it offline. Dry-run output is shortened. Review the full source files alongside it. Catalog drift checks compare selected object definitions and table presence, but do not compare column definitions or nullability.
Read the source
Inspect SQL files and their explicit execution order.
Review the plan
Check inputs offline and examine the dry-run plan.
Keep the evidence
Inspect audit records and compare SQL-file snapshots.
Schema Migrator / change review
Synthetic data01 / Review the source
An ordered file adds a table for observations. Read the SQL before planning a change.
001_observations.sql
CREATE TABLE observations (
id bigint PRIMARY KEY,
observed_at timestamptz NOT NULL
);SQL-file snapshots preserve source files and checksums. They do not back up database data.
SQL means Structured Query Language. This demo never connects to or changes a database.
Read all steps as text
01 / Review the source
An ordered file adds a table for observations. Read the SQL before planning a change.
001_observations.sql
CREATE TABLE observations (
id bigint PRIMARY KEY,
observed_at timestamptz NOT NULL
);02 / Check the files
Sample offline validation: the file is readable and the ordered input is consistent. Offline validation does not prove that SQL will succeed against every target.
File: 001_observations.sql
Offline checks: passed (sample)
Target connection: not attempted03 / See the proposed order
A sample dry-run plan places the table before its index. Nothing in this demonstration is applied to a database.
1. tables/001_observations.sql
2. indexes/002_observed_at.sql
Mode: dry-run
Database changes: none04 / Read the evidence
A fictional completed run records the target, outcome, and ordered files. This is an example audit record, not a live execution.
Run: sample-run-004
Target: example-postgres
Outcome: completed (synthetic)
Files: 2
SQL-file snapshot: sample-snapshot-004Validation, planning, and run history
File validation and planning run offline. Catalog checks and execution need a configured target. The public demonstration executes no SQL.
- 01 / Validate
Validate
- Input
- SQL files and ordering configuration
- Processing
- Discover files and validate inputs offline
- Output
- Ordered files and validation findings
- 02 / Plan
Plan
- Input
- Validated files and target configuration
- Processing
- Generate a dry-run plan
- Output
- Proposed execution order for review
- 03 / Review the record
Review the record
- Input
- A completed run and saved SQL files
- Processing
- Retrieve execution evidence and compare checksums
- Output
- Run history and file changes
Working with Migrator
Platform and database engineers reviewing SQL execution order, target drift, and evidence of previous changes.
FOR OPERATORS AND REVIEWERS
Keep run records and target configuration.
- Use stored run and audit records to support change reviews and operational handoffs.
- Keep target configuration and encrypted credentials in one service, with connection checks and guarded execution.
Implementation details
Inspect supported catalog drift
Compare functions, procedures, triggers, views, materialized views and indexes, plus table presence. Column definitions and nullability are outside the comparison scope.
Compare SQL files
Capture SQL-file snapshots and compare their checksums. These snapshots preserve source files. They are not database backups.
Manage your targets
Keep target configuration in one place, check connections, and store target credentials encrypted. PostgreSQL is the primary target story.
Record migration runs
Discover files in order, validate offline, review dry-run plans, and inspect guarded migration runs and audit records. Review before applying to a real target.
SQL-FILE SNAPSHOTS
SQL-file snapshots preserve source files and checksums. They do not back up database data.
The sample review records are synthetic. It connects to no target and executes no SQL.
Terms
- SQL
- Structured Query Language: the language used to define and query database structures.
- Schema
- The structure of a database, including tables, columns, and related objects.
- Dry-run
- A plan of proposed work without applying the changes.
- Drift
- A difference between expected definitions and the database structure.
- SQL-file snapshot
- A saved set of SQL source files and checksums. It does not back up database data.
- Checksum
- A value calculated from a file to detect changes to its contents.
Technical guides
Worked examples for SQL migrations, wireless investigations, and hybrid search.
PostgreSQL migration dry runs
Run a two-file PostgreSQL migration example through discovery, validation, and an offline dry-run preview. Learn what the checks establish and what needs a target.
Investigating PostgreSQL schema drift
Compare expected PostgreSQL definitions, live catalog objects, and migration control state in a worked schema drift example. Separate detection from remediation.
Migration checksum mismatches
Reproduce a SQL-file checksum mismatch locally, interpret added, removed, and changed snapshot paths, and distinguish file integrity from PostgreSQL schema drift.