Databases
Start with measured support, then open the guide for the target engine. A dialect name tells Ptah which SQL rules to use; a release-line support level tells you what this repository continuously exercises. Neither is a promise that every construct renders on every target.
Common tasks
Section titled “Common tasks”Check database supportRead the generated release-line census, support policy, and evidence boundaries.
Connect to PostgreSQLUse PostgreSQL URLs, schemas, extensions, capabilities, and inspection behavior.
Connect to MySQL or MariaDBCheck URL forms, release lines, network transports, and family differences.
Use SQLite or TursoDistinguish local SQLite files from remote libSQL targets and their URL forms.
Check the live serverRun capability discovery and interpret how Ptah resolved the server profile.
Choose by engine
Section titled “Choose by engine”The sidebar includes focused guidance for SQL Server, Oracle, ClickHouse, and distributed SQL engines in addition to the common targets above. If an engine page and the generated matrix appear to disagree, the generated declaration and live capability output take precedence.
Reference
Section titled “Reference”Database URLs and dev databases separates connection scope from disposable replay targets. Dialects and capabilities explains why an accepted dialect can still refuse a concrete construct.