Skip to content

feat(jdbc): jdbc:vortex: driver to query Vortex files from any JDBC tool #566

Closed

Description

@dfa1

Query .vortex files from any JDBC tool (IntelliJ Database, DBeaver, sqlline) without writing Java.

Shape

  • jdbc:vortex:<dir>?schema=vtx&table.<name>=<path>: a thin java.sql.Driver in vortex-jdbc, registered via META-INF/services/java.sql.Driver. It parses the URL into options and delegates to VortexCalcite.connect(opts). Unknown keys fail loudly.
  • vortex-jdbc is the user-facing artifact: embedded engine (Calcite + Babel + vortex-*), shipped as a self-contained shaded jar like duckdb_jdbc / sqlite-jdbc. Needs JDK 25+ (document it). vortex-calcite stays the plain library.

Prerequisites in vortex-calcite

  • VortexCalciteOptions (final class, private ctor, defaults() + withXxx): schema name, directory, explicit tables. Replaces connect(String, Map) / new VortexSchema(Map); Breaking.
  • Directory discovery: every *.vortex becomes a table named after the file stem. Column schema already comes from the file dtype.
  • VortexSchemaFactory for jdbc:calcite:model=..., parsing operand into the same options.
  • JAVA lex policy and the Babel parser stay hardcoded (needed for correct column-name resolution).

Shading

Merge META-INF/services (Driver entries must survive), strip *.SF/*.RSA/*.DSA, do not relocate Calcite packages (reflective class-name lookups in model JSON). Smoke test: load the shaded jar in a fresh classloader, open jdbc:vortex:<dir>, run one query.

Docs

docs/reference.md, README module list, CHANGELOG, a short "use from IntelliJ / DBeaver" section.

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    enhancementNew feature or request

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions