Extract tables from Duckdb Action

Extract some configurable geo tables into gpkg file

A GitHub Action for extracting a configurable tables from a configurable duckdb. Extract just materialize the table into a GPKG

Features

Extract tables from DuckDB Action

Extract and convert tables from DuckDB files into GeoPackage (GPKG) or Parquet outputs. This repository contains an action and a workflow to extract a table from a DuckDB file changed in a commit or when run manually.

Features

• Extract a named table from a DuckDB file using the DuckDB Python package
• Optionally convert the result to GPKG using ogr2ogr (GDAL)
• Triggered automatically on commits that modify **/*.duckdb files, or manually via workflow_dispatch

Workflow

The workflow is defined in .github/workflows/extract-duckdb.yml.

• Triggers: push on **/*.duckdb and workflow_dispatch for manual runs.
• Environment: ubuntu-latest runner (the workflow installs gdal-bin and the duckdb Python package).

Inputs (manual run)

• table_name (required): table name to extract from the DuckDB file.
• output_format (optional): GPKG (default) or PARQUET.

Behavior

• On push: the workflow locates the first .duckdb file changed in the commit and extracts the requested table.
• On manual dispatch: provide table_name (and optional output_format) in the Run workflow form.
• Outputs: the workflow writes an output file named <duckdb-base>-<table>.(gpkg|parquet) and sets output_file in the job outputs.

Usage examples

Manual run (UI): go to the Actions tab, pick "Extract DuckDB Table", click "Run workflow" and set table_name.

Programmatic example (manual workflow_dispatch in another workflow using workflow_call or via API):

# Example consumer workflow (concept)
name: Invoke Extract
on:
  workflow_dispatch:
    inputs:
      table_name:
        required: true
jobs:
  call:
    runs-on: ubuntu-latest
    steps:
      - name: Trigger extract workflow (example)
        run: echo "Use repository dispatch or call the workflow from the UI with table_name=${{ github.event.inputs.table_name }}"

Notes

• The workflow expects table_name via workflow_dispatch input or an environment variable; for fully automated push-based extraction you may add a small convention (file, metadata, or wrapper workflow) to provide the table name.
• Conversion to GPKG uses ogr2ogr; ensure your data schema is compatible with GDAL.

Development

Follow the existing development instructions in this README for running tests and linters locally.