Path Stage

What the stage does

• Source / output paths — Read from a context field and write results to a target path (defaults to {source}_path).
• JSONPath queries — Enter one or more paths like $.items[*].name or $[0].data. Queries run in order.
• Match modes — All matches returns an array of matches; First match returns a single value.
• Fallbacks — Optional JSON fallback used when no match is found (parsed from the input string).
• Multi-query output — If multiple JSONPaths are configured, the stage outputs an array of results (one per query). With a single query, it outputs that query's result directly.
• Error handling — Throws when the source field is missing/undefined or when a JSONPath is empty/invalid.

Configure the Path stage

1. Choose the Source field (context path of the JSON you want to query).
2. Set an Output field (defaults to {source}_path) to control where results are stored.
3. Add one or more JSONPath entries:
   • Enter the path (auto-prefixes with $ if missing).
   • Choose All matches or First match.
   • Optionally set a JSON fallback (e.g., [] or null) when no match is found.
4. Click Run Stage to preview, or Run All to execute the pipeline. Results are written to the output path and logged.

Example: extract product names

Source field: catalog

• JSONPath: $.products[*].name
• Match mode: All matches
• Output field: catalog_names

The stage writes an array of product names to context.catalog_names, ready for Filter or Visualize stages.

Tips for reliable extraction

• Validate upstream: Use Validate to ensure the source field exists and has the expected shape.
• Multiple queries: Add multiple JSONPaths when you need several slices at once; outputs will be an array of results in the same order.
• Branch outputs: Write extractions to new paths so you can inspect them without altering the source dataset.