Power Document ExtractorInstall

Summary

Extracts structured content from a binary document attached to a specified property, with detail level configurable via structuredLevel.

Inputs

• binaryPropertyName — Name of the binary property containing the file. Defaults to 'data'.
• structuredLevel — Level of detail in extracted blocks. Options: Raw, Basic, Full. Defaults to 'full'.

Output shape

Each input item produces one output item with json.blocks (extracted content blocks) and json.metadata (file metadata including format, size, MIME type, and any parser-level errors), paired with the original item index via pairedItem.

The operation parameter ('Extract XML') is used with extractFormatFromOperation to determine the file format, which routes to a dedicated parser (e.g., parseLibreOffice for Office formats, parsePdf, parseTxt, parseXlsx, parseCsv, parseMarkdown, parseEpub, parseFb2, parseOdg). For Office formats where LibreOffice conversion fails, the node falls back to text parsing and records a libreOfficeError in metadata. The structuredLevel parameter is passed to each parser to control block detail.