Parse Document
AI-generatedSummary
Parses a document supplied as workflow binary data or fetched from a URL, with configurable output format, OCR, page selection, password, and optional image extraction.
Inputs
- Input Type (required) — Choose Binary Data (default) to read the configured binary property, or URL to fetch the document from the configured URL.
- Input Binary Field (required) — Binary property name to read when Input Type is Binary Data; defaults to
data. - URL (required) — Document URL to fetch when Input Type is URL.
- Skip SSL Verification — When enabled for URL input, skips SSL certificate verification; defaults to false.
- Output Format — Select Markdown (default), JSON, or plain Text output.
- OCR Language — Language for OCR recognition; defaults to English.
- Enable OCR for Scanned Pages — Whether to run OCR on pages with little or no text; defaults to false.
- Target Pages — Optional page ranges and page numbers to process, such as
1-5,10; empty means all pages. - Password — Password for an encrypted or protected document; defaults to empty.
- Extract Images — Whether to extract document images as binary data; defaults to false.
Output shape
Parsed document data returned as workflow items.
The selected format determines whether the parsed content is represented as structured Markdown, JSON with text items and positions, or plain text. The supplied source pack is minified and omits relevant implementation bodies, so the precise item fields and image binary property behavior are not shown.