ParsePdfDocument
Description
Parses a PDF file, extracting the text and additional information into a structured JSON document. Additionally, any images or tables found in the document are extracted and routed to separate relationships. Their contents are optionally stored in the Document JSON itself, either as a Base64 encoded image or as text extracted from the image (or both). This Processor extracts the information in such a manner as to preserve the original document's layout, including the hierarchy of the sections within the document.
Tags
datavolo, document, extract, image, ocr, parse, pdf, rag, retrieval augmented generation, table, text, unstructured
Properties
In the list below required Properties are shown with an asterisk (*). Other properties are considered optional. The table also indicates any default values, and whether a property supports the NiFi Expression Language.
| Display Name | API Name | Default Value | Allowable Values | Description |
|---|---|---|---|---|
| Service Location Strategy * | Service Location Strategy | Default |
| Determines how Service Locations are configured within this processor for the Datavolo Document Element Detection Service. |
| Custom Element Detection Service URL * | Custom Element Detection Service URL | The Custom URL to the Datavolo Document Element Detection Service. This property is only considered if:
| ||
| OCR Service * | OCR Service | Controller Service: OCRService Implementations: StandardOCRService | An OCR Service for reading files to output text. | |
| Table Embedding Strategy * | Table Embedding Strategy | Extract as Text |
| When a table is found in the document, this property specifies how the table should be embedded into the document. |
| Image Embedding Strategy * | Image Embedding Strategy | Extract as Text |
| When an image is found in the document, this property specifies how the image should be embedded into the document. |
| Communication Timeout * | Communication Timeout | 60 sec | The amount of time to wait for a response from the microservices before timing out. |
Dynamic Properties
This component does not support dynamic properties.
Relationships
| Name | Description |
|---|---|
| comms.failure | If the processor is unable to communicate with one of the necessary services, the input FlowFile will be routed to this relationship. |
| failure | If the text of a FlowFile cannot be extracted for any reason, the input FlowFile will be routed to this relationship. |
| images | If an image is found in the document, the image will be routed to this relationship. |
| success | The text of the PDF is routed to the success relationship. |
| tables | If a table is found in the document, the image of the table will be routed to this relationship. |
Reads Attributes
This processor does not read attributes.
Writes Attributes
| Name | Description |
|---|---|
| container.scope | The scope of the container is set to DOCUMENT for the JSON Document, TABLE for tables, and FIGURE for any figures/images identified |
| document.id | A unique UUID for the document |
| fragment.count | The total number of fragments |
| fragment.index | The index of the fragment |
| mime.type | The MIME type is set to 'application/json' for the JSON document, 'image/png' for any extracted images. |
| page.count | The number of pages in the PDF file is added to the JSON document. |
State Management
This component does not store state.
Restricted
This component is not restricted.
Input Requirement
This component requires an incoming relationship.
System Resource Considerations
This component does not specify system resource considerations.
See Also
ChunkText, ChunkDocument, MergeDocumentElements, ParseTableImage