File Document Loader
File Document Loader
Section titled “File Document Loader”Load a file (PDF, HTML, text) as text to feed into the Indexer.
What it does
Section titled “What it does”Takes a file/binary payload attached to the incoming item, decodes it to plain text based on its
type, and outputs that text under a configurable field (default text). Connect that output to
the Indexer node’s Input port to index a file.
Supported formats:
- PDF — text is extracted with the same engine as the PDF Parser.
- HTML — converted to clean plain text.
- Text formats —
csv,json,xml,markdown,txt(and anytext/*) are decoded as text.
When to use it
Section titled “When to use it”Use it to index the contents of an uploaded or fetched file — a PDF report, an exported CSV, a Markdown doc — into a knowledge base.
Inputs and settings
Section titled “Inputs and settings”| Setting | Notes |
|---|---|
| Binary key / field | The binary attachment key on the input item (its $binary map), or a field holding a base64 / data-URL string. Defaults to data. |
| Output field name | Field that holds the extracted text. Defaults to text. |
| MIME type | Optional hint used to pick the decoder. Required for a raw base64 string with no data-URL prefix. |
| File name | Optional hint — its extension is used to pick the decoder when no MIME type is known. |
| Include input fields | When enabled, the incoming item’s fields are merged into the output. |
Outputs
Section titled “Outputs”An item with the extracted text under the chosen output field (default text).
- The loader first looks for a real binary attachment on the item’s
$binarymap under the configured key; if none is found it falls back to a plain field of the same name holding a base64 string or adata:URL. - For a PDF you already have on an item you can use either this loader or the PDF Parser — this loader is the choice when the goal is to feed the Indexer and you may be handling PDF, HTML, or text interchangeably.
Troubleshooting
Section titled “Troubleshooting”No source node is connected to this node.— connect a node that provides the file.No binary attachment or field "…" found— check the Binary key matches the attachment or field carrying the payload.Unsupported document type …— set a MIME type or file name so the loader can pick a decoder, or convert the file to a supported format first.