Skip to main content
View source

Parser

View as Markdown

A RocketRide filter that extracts structured content from a wide variety of document types, routing each kind of embedded content to its own output lane.

What it does​

Extracts structured content from a wide variety of document types. The parser automatically identifies embedded content and routes it to the appropriate output lane, making text, tables, images, audio, and video accessible for downstream processing.

Lanes​

Lane inLane outDescription
tagstextExtracted plain text.
tagstableExtracted tables.
tagsimageExtracted images.
tagsvideoExtracted video streams.
tagsaudioExtracted audio streams.

Configuration​

This node declares no fields of its own.

Notes​

Provenance — not for audit-grade extraction. Output is flattened to Markdown: page boundaries, bbox/polygon coordinates, and the table-HTML cell grid are dropped, so cell-level provenance and coordinate-based review cannot be reconstructed. Do not use this node where every value must trace back to its exact source cell, such as financial-table or regulatory extraction. For structure-preserving parsing that retains table HTML plus page and coordinate data, use the datalab_parse node instead.

Media is extracted only for connected lanes. Images, audio and video are pulled out only when something is connected to the image, audio or video lane. This covers media embedded in documents parsed by Tika and standalone image, audio and video files, which still yield their metadata. A pipeline that uses only text or table skips them, which keeps PDFs that draw their pages as thousands of small images as fast to parse as any other. Parsers loaded by reflection, such as the optional Aspose PDF parser, still extract the images; only sending them to the lane is skipped. Connect the image lane, for example to an OCR node, to receive them.

Schema​

No configuration fields.