Processor #

A processor performs specific actions on the input document when it flows through the pipeline:

[Doc input] --> [Processor] --> [Doc output]

Pipeline #

A pipeline is basically multiple processors chained together, users can achieve sophisticated document processing via pipelines:

[Doc input] --> [Processor A] --> [Processor B] --> [Processor C] --> [Doc output]

Processor categories #

Processors are grouped by the type of input they expect.

Document Processors #

Operate on documents. Each pipeline message carries a serialized document.

Attachment Processors #

Operate on attachments. Each pipeline message carries the serialized attachment metadata. Attachment processors load the binary content themselves when they need it.