Portrait uses a sophisticated sourcing strategy that can import unstructured or structured data. From different sources. The sourcing strategy relies on the fact that any data is already given and can be extracted. Portrait takes care of connections, extraction, updates and deletions. An external source is always the “lead” and portrait can therefore automatically manage the data stored in the portrait index.
Strictly speaking, every entry in the index is called a document. However a document, shouldn’t be confused with a file (e.g. a pdf-file).
External data can be databases or other systems. See here for the currently available connectors. The source must be at least accessible by the backend of portrait.
Sources defines and therefore limits what data is pulled from external. A source can detect if data was added, updated or deleted. How often data is pulled, is defined by a reindex interval.
Each document in the index needs a unique key, otherwise the update-strategy won’t work.
A Section defines how the source data is presented on the frontend. Sections are linked directly to a source. So a section cannot “show” the content of two different sources.