orchardcore-ai-documents-extractors

Orchard Core AI Documents PDF and OpenXml Extractors

Configure document text extraction

You are an Orchard Core expert. Add optional PDF and Microsoft Open XML document extractors to the CrestApps AI Documents processing pipeline. Enable only the file format support required by the application, then combine it with a document context and optional vector indexing backend.

Guidelines

• Enable the exact feature IDs CrestApps.OrchardCore.AI.Documents.Pdf and CrestApps.OrchardCore.AI.Documents.OpenXml for PDF and Office Open XML support.
• These modules intentionally have no manifest dependency on the base AI Documents feature. They can be installed independently, but are useful only when an AI Documents capability processes uploaded files.
• An AI Documents context or backend enables the base capability by dependency. Examples include Chat Interactions, Profiles, Chat Sessions, Azure AI Search, and Elasticsearch.
• PDF registration calls AddCoreAIPdfDocumentProcessing; OpenXml registration calls AddCoreAIOpenXmlDocumentProcessing.
• Install both packages in the web or startup project if users need both PDF and Office files.
• Extracted text is input to the normal pipeline. It does not itself create chunks, embeddings, or vector indexes.
• Use an embedding provider and a configured index profile when the application needs retrieved document context rather than only extraction.

Feature overview

The PDF module uses PdfPig as an IngestionDocumentReader. The OpenXml module uses the Microsoft DocumentFormat.OpenXml SDK to extract supported Office document text.

Enable both extractors

{
  "steps": [
    {
      "name": "Feature",
      "enable": [
        "CrestApps.OrchardCore.AI.Documents.ChatInteractions",
        "CrestApps.OrchardCore.AI.Documents.AzureAI",
        "CrestApps.OrchardCore.AI.Documents.Pdf",
        "CrestApps.OrchardCore.AI.Documents.OpenXml"
      ],
      "disable": []
    }
  ]
}