schema_version: 3
provider: fai
name: text-extract
version: 0.1.0

# Capability provided by this module.
provides:
  - capability: text.extract
    version: 0.1.0

# Inputs the invoke function accepts.
inputs:
  document:
    type: bytes
    description:
      en: |
        The document to extract text from. Supported MIME types:
          application/pdf
          application/x-pdf
          application/vnd.openxmlformats-officedocument.wordprocessingml.document
      de: |
        Das Dokument, aus dem Text extrahiert werden soll. Unterstützte MIME-Typen:
          application/pdf
          application/x-pdf
          application/vnd.openxmlformats-officedocument.wordprocessingml.document

# Outputs produced.
outputs:
  extracted:
    type: json
    description:
      en: |
        JSON-encoded ExtractedDocument:
          { engine, engine_version, pages: [{number, text, confidence?, bbox?}] }
        confidence and bbox are reserved for a future service-mode variant
        (pdfium / mupdf via host services). v0.1.0 omits them.
      de: |
        JSON-codiertes ExtractedDocument:
          { engine, engine_version, pages: [{number, text, confidence?, bbox?}] }
        confidence und bbox sind für eine künftige Service-Mode-Variante reserviert
        (pdfium / mupdf über Host-Services). v0.1.0 liefert sie nicht.

# Permissions required.
#
# v0.1.0 runs entirely in-WASM with pure-Rust crates (pdf-extract,
# zip, quick-xml). No filesystem, no network. The empty list makes
# this explicit; the hub enforces it.
permissions: []

# MIME-type allow-list for the `document` input. Studio reads this
# through ModuleInfoResponse.accepts_mime and uses it as the OS
# file-picker filter so operators don't accidentally pick a `.png`
# for a PDF-extract step. Same list documented inline next to
# `inputs.document` above; kept synchronised by hand for now.
accepts_mime:
  - application/pdf
  - application/x-pdf
  - application/vnd.openxmlformats-officedocument.wordprocessingml.document