ocrd-tool.json

{
  "version": "0.11.2",
  "git_url": "https://github.com/OCR-D/ocrd_fileformat",
  "tools": {
    "ocrd-fileformat-transform": {
      "executable": "ocrd-fileformat-transform",
      "description": "Convert between OCR file formats",
      "categories": [
        "Image preprocessing"
      ],
      "steps": [
        "preprocessing/optimization"
      ],
      "input_file_grp": [
        "OCR-D-OCR-PAGE",
        "OCR-D-OCR-ALTO",
        "OCR-D-OCR-HOCR"
      ],
      "output_file_grp": [
        "OCR-D-OCR-PAGE",
        "OCR-D-OCR-ALTO",
        "OCR-D-OCR-HOCR"
      ],
      "parameters": {
        "from-to": {
          "description": "Transformation scenario, see ocr-fileformat -L",
          "type": "string",
          "default": "page alto",
          "enum": [
            "abbyy hocr",
            "abbyy page",
            "alto2.0 alto3.0",
            "alto2.0 alto3.1",
            "alto2.0 hocr",
            "alto2.1 alto3.0",
            "alto2.1 alto3.1",
            "alto2.1 hocr",
            "alto page",
            "alto text",
            "gcv hocr",
            "gcv page",
            "hocr alto2.0",
            "hocr alto2.1",
            "hocr page",
            "hocr text",
            "page alto",
            "page alto_legacy",
            "page hocr",
            "page page2019",
            "page text",
            "tei hocr",
            "textract page"
          ]
        },
        "ext": {
          "description": "Output extension. Set to empty string to derive extension from the media type.",
          "type": "string",
          "default": ""
        },
        "script-args": {
          "description": "Arguments to Saxon (for XSLT transformations) or to transformation script",
          "type": "string",
          "default": ""
        }
      }
    }
  }
}