page2tsv/qurator/tsvtools/ocrd-tool.json

{
  "version": "0.0.1",
  "git_url": "https://github.com/qurator-spk/page2tsv",
  "tools": {
    "ocrd-neat-export": {
      "executable": "ocrd-neat-export",
      "description": "Convert PAGE-XML to neat-loadable TSV",
      "categories": [ "Format-Conversion" ],
      "steps": [ "format-conversion" ],
      "input_file_grp": ["INPUT"],
      "output_file_grp": ["OUTPUT"],
      "parameters": {
        "iiif_url_template": {
          "type": "string",
          "description": "URL template for lookup of images via IIIF based on {{ unique_identifier }}, {{ page_id }}, {{ page_no }} and {{ PPN }}. 'left', 'top', 'right', 'bottom', 'width', and 'height' are replaced by the neat JS.",
          "default": "https://content.staatsbibliothek-berlin.de/dc/{{ PPN }}-{{ page_no }}/left,top,width,height/full/0/default.jpg"
        },
        "scale_filegrp": {
          "type": "string",
          "description": "If the OCR was run on images with a different resolution than the 'full' IIIF size, use the images in this file group to scale. Set to empty string to disable",
          "default": ""
        },
        "noproxy": {
          "type": "boolean",
          "description": "Disable proxy if set",
          "default": true
        }
      }
    },
    "ocrd-neat-import": {
      "executable": "ocrd-neat-export",
      "description": "Re-integrate TSV into PAGE-XML",
      "categories": [ "Format-Conversion" ],
      "steps": [ "format-conversion" ],
      "input_file_grp": ["PAGE-GRP,TSV-GRP"],
      "output_file_grp": ["OUTPUT"],
      "parameters": {
        "keep_words": {
          "type": "boolean",
          "description": "After updating the line TextEquiv, remove (false) or keep (true) existing and probably inconsistent pc:Word",
          "default": false
        }
      }
    }
  }
}