{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "$id": "https://raw.githubusercontent.com/api-evangelist/docling/refs/heads/main/json-schema/docling-document-schema.json",
  "title": "DoclingDocument",
  "description": "Lossless representation of a parsed document produced by Docling. Captures structural elements (texts, tables, pictures, key-value items), provenance, layout, and hierarchy across pages.",
  "type": "object",
  "required": ["schema_name", "version", "name"],
  "properties": {
    "schema_name": {
      "type": "string",
      "const": "DoclingDocument"
    },
    "version": {
      "type": "string",
      "description": "Schema version, e.g. 1.4.0."
    },
    "name": {
      "type": "string",
      "description": "Logical document name (typically the source filename without extension)."
    },
    "origin": {
      "type": "object",
      "description": "Provenance of the source artifact.",
      "properties": {
        "mimetype": {"type": "string"},
        "binary_hash": {"type": "string"},
        "filename": {"type": "string"},
        "uri": {"type": "string", "format": "uri"}
      }
    },
    "furniture": {
      "type": "array",
      "description": "Non-content elements (headers, footers, page numbers).",
      "items": {"$ref": "#/$defs/RefItem"}
    },
    "body": {
      "type": "object",
      "description": "Root of the structural hierarchy.",
      "properties": {
        "self_ref": {"type": "string"},
        "children": {
          "type": "array",
          "items": {"$ref": "#/$defs/RefItem"}
        }
      }
    },
    "groups": {
      "type": "array",
      "description": "Grouping nodes (sections, lists).",
      "items": {"$ref": "#/$defs/GroupItem"}
    },
    "texts": {
      "type": "array",
      "items": {"$ref": "#/$defs/TextItem"}
    },
    "tables": {
      "type": "array",
      "items": {"$ref": "#/$defs/TableItem"}
    },
    "pictures": {
      "type": "array",
      "items": {"$ref": "#/$defs/PictureItem"}
    },
    "key_value_items": {
      "type": "array",
      "items": {"$ref": "#/$defs/KeyValueItem"}
    },
    "pages": {
      "type": "object",
      "description": "Per-page metadata keyed by page number.",
      "additionalProperties": {"$ref": "#/$defs/PageItem"}
    }
  },
  "$defs": {
    "RefItem": {
      "type": "object",
      "properties": {
        "$ref": {"type": "string", "description": "JSON pointer reference to another element."}
      }
    },
    "BoundingBox": {
      "type": "object",
      "required": ["l", "t", "r", "b"],
      "properties": {
        "l": {"type": "number"},
        "t": {"type": "number"},
        "r": {"type": "number"},
        "b": {"type": "number"},
        "coord_origin": {"type": "string", "enum": ["TOPLEFT", "BOTTOMLEFT"]}
      }
    },
    "Provenance": {
      "type": "object",
      "properties": {
        "page_no": {"type": "integer"},
        "bbox": {"$ref": "#/$defs/BoundingBox"},
        "charspan": {
          "type": "array",
          "items": {"type": "integer"},
          "minItems": 2,
          "maxItems": 2
        }
      }
    },
    "TextItem": {
      "type": "object",
      "required": ["self_ref", "label", "text"],
      "properties": {
        "self_ref": {"type": "string"},
        "parent": {"$ref": "#/$defs/RefItem"},
        "children": {
          "type": "array",
          "items": {"$ref": "#/$defs/RefItem"}
        },
        "label": {
          "type": "string",
          "enum": [
            "title",
            "section_header",
            "paragraph",
            "list_item",
            "caption",
            "footnote",
            "page_header",
            "page_footer",
            "code",
            "formula",
            "text"
          ]
        },
        "text": {"type": "string"},
        "orig": {"type": "string"},
        "prov": {
          "type": "array",
          "items": {"$ref": "#/$defs/Provenance"}
        },
        "level": {"type": "integer"}
      }
    },
    "TableItem": {
      "type": "object",
      "required": ["self_ref", "data"],
      "properties": {
        "self_ref": {"type": "string"},
        "label": {"type": "string", "const": "table"},
        "captions": {
          "type": "array",
          "items": {"$ref": "#/$defs/RefItem"}
        },
        "data": {
          "type": "object",
          "properties": {
            "num_rows": {"type": "integer"},
            "num_cols": {"type": "integer"},
            "grid": {
              "type": "array",
              "items": {
                "type": "array",
                "items": {"$ref": "#/$defs/TableCell"}
              }
            }
          }
        },
        "prov": {
          "type": "array",
          "items": {"$ref": "#/$defs/Provenance"}
        }
      }
    },
    "TableCell": {
      "type": "object",
      "properties": {
        "text": {"type": "string"},
        "row_span": {"type": "integer"},
        "col_span": {"type": "integer"},
        "start_row_offset_idx": {"type": "integer"},
        "end_row_offset_idx": {"type": "integer"},
        "start_col_offset_idx": {"type": "integer"},
        "end_col_offset_idx": {"type": "integer"},
        "column_header": {"type": "boolean"},
        "row_header": {"type": "boolean"},
        "row_section": {"type": "boolean"}
      }
    },
    "PictureItem": {
      "type": "object",
      "required": ["self_ref"],
      "properties": {
        "self_ref": {"type": "string"},
        "label": {"type": "string", "const": "picture"},
        "image": {
          "type": "object",
          "properties": {
            "mimetype": {"type": "string"},
            "dpi": {"type": "integer"},
            "size": {
              "type": "object",
              "properties": {"width": {"type": "number"}, "height": {"type": "number"}}
            },
            "uri": {"type": "string"}
          }
        },
        "captions": {
          "type": "array",
          "items": {"$ref": "#/$defs/RefItem"}
        },
        "annotations": {
          "type": "array",
          "items": {
            "type": "object",
            "properties": {
              "kind": {"type": "string", "enum": ["classification", "description"]},
              "text": {"type": "string"},
              "predicted_classes": {
                "type": "array",
                "items": {
                  "type": "object",
                  "properties": {
                    "class_name": {"type": "string"},
                    "confidence": {"type": "number"}
                  }
                }
              }
            }
          }
        },
        "prov": {
          "type": "array",
          "items": {"$ref": "#/$defs/Provenance"}
        }
      }
    },
    "KeyValueItem": {
      "type": "object",
      "required": ["self_ref"],
      "properties": {
        "self_ref": {"type": "string"},
        "label": {"type": "string", "const": "key_value_region"},
        "graph": {
          "type": "object",
          "properties": {
            "cells": {
              "type": "array",
              "items": {
                "type": "object",
                "properties": {
                  "cell_id": {"type": "integer"},
                  "text": {"type": "string"},
                  "label": {"type": "string", "enum": ["key", "value"]}
                }
              }
            },
            "links": {
              "type": "array",
              "items": {
                "type": "object",
                "properties": {
                  "source_cell_id": {"type": "integer"},
                  "target_cell_id": {"type": "integer"},
                  "label": {"type": "string"}
                }
              }
            }
          }
        }
      }
    },
    "GroupItem": {
      "type": "object",
      "properties": {
        "self_ref": {"type": "string"},
        "label": {"type": "string", "enum": ["section", "list", "ordered_list", "unordered_list"]},
        "name": {"type": "string"},
        "children": {
          "type": "array",
          "items": {"$ref": "#/$defs/RefItem"}
        }
      }
    },
    "PageItem": {
      "type": "object",
      "properties": {
        "page_no": {"type": "integer"},
        "size": {
          "type": "object",
          "properties": {"width": {"type": "number"}, "height": {"type": "number"}}
        },
        "image": {
          "type": "object",
          "properties": {"uri": {"type": "string"}, "dpi": {"type": "integer"}}
        }
      }
    }
  }
}