start = element pdf2xml { Firstpage? | (Firstpage, Page+) }
Firstpage = element page { PageAttributes, Fontspec+, Text* }
Page = element page { PageAttributes, Text* }
Fontspec =
    element fontspec {
        attribute color { text },
        attribute family { xsd:NCName },
        attribute id { xsd:integer },
        attribute size { xsd:integer }
    }
Text =
    element text {
        attribute font { xsd:integer },
        BoundingBox,
        FreeText+
    }
FreeText =
    mixed {
        element i { FreeText }
        | element a {
              FreeText,
              attribute href { xsd:anyURI }
          }
        | element b { FreeText }
    }
    
# attributes
PageAttributes = BoundingBox, Number, Position
BoundingBox = Height, Left, Top, Width
Number = attribute number { xsd:integer }
Position = attribute position { xsd:NCName }
Height = attribute height { xsd:integer }
Left = attribute left { xsd:integer }
Top = attribute top { xsd:integer }
Width = attribute width { xsd:integer }