page

Documentation

Index

• Constants
• Variables
• type AlternativeImage
• type Coords
  • func (c *Coords) MarshalXML(e *xml.Encoder, start xml.StartElement) error
  • func (c *Coords) UnmarshalXML(d *xml.Decoder, start xml.StartElement) error
• type Glyph
• type Metadata
  • func (m Metadata) MarshalXML(e *xml.Encoder, start xml.StartElement) error
  • func (m Metadata) UnmarshalXML(d *xml.Decoder, start xml.StartElement) error
• type OrderedGroup
• type Page
• type PcGts
  • func Open(path string) (*PcGts, error)
  • func OpenFromHOCR(file string) (*PcGts, error)
  • func Read(r io.Reader) (*PcGts, error)
  • func ReadFromHOCR(in io.Reader) (*PcGts, error)
  • func (p *PcGts) UnmarshalXML(d *xml.Decoder, start xml.StartElement) error
• type PrintSpace
• type ReadingOrder
• type RegionRefIndexed
• type TextEquiv
• type TextLine
  • func (tl *TextLine) UpdateWords(str string)
• type TextRegion
• type TextRegionBase
• type TextStyle
• type Word

Constants

const (
	XMLNameSpace      = "http://schema.primaresearch.org/PAGE/gts/pagecontent/2013-07-15"
	XMLSchemaInstance = "http://www.w3.org/2001/XMLSchema-instance"
	XMLSchemaLocation = "http://schema.primaresearch.org/PAGE/gts/pagecontent/2013-07-15" +
		" http://schema.primaresearch.org/PAGE/gts/pagecontent/2013-07-15/pagecontent.xsd"
)

XML namespace, schema instance and location.

Variables

var PcGtsXMLHeader = []xml.Attr{
	xml.Attr{

		Value: XMLNameSpace,
	},
	xml.Attr{
		Name:  xml.Name{Space: "xmlns", Local: "xsi"},
		Value: XMLSchemaInstance,
	},
	xml.Attr{
		Name:  xml.Name{Space: "xsi", Local: "schemaLocation"},
		Value: XMLSchemaLocation,
	},
}

PcGtsXMLHeader defines the default xml namespace header.

Types

type AlternativeImage

type AlternativeImage struct {
	Filename string  `xml:"filename,attr"`
	Comments string  `xml:"comments,attr"`
	Conf     float64 `xml:"conf,attr"`
}

AlternativeImage defines optional alternative image information.

type Coords

type Coords struct {
	Points []image.Point `xml:"points,attr"`
}

Coords are rectangles of points.

func (*Coords) MarshalXML

func (c *Coords) MarshalXML(e *xml.Encoder, start xml.StartElement) error

MarshalXML marshals a Coords instance. <Coords points="x0,y0 x1,y1 x2,y2,..."/>

func (*Coords) UnmarshalXML

func (c *Coords) UnmarshalXML(d *xml.Decoder, start xml.StartElement) error

UnmarshalXML unmarshals a Coords instance.

type Glyph

type Glyph struct {
	TextRegionBase
}

Glyph is a single character in a word.

type Metadata

type Metadata map[string]string

Metadata defines

func (Metadata) MarshalXML

func (m Metadata) MarshalXML(e *xml.Encoder, start xml.StartElement) error

MarshalXML marshals the Metadata of a PcGts structure to xml. <Metadata> <Creator>OCR-D</Creator> ... </Metadata>

func (Metadata) UnmarshalXML

func (m Metadata) UnmarshalXML(d *xml.Decoder, start xml.StartElement) error

UnmarshalXML unmarshals the Metadata of a PcGts structure from xml.

type OrderedGroup

type OrderedGroup struct {
	ID               string `xml:"id,attr"`
	Caption          string `xml:"caption,attr"`
	RegionRefIndexed []RegionRefIndexed
}

OrderedGroup is a collection of regions.

type Page

type Page struct {
	ImageFilename string `xml:"imageFilename,attr"`
	ImageHeight   int    `xml:"imageHeight,attr"`
	ImageWidth    int    `xml:"imageWidth,attr"`
	Type          string `xml:"type,attr"`
	PrintSpace    PrintSpace
	ReadingOrder  ReadingOrder
	TextRegion    []TextRegion
}

Page is a page in a PcGts structure.

type PcGts

type PcGts struct {
	Attributes []xml.Attr `xml:",attr"`
	Metadata   Metadata   `xml:"Metadata"`
	Page       Page       `xml:"Page"`
}

PcGts is the top level node of page XML files.

func Open

func Open(path string) (*PcGts, error)

Open reads a new page xml file from the given file path.

func OpenFromHOCR

func OpenFromHOCR(file string) (*PcGts, error)

OpenFromHOCR reads a hOCR file. Returns the hOCR content as PageXML structure. This method assumes one page per hOCR document.

func Read

func Read(r io.Reader) (*PcGts, error)

Read reads a new page xml file from an input stream.

func ReadFromHOCR

func ReadFromHOCR(in io.Reader) (*PcGts, error)

ReadFromHOCR parses a hOCR file. Returns the hOCR content as PageXML structure. This method assumes one page per hOCR document.

func (*PcGts) UnmarshalXML

func (p *PcGts) UnmarshalXML(d *xml.Decoder, start xml.StartElement) error

UnmarshalXML unmarshals the top-level PcGts node of page xml files.

type PrintSpace

type PrintSpace struct {
	Coords Coords
}

PrintSpace defines the print space of a page.

type ReadingOrder

type ReadingOrder struct {
	OrderedGroup []OrderedGroup
}

ReadingOrder is a collection of ordered groups.

type RegionRefIndexed

type RegionRefIndexed struct {
	Index     int    `xml:"index,attr"`
	RegionRef string `xml:"regionRef,attr"`
}

RegionRefIndexed is a index region.

type TextEquiv

type TextEquiv struct {
	PlainText []string
	Unicode   []string
	DT        string  `xml:"dataType,attr,omitempty"`
	DTD       string  `xml:"dataTypeDetails,attr,omitempty"`
	Index     int     `xml:"index,attr,omitempty"`
	Conf      float64 `xml:"conf,attr,omitempty"`
}

TextEquiv defines the text string of text regions.

type TextLine

type TextLine struct {
	TextRegionBase
	BaseLine Coords `xml:"Baseline"`
	Word     []Word
}

TextLine is a line of text in a text region.

func (*TextLine) UpdateWords

func (tl *TextLine) UpdateWords(str string)

UpdateWords updates the words of this text line using the given string (should be most likely TextEquiv.Unicode[0]). Any existing words are discared. The bounding boxes are approximatly calculated based on the number of (unicode) characters of the different words.

type TextRegion

type TextRegion struct {
	TextRegionBase
	Type     string `xml:"type,attr"`
	TextLine []TextLine
}

TextRegion is a region of text (paragraph, block, ...)

type TextRegionBase

type TextRegionBase struct {
	ID               string `xml:"id,attr"`
	Custom           string `xml:"custom,attr"`
	Coords           Coords
	TextStyle        TextStyle
	TextEquiv        TextEquiv         // TODO: multiple TextEquivs are allowed
	AlternativeImage *AlternativeImage `xml:"AlternativeImage,omitempty"`
}

TextRegionBase defines the base data structure for all text regions (TextRegion, Line, Word, Glyph) in a page XML document.

type TextStyle

type TextStyle struct {
	FontFamaily  string  `xml:"fontFamily,attr,omitempty"`
	Serif        bool    `xml:"serif,attr,omitempty"`
	Monospace    bool    `xml:"monospace,attr,omitempty"`
	FontSize     float32 `xml:"fontSize,attr,omitempty"`
	Kerning      int     `xml:"kerning,attr,omitempty"`
	TextColor    string  `xml:"textColour,attr,omitempty"`
	TextColorRGB int     `xml:"textColourRgb,attr,omitempty"`
}

TextStyle specifies font information of any text region.

type Word

type Word struct {
	TextRegionBase
	Glyph []Glyph
}

Word is a token in a line.