|
|
<?xml version="1.0" encoding="UTF-8" ?>
|
|
|
<schema targetNamespace="http://schema.primaresearch.org/PAGE/gts/pagecontent/2019-07-15"
|
|
|
xmlns:pc="http://schema.primaresearch.org/PAGE/gts/pagecontent/2019-07-15"
|
|
|
elementFormDefault="qualified"
|
|
|
xmlns="http://www.w3.org/2001/XMLSchema"
|
|
|
xmlns:xsd="http://www.w3.org/2001/XMLSchema">
|
|
|
<!-- defaults in XMLSchema:
|
|
|
attribute use="optional"
|
|
|
element minOccurs="1" maxOccurs="1" abstract="false" nillable="false"
|
|
|
-->
|
|
|
|
|
|
<element name="PcGts" type="pc:PcGtsType">
|
|
|
<annotation>
|
|
|
<documentation>Page Content - Ground Truth and Storage</documentation>
|
|
|
</annotation>
|
|
|
</element>
|
|
|
<complexType name="PcGtsType">
|
|
|
<sequence>
|
|
|
<element name="Metadata" type="pc:MetadataType"></element>
|
|
|
<element name="Page" type="pc:PageType"></element>
|
|
|
</sequence>
|
|
|
<attribute name="pcGtsId" type="ID"/>
|
|
|
</complexType>
|
|
|
<complexType name="MetadataType">
|
|
|
<sequence>
|
|
|
<element name="Creator" type="string"/>
|
|
|
<element name="Created" type="dateTime">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The timestamp has to be in UTC (Coordinated
|
|
|
Universal Time) and not local time.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</element>
|
|
|
<element name="LastChange" type="dateTime">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The timestamp has to be in UTC
|
|
|
(Coordinated Universal Time)
|
|
|
and not local time.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</element>
|
|
|
<element name="Comments" type="string"
|
|
|
minOccurs="0" maxOccurs="1">
|
|
|
</element>
|
|
|
<element name="UserDefined" type="pc:UserDefinedType"
|
|
|
minOccurs="0" maxOccurs="1">
|
|
|
</element>
|
|
|
<element name="MetadataItem" type="pc:MetadataItemType"
|
|
|
minOccurs="0" maxOccurs="unbounded">
|
|
|
</element>
|
|
|
</sequence>
|
|
|
<attribute name="externalRef" type="string">
|
|
|
<annotation>
|
|
|
<documentation>External reference of any kind</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
</complexType>
|
|
|
<complexType name="MetadataItemType">
|
|
|
<sequence>
|
|
|
<element name="Labels" type="pc:LabelsType"
|
|
|
minOccurs="0" maxOccurs="unbounded">
|
|
|
<annotation>
|
|
|
<documentation>Semantic labels / tags</documentation>
|
|
|
</annotation>
|
|
|
</element>
|
|
|
</sequence>
|
|
|
<attribute name="type">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Type of metadata (e.g. author)
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
<simpleType>
|
|
|
<restriction base="string">
|
|
|
<enumeration value="author"/>
|
|
|
<enumeration value="imageProperties"/>
|
|
|
<enumeration value="processingStep"/>
|
|
|
<enumeration value="other"/>
|
|
|
</restriction>
|
|
|
</simpleType>
|
|
|
</attribute>
|
|
|
<attribute name="name" type="string">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
E.g. imagePhotometricInterpretation
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="value" type="string" use="required">
|
|
|
<annotation>
|
|
|
<documentation>E.g. RGB</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="date" type="dateTime"/>
|
|
|
</complexType>
|
|
|
<complexType name="LabelsType">
|
|
|
<sequence>
|
|
|
<element name="Label" type="pc:LabelType"
|
|
|
minOccurs="0" maxOccurs="unbounded">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
A semantic label / tag
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</element>
|
|
|
</sequence>
|
|
|
<attribute name="externalModel" type="string">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Reference to external model / ontology / schema
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="externalId" type="string">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
E.g. an RDF resource identifier
|
|
|
(to be used as subject or object of an RDF triple)
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="prefix" type="string">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Prefix for all labels (e.g. first part of an URI)
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="comments" type="string"/>
|
|
|
</complexType>
|
|
|
<complexType name="LabelType">
|
|
|
<annotation>
|
|
|
<documentation>Semantic label</documentation>
|
|
|
</annotation>
|
|
|
<attribute name="value" type="string" use="required">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The label / tag (e.g. 'person').
|
|
|
Can be an RDF resource identifier
|
|
|
(e.g. object of an RDF triple).
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="type" type="string">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Additional information on the label
|
|
|
(e.g. 'YYYY-mm-dd' for a date label).
|
|
|
Can be used as predicate of an RDF triple.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="comments" type="string">
|
|
|
</attribute>
|
|
|
</complexType>
|
|
|
<complexType name="PageType">
|
|
|
<sequence>
|
|
|
<element name="AlternativeImage" type="pc:AlternativeImageType"
|
|
|
minOccurs="0" maxOccurs="unbounded">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Alternative document page images
|
|
|
(e.g. black-and-white).
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</element>
|
|
|
<element name="Border" type="pc:BorderType" minOccurs="0"
|
|
|
maxOccurs="1">
|
|
|
</element>
|
|
|
<element name="PrintSpace" type="pc:PrintSpaceType"
|
|
|
minOccurs="0" maxOccurs="1">
|
|
|
</element>
|
|
|
<element name="ReadingOrder" type="pc:ReadingOrderType"
|
|
|
minOccurs="0" maxOccurs="1">
|
|
|
<annotation>
|
|
|
<documentation>Order of blocks within the page.</documentation>
|
|
|
</annotation>
|
|
|
</element>
|
|
|
<element name="Layers" type="pc:LayersType" minOccurs="0"
|
|
|
maxOccurs="1">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Unassigned regions are considered to be in the
|
|
|
(virtual) default layer which is to be treated
|
|
|
as below any other layers.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</element>
|
|
|
<element name="Relations" type="pc:RelationsType"
|
|
|
minOccurs="0">
|
|
|
</element>
|
|
|
<element name="TextStyle" type="pc:TextStyleType"
|
|
|
minOccurs="0" maxOccurs="1">
|
|
|
<annotation>
|
|
|
<documentation>Default text style</documentation>
|
|
|
</annotation>
|
|
|
</element>
|
|
|
<element name="UserDefined" type="pc:UserDefinedType"
|
|
|
minOccurs="0" maxOccurs="1">
|
|
|
</element>
|
|
|
<element name="Labels" type="pc:LabelsType"
|
|
|
minOccurs="0" maxOccurs="unbounded">
|
|
|
<annotation>
|
|
|
<documentation>Semantic labels / tags</documentation>
|
|
|
</annotation>
|
|
|
</element>
|
|
|
<choice minOccurs="0" maxOccurs="unbounded">
|
|
|
<element name="TextRegion" type="pc:TextRegionType"/>
|
|
|
<element name="ImageRegion" type="pc:ImageRegionType"/>
|
|
|
<element name="LineDrawingRegion" type="pc:LineDrawingRegionType"/>
|
|
|
<element name="GraphicRegion" type="pc:GraphicRegionType"/>
|
|
|
<element name="TableRegion" type="pc:TableRegionType"/>
|
|
|
<element name="ChartRegion" type="pc:ChartRegionType"/>
|
|
|
<element name="MapRegion" type="pc:MapRegionType"/>
|
|
|
<element name="SeparatorRegion" type="pc:SeparatorRegionType"/>
|
|
|
<element name="MathsRegion" type="pc:MathsRegionType"/>
|
|
|
<element name="ChemRegion" type="pc:ChemRegionType"/>
|
|
|
<element name="MusicRegion" type="pc:MusicRegionType"/>
|
|
|
<element name="AdvertRegion" type="pc:AdvertRegionType"/>
|
|
|
<element name="NoiseRegion" type="pc:NoiseRegionType"/>
|
|
|
<element name="UnknownRegion" type="pc:UnknownRegionType"/>
|
|
|
<element name="CustomRegion" type="pc:CustomRegionType"/>
|
|
|
</choice>
|
|
|
</sequence>
|
|
|
<attribute name="imageFilename" type="string" use="required">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Contains the image file name including the file extension.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="imageWidth" type="int" use="required">
|
|
|
<annotation>
|
|
|
<documentation>Specifies the width of the image.</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="imageHeight" type="int" use="required">
|
|
|
<annotation>
|
|
|
<documentation>Specifies the height of the image.</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="imageXResolution" type="float">
|
|
|
<annotation>
|
|
|
<documentation>Specifies the image resolution in width.</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="imageYResolution" type="float">
|
|
|
<annotation>
|
|
|
<documentation>Specifies the image resolution in height.</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="imageResolutionUnit">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Specifies the unit of the resolution information
|
|
|
referring to a standardised unit of measurement
|
|
|
(pixels per inch, pixels per centimeter or other).
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
<simpleType>
|
|
|
<restriction base="string">
|
|
|
<enumeration value="PPI"/>
|
|
|
<enumeration value="PPCM"/>
|
|
|
<enumeration value="other"/>
|
|
|
</restriction>
|
|
|
</simpleType>
|
|
|
</attribute>
|
|
|
<attribute name="custom" type="string">
|
|
|
<annotation>
|
|
|
<documentation>For generic use</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="orientation" type="float">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The angle the rectangle encapsulating the page
|
|
|
(or its Border) has to be rotated in clockwise direction
|
|
|
in order to correct the present skew
|
|
|
(negative values indicate anti-clockwise rotation).
|
|
|
(The rotated image can be further referenced
|
|
|
via “AlternativeImage”.)
|
|
|
Range: -179.999,180
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="type" type="pc:PageTypeSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The type of the page within the document
|
|
|
(e.g. cover page).
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="primaryLanguage" type="pc:LanguageSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The primary language used in the page
|
|
|
(lower-level definitions override the page-level definition).
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="secondaryLanguage" type="pc:LanguageSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The secondary language used in the page
|
|
|
(lower-level definitions override the page-level definition).
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="primaryScript" type="pc:ScriptSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The primary script used in the page
|
|
|
(lower-level definitions override the page-level definition).
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="secondaryScript" type="pc:ScriptSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The secondary script used in the page
|
|
|
(lower-level definitions override the page-level definition).
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="readingDirection" type="pc:ReadingDirectionSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The direction in which text within lines
|
|
|
should be read (order of words and characters),
|
|
|
in addition to “textLineOrder”
|
|
|
(lower-level definitions override the page-level definition).
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="textLineOrder" type="pc:TextLineOrderSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The order of text lines within a block,
|
|
|
in addition to “readingDirection”
|
|
|
(lower-level definitions override the page-level definition).
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="conf" type="pc:ConfSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>Confidence value for whole page (between 0 and 1)</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
</complexType>
|
|
|
<complexType name="TextRegionType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Pure text is represented as a text region. This includes
|
|
|
drop capitals, but practically ornate text may be
|
|
|
considered as a graphic.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
<complexContent>
|
|
|
<extension base="pc:RegionType">
|
|
|
<sequence>
|
|
|
<element name="TextLine" type="pc:TextLineType"
|
|
|
minOccurs="0" maxOccurs="unbounded"/>
|
|
|
<element name="TextEquiv" type="pc:TextEquivType"
|
|
|
minOccurs="0" maxOccurs="unbounded"/>
|
|
|
<element name="TextStyle" type="pc:TextStyleType"
|
|
|
minOccurs="0" maxOccurs="1"/>
|
|
|
</sequence>
|
|
|
<attribute name="orientation" type="float">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The angle the rectangle encapsulating the region
|
|
|
has to be rotated in clockwise direction
|
|
|
in order to correct the present skew
|
|
|
(negative values indicate anti-clockwise rotation).
|
|
|
(The rotated image can be further referenced
|
|
|
via “AlternativeImage”.)
|
|
|
Range: -179.999,180
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="type" type="pc:TextTypeSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The nature of the text in the region
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="leading" type="int">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The degree of space in points between the lines of
|
|
|
text (line spacing)
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="readingDirection" type="pc:ReadingDirectionSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The direction in which text within lines
|
|
|
should be read (order of words and characters),
|
|
|
in addition to “textLineOrder”.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="textLineOrder" type="pc:TextLineOrderSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The order of text lines within the block,
|
|
|
in addition to “readingDirection”.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="readingOrientation" type="float">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The angle the baseline of text within the region
|
|
|
has to be rotated (relative to the rectangle
|
|
|
encapsulating the region) in clockwise direction
|
|
|
in order to correct the present skew,
|
|
|
in addition to “orientation”
|
|
|
(negative values indicate anti-clockwise rotation).
|
|
|
Range: -179.999,180
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="indented" type="boolean">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Defines whether a region of text is indented or not
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="align" type="pc:AlignSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>Text align</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="primaryLanguage" type="pc:LanguageSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The primary language used in the region
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="secondaryLanguage" type="pc:LanguageSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The secondary language used in the region
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="primaryScript" type="pc:ScriptSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The primary script used in the region
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="secondaryScript" type="pc:ScriptSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The secondary script used in the region
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="production" type="pc:ProductionSimpleType"/>
|
|
|
</extension>
|
|
|
</complexContent>
|
|
|
</complexType>
|
|
|
<complexType name="CoordsType">
|
|
|
<attribute name="points" type="pc:PointsType" use="required">
|
|
|
<annotation>
|
|
|
<documentation>Point list with format "x1,y1 x2,y2 ..."</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="conf" type="pc:ConfSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>Confidence value (between 0 and 1)</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
</complexType>
|
|
|
<complexType name="TextLineType">
|
|
|
<sequence>
|
|
|
<element name="AlternativeImage" type="pc:AlternativeImageType"
|
|
|
minOccurs="0" maxOccurs="unbounded">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Alternative text line images (e.g.
|
|
|
black-and-white)
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</element>
|
|
|
<element name="Coords" type="pc:CoordsType"></element>
|
|
|
<element name="Baseline" type="pc:BaselineType"
|
|
|
minOccurs="0">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Multiple connected points that mark the baseline
|
|
|
of the glyphs
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</element>
|
|
|
<element name="Word" type="pc:WordType" minOccurs="0"
|
|
|
maxOccurs="unbounded">
|
|
|
</element>
|
|
|
<element name="TextEquiv" type="pc:TextEquivType"
|
|
|
minOccurs="0" maxOccurs="unbounded">
|
|
|
</element>
|
|
|
<element name="TextStyle" type="pc:TextStyleType"
|
|
|
minOccurs="0">
|
|
|
</element>
|
|
|
<element name="UserDefined" type="pc:UserDefinedType"
|
|
|
minOccurs="0" maxOccurs="1">
|
|
|
</element>
|
|
|
<element name="Labels" type="pc:LabelsType"
|
|
|
minOccurs="0" maxOccurs="unbounded">
|
|
|
<annotation>
|
|
|
<documentation>Semantic labels / tags</documentation>
|
|
|
</annotation>
|
|
|
</element>
|
|
|
</sequence>
|
|
|
<attribute name="id" type="ID" use="required"/>
|
|
|
<attribute name="primaryLanguage" type="pc:LanguageSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Overrides primaryLanguage attribute of parent text
|
|
|
region
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="primaryScript" type="pc:ScriptSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The primary script used in the text line
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="secondaryScript" type="pc:ScriptSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The secondary script used in the text line
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="readingDirection" type="pc:ReadingDirectionSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The direction in which text within the line
|
|
|
should be read (order of words and characters).
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="production" type="pc:ProductionSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Overrides the production attribute of the parent
|
|
|
text region
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="custom" type="string">
|
|
|
<annotation>
|
|
|
<documentation>For generic use</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="comments" type="string"/>
|
|
|
<attribute name="index" type="int">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Position (order number) of this text line within the
|
|
|
parent text region.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
</complexType>
|
|
|
<complexType name="WordType">
|
|
|
<sequence>
|
|
|
<element name="AlternativeImage" type="pc:AlternativeImageType"
|
|
|
minOccurs="0" maxOccurs="unbounded">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Alternative word images (e.g.
|
|
|
black-and-white)
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</element>
|
|
|
<element name="Coords" type="pc:CoordsType">
|
|
|
</element>
|
|
|
<element name="Glyph" type="pc:GlyphType" minOccurs="0"
|
|
|
maxOccurs="unbounded">
|
|
|
</element>
|
|
|
<element name="TextEquiv" type="pc:TextEquivType"
|
|
|
minOccurs="0" maxOccurs="unbounded">
|
|
|
</element>
|
|
|
<element name="TextStyle" type="pc:TextStyleType"
|
|
|
minOccurs="0">
|
|
|
</element>
|
|
|
<element name="UserDefined" type="pc:UserDefinedType"
|
|
|
minOccurs="0" maxOccurs="1">
|
|
|
</element>
|
|
|
<element name="Labels" type="pc:LabelsType" minOccurs="0" maxOccurs="unbounded">
|
|
|
<annotation>
|
|
|
<documentation>Semantic labels / tags</documentation>
|
|
|
</annotation>
|
|
|
</element>
|
|
|
</sequence>
|
|
|
<attribute name="id" type="ID" use="required"/>
|
|
|
<attribute name="language" type="pc:LanguageSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Overrides primaryLanguage attribute of parent line
|
|
|
and/or text region
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="primaryScript" type="pc:ScriptSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The primary script used in the word
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="secondaryScript" type="pc:ScriptSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The secondary script used in the word
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="readingDirection" type="pc:ReadingDirectionSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The direction in which text within the word
|
|
|
should be read (order of characters).
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="production" type="pc:ProductionSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Overrides the production attribute of the parent
|
|
|
text line and/or text region.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="custom" type="string">
|
|
|
<annotation>
|
|
|
<documentation>For generic use</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="comments" type="string"/>
|
|
|
</complexType>
|
|
|
<complexType name="GlyphType">
|
|
|
<sequence>
|
|
|
<element name="AlternativeImage" type="pc:AlternativeImageType"
|
|
|
minOccurs="0" maxOccurs="unbounded">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Alternative glyph images (e.g.
|
|
|
black-and-white)
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</element>
|
|
|
<element name="Coords" type="pc:CoordsType"></element>
|
|
|
<element name="Graphemes" type="pc:GraphemesType"
|
|
|
minOccurs="0" maxOccurs="1">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Container for graphemes, grapheme groups and
|
|
|
non-printing characters
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</element>
|
|
|
<element name="TextEquiv" type="pc:TextEquivType"
|
|
|
minOccurs="0" maxOccurs="unbounded">
|
|
|
</element>
|
|
|
<element name="TextStyle" type="pc:TextStyleType"
|
|
|
minOccurs="0">
|
|
|
</element>
|
|
|
<element name="UserDefined" type="pc:UserDefinedType"
|
|
|
minOccurs="0" maxOccurs="1">
|
|
|
</element>
|
|
|
<element name="Labels" type="pc:LabelsType"
|
|
|
minOccurs="0" maxOccurs="unbounded">
|
|
|
<annotation>
|
|
|
<documentation>Semantic labels / tags</documentation>
|
|
|
</annotation>
|
|
|
</element>
|
|
|
</sequence>
|
|
|
<attribute name="id" type="ID" use="required"/>
|
|
|
<attribute name="ligature" use="optional" type="boolean"/>
|
|
|
<attribute name="symbol" use="optional" type="boolean"/>
|
|
|
<attribute name="script" type="pc:ScriptSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The script used for the glyph
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="production" type="pc:ProductionSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Overrides the production attribute of the parent
|
|
|
word / text line / text region.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="custom" type="string">
|
|
|
<annotation>
|
|
|
<documentation>For generic use</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="comments" type="string"/>
|
|
|
</complexType>
|
|
|
<complexType name="TextEquivType">
|
|
|
<sequence>
|
|
|
<element name="PlainText" type="string" minOccurs="0">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Text in a "simple" form (ASCII or extended ASCII
|
|
|
as mostly used for typing). I.e. no use of
|
|
|
special characters for ligatures (should be
|
|
|
stored as two separate characters) etc.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</element>
|
|
|
<element name="Unicode" type="string">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Correct encoding of the original, always using
|
|
|
the corresponding Unicode code point. I.e.
|
|
|
ligatures have to be represented as one
|
|
|
character etc.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</element>
|
|
|
</sequence>
|
|
|
<attribute name="index" use="optional">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Used for sort order in case multiple TextEquivs are defined.
|
|
|
The text content with the lowest index should be interpreted
|
|
|
as the main text content.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
<simpleType>
|
|
|
<restriction base="integer">
|
|
|
<minInclusive value="0"></minInclusive>
|
|
|
</restriction>
|
|
|
</simpleType>
|
|
|
</attribute>
|
|
|
<attribute name="conf" type="pc:ConfSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>OCR confidence value (between 0 and 1)</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="dataType" type="pc:TextDataTypeSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Type of text content (is it free text or a number, for instance).
|
|
|
This is only a descriptive attribute, the text type
|
|
|
is not checked during XML validation.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="dataTypeDetails" type="string">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Refinement for dataType attribute. Can be a regular expression, for instance.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<!-- <attribute name="mergeWithNextRule" type="pc:TextMergeRuleSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>Rule for merging consecutive text objects. The rule applies to the first object of a pair (i.e. 'remove-last' removes the last
|
|
|
character of the first region, can be used to remove hyphen, for example)</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="mergeWithNextRuleData" type="string">
|
|
|
<annotation>
|
|
|
<documentation>Custom data for mergeRule attribute. Can number of characters to be removed, for example.</documentation>
|
|
|
</annotation>
|
|
|
</attribute> -->
|
|
|
<attribute name="comments" type="string"/>
|
|
|
</complexType>
|
|
|
<complexType name="ImageRegionType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
An image is considered to be more intricate and complex
|
|
|
than a graphic. These can be photos or drawings.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
<complexContent>
|
|
|
<extension base="pc:RegionType">
|
|
|
<attribute name="orientation" type="float">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The angle the rectangle encapsulating a region
|
|
|
has to be rotated in clockwise direction
|
|
|
in order to correct the present skew
|
|
|
(negative values indicate anti-clockwise rotation).
|
|
|
Range: -179.999,180
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="colourDepth" type="pc:ColourDepthSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The colour bit depth required for the region
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="bgColour" type="pc:ColourSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The background colour of the region
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="embText" type="boolean">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Specifies whether the region also contains
|
|
|
text
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
</extension>
|
|
|
</complexContent>
|
|
|
</complexType>
|
|
|
<complexType name="LineDrawingRegionType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
A line drawing is a single colour illustration without
|
|
|
solid areas.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
<complexContent>
|
|
|
<extension base="pc:RegionType">
|
|
|
<attribute name="orientation" type="float">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The angle the rectangle encapsulating a region
|
|
|
has to be rotated in clockwise direction
|
|
|
in order to correct the present skew
|
|
|
(negative values indicate anti-clockwise rotation).
|
|
|
Range: -179.999,180
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="penColour" type="pc:ColourSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The pen (foreground) colour of the region
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="bgColour" type="pc:ColourSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The background colour of the region
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="embText" type="boolean">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Specifies whether the region also contains
|
|
|
text
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
</extension>
|
|
|
</complexContent>
|
|
|
</complexType>
|
|
|
<complexType name="GraphicRegionType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Regions containing simple graphics, such as a company
|
|
|
logo, should be marked as graphic regions.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
<complexContent>
|
|
|
<extension base="pc:RegionType">
|
|
|
<attribute name="orientation" type="float">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The angle the rectangle encapsulating a region
|
|
|
has to be rotated in clockwise direction
|
|
|
in order to correct the present skew
|
|
|
(negative values indicate anti-clockwise rotation).
|
|
|
Range: -179.999,180
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="type" type="pc:GraphicsTypeSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The type of graphic in the region
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="numColours" type="int">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
An approximation of the number of colours
|
|
|
used in the region
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="embText" type="boolean">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Specifies whether the region also contains
|
|
|
text.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
</extension>
|
|
|
</complexContent>
|
|
|
</complexType>
|
|
|
<complexType name="TableRegionType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Tabular data in any form is represented with a table
|
|
|
region. Rows and columns may or may not have separator
|
|
|
lines; these lines are not separator regions.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
<complexContent>
|
|
|
<extension base="pc:RegionType">
|
|
|
<sequence>
|
|
|
<element name="Grid" type="pc:GridType"
|
|
|
minOccurs="0" maxOccurs="1">
|
|
|
<annotation>
|
|
|
<documentation>Table grid (visible or virtual grid lines)</documentation>
|
|
|
</annotation>
|
|
|
</element>
|
|
|
</sequence>
|
|
|
<attribute name="orientation" type="float">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The angle the rectangle encapsulating a region
|
|
|
has to be rotated in clockwise direction
|
|
|
in order to correct the present skew
|
|
|
(negative values indicate anti-clockwise rotation).
|
|
|
Range: -179.999,180
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="rows" type="int">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The number of rows present in the table
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="columns" type="int">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The number of columns present in the table
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="lineColour" type="pc:ColourSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The colour of the lines used in the region
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="bgColour" type="pc:ColourSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The background colour of the region
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="lineSeparators" type="boolean">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Specifies the presence of line separators
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="embText" type="boolean">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Specifies whether the region also contains
|
|
|
text
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
</extension>
|
|
|
</complexContent>
|
|
|
</complexType>
|
|
|
<complexType name="GridType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Matrix of grid points defining the table grid on the page.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
<sequence>
|
|
|
<element name="GridPoints" type="pc:GridPointsType"
|
|
|
minOccurs="2" maxOccurs="unbounded">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
One row in the grid point matrix.
|
|
|
Points with x,y coordinates.
|
|
|
(note: for a table with n table rows there should be n+1 grid rows)
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</element>
|
|
|
</sequence>
|
|
|
</complexType>
|
|
|
<complexType name="GridPointsType">
|
|
|
<annotation>
|
|
|
<documentation>Points with x,y coordinates.</documentation>
|
|
|
</annotation>
|
|
|
<attribute name="index" type="int" use="required">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The grid row index
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="points" type="pc:PointsType"
|
|
|
use="required"/>
|
|
|
</complexType>
|
|
|
<complexType name="ChartRegionType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Regions containing charts or graphs of any type, should
|
|
|
be marked as chart regions.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
<complexContent>
|
|
|
<extension base="pc:RegionType">
|
|
|
<attribute name="orientation" type="float">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The angle the rectangle encapsulating a region
|
|
|
has to be rotated in clockwise direction
|
|
|
in order to correct the present skew
|
|
|
(negative values indicate anti-clockwise rotation).
|
|
|
Range: -179.999,180
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="type" type="pc:ChartTypeSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The type of chart in the region
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="numColours" type="int">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
An approximation of the number of colours
|
|
|
used in the region
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="bgColour" type="pc:ColourSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The background colour of the region
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="embText" type="boolean">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Specifies whether the region also contains
|
|
|
text
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
</extension>
|
|
|
</complexContent>
|
|
|
</complexType>
|
|
|
<complexType name="SeparatorRegionType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Separators are lines that lie between columns and
|
|
|
paragraphs and can be used to logically separate
|
|
|
different articles from each other.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
<complexContent>
|
|
|
<extension base="pc:RegionType">
|
|
|
<attribute name="orientation" type="float">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The angle the rectangle encapsulating a region
|
|
|
has to be rotated in clockwise direction
|
|
|
in order to correct the present skew
|
|
|
(negative values indicate anti-clockwise rotation).
|
|
|
Range: -179.999,180
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="colour" type="pc:ColourSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The colour of the separator
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
</extension>
|
|
|
</complexContent>
|
|
|
</complexType>
|
|
|
<complexType name="MathsRegionType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Regions containing equations and mathematical symbols
|
|
|
should be marked as maths regions.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
<complexContent>
|
|
|
<extension base="pc:RegionType">
|
|
|
<attribute name="orientation" type="float">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The angle the rectangle encapsulating a region
|
|
|
has to be rotated in clockwise direction
|
|
|
in order to correct the present skew
|
|
|
(negative values indicate anti-clockwise rotation).
|
|
|
Range: -179.999,180
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="bgColour" type="pc:ColourSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The background colour of the region
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
</extension>
|
|
|
</complexContent>
|
|
|
</complexType>
|
|
|
<complexType name="ChemRegionType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Regions containing chemical formulas.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
<complexContent>
|
|
|
<extension base="pc:RegionType">
|
|
|
<attribute name="orientation" type="float">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The angle the rectangle encapsulating a
|
|
|
region has to be rotated in clockwise
|
|
|
direction in order to correct the present
|
|
|
skew (negative values indicate
|
|
|
anti-clockwise rotation). Range:
|
|
|
-179.999,180
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="bgColour" type="pc:ColourSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The background colour of the region
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
</extension>
|
|
|
</complexContent>
|
|
|
</complexType>
|
|
|
<complexType name="MapRegionType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Regions containing maps.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
<complexContent>
|
|
|
<extension base="pc:RegionType">
|
|
|
<attribute name="orientation" type="float">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The angle the rectangle encapsulating a
|
|
|
region has to be rotated in clockwise
|
|
|
direction in order to correct the present
|
|
|
skew (negative values indicate
|
|
|
anti-clockwise rotation). Range:
|
|
|
-179.999,180
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
</extension>
|
|
|
</complexContent>
|
|
|
</complexType>
|
|
|
<complexType name="MusicRegionType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Regions containing musical notations.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
<complexContent>
|
|
|
<extension base="pc:RegionType">
|
|
|
<attribute name="orientation" type="float">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The angle the rectangle encapsulating a region
|
|
|
has to be rotated in clockwise direction
|
|
|
in order to correct the present skew
|
|
|
(negative values indicate anti-clockwise rotation).
|
|
|
Range: -179.999,180
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="bgColour" type="pc:ColourSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The background colour of the region
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
</extension>
|
|
|
</complexContent>
|
|
|
</complexType>
|
|
|
<complexType name="AdvertRegionType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Regions containing advertisements.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
<complexContent>
|
|
|
<extension base="pc:RegionType">
|
|
|
<attribute name="orientation" type="float">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The angle the rectangle encapsulating a region
|
|
|
has to be rotated in clockwise direction
|
|
|
in order to correct the present skew
|
|
|
(negative values indicate anti-clockwise rotation).
|
|
|
Range: -179.999,180
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="bgColour" type="pc:ColourSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The background colour of the region
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
</extension>
|
|
|
</complexContent>
|
|
|
</complexType>
|
|
|
<complexType name="NoiseRegionType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Noise regions are regions where no real data lies, only
|
|
|
false data created by artifacts on the document or
|
|
|
scanner noise.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
<complexContent>
|
|
|
<extension base="pc:RegionType"></extension>
|
|
|
</complexContent>
|
|
|
</complexType>
|
|
|
<complexType name="UnknownRegionType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
To be used if the region type cannot be ascertained.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
<complexContent>
|
|
|
<extension base="pc:RegionType"></extension>
|
|
|
</complexContent>
|
|
|
</complexType>
|
|
|
<complexType name="CustomRegionType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Regions containing content that is not covered
|
|
|
by the default types (text, graphic, image,
|
|
|
line drawing, chart, table, separator, maths,
|
|
|
map, music, chem, advert, noise, unknown).
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
<complexContent>
|
|
|
<extension base="pc:RegionType">
|
|
|
<attribute name="type" type="string">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Information on the type of content represented by this region
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
</extension>
|
|
|
</complexContent>
|
|
|
</complexType>
|
|
|
<complexType name="PrintSpaceType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Determines the effective area on the paper of a printed page.
|
|
|
Its size is equal for all pages of a book
|
|
|
(exceptions: titlepage, multipage pictures).
|
|
|
It contains all living elements (except marginals)
|
|
|
like body type, footnotes, headings, running titles.
|
|
|
It does not contain pagenumber (if not part of running title),
|
|
|
marginals, signature mark, preview words.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
<sequence>
|
|
|
<element name="Coords" type="pc:CoordsType"/>
|
|
|
</sequence>
|
|
|
</complexType>
|
|
|
<complexType name="ReadingOrderType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Definition of the reading order within the page.
|
|
|
To express a reading order between elements
|
|
|
they have to be included in an OrderedGroup.
|
|
|
Groups may contain further groups.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
<choice minOccurs="1" maxOccurs="1">
|
|
|
<element name="OrderedGroup" type="pc:OrderedGroupType"/>
|
|
|
<element name="UnorderedGroup" type="pc:UnorderedGroupType"/>
|
|
|
</choice>
|
|
|
<attribute name="conf" type="pc:ConfSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>Confidence value (between 0 and 1)</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
</complexType>
|
|
|
<complexType name="RegionRefIndexedType">
|
|
|
<annotation>
|
|
|
<documentation>Numbered region</documentation>
|
|
|
</annotation>
|
|
|
<attribute name="index" type="int" use="required">
|
|
|
<annotation>
|
|
|
<documentation>Position (order number) of this item within the current hierarchy level.</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="regionRef" type="IDREF" use="required"/>
|
|
|
</complexType>
|
|
|
<complexType name="OrderedGroupIndexedType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Indexed group containing ordered elements
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
<sequence>
|
|
|
<element name="UserDefined" type="pc:UserDefinedType"
|
|
|
minOccurs="0" maxOccurs="1"/>
|
|
|
<element name="Labels" type="pc:LabelsType"
|
|
|
minOccurs="0" maxOccurs="unbounded">
|
|
|
<annotation>
|
|
|
<documentation>Semantic labels / tags</documentation>
|
|
|
</annotation>
|
|
|
</element>
|
|
|
<choice minOccurs="1" maxOccurs="unbounded">
|
|
|
<element name="RegionRefIndexed" type="pc:RegionRefIndexedType"/>
|
|
|
<element name="OrderedGroupIndexed" type="pc:OrderedGroupIndexedType"/>
|
|
|
<element name="UnorderedGroupIndexed" type="pc:UnorderedGroupIndexedType"/>
|
|
|
</choice>
|
|
|
</sequence>
|
|
|
<attribute name="id" type="ID" use="required"/>
|
|
|
<attribute name="regionRef" type="IDREF">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Optional link to a parent region of nested regions.
|
|
|
The parent region doubles as reading order group.
|
|
|
Only the nested regions should be allowed as group members.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="index" type="int" use="required">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Position (order number) of this item within the
|
|
|
current hierarchy level.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="caption" type="string"/>
|
|
|
<attribute name="type" type="pc:GroupTypeSimpleType"/>
|
|
|
<attribute name="continuation" type="boolean">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Is this group a continuation of another group (from
|
|
|
previous column or page, for example)?
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="custom" type="string">
|
|
|
<annotation>
|
|
|
<documentation>For generic use</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="comments" type="string"/>
|
|
|
</complexType>
|
|
|
<complexType name="UnorderedGroupIndexedType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Indexed group containing unordered elements
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
<sequence>
|
|
|
<element name="UserDefined" type="pc:UserDefinedType"
|
|
|
minOccurs="0" maxOccurs="1">
|
|
|
</element>
|
|
|
<element name="Labels" type="pc:LabelsType"
|
|
|
minOccurs="0" maxOccurs="unbounded">
|
|
|
<annotation>
|
|
|
<documentation>Semantic labels / tags</documentation>
|
|
|
</annotation>
|
|
|
</element>
|
|
|
<choice minOccurs="1" maxOccurs="unbounded">
|
|
|
<element name="RegionRef" type="pc:RegionRefType"/>
|
|
|
<element name="OrderedGroup" type="pc:OrderedGroupType"/>
|
|
|
<element name="UnorderedGroup" type="pc:UnorderedGroupType"/>
|
|
|
</choice>
|
|
|
</sequence>
|
|
|
<attribute name="id" type="ID" use="required"/>
|
|
|
<attribute name="regionRef" type="IDREF">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Optional link to a parent region of nested regions.
|
|
|
The parent region doubles as reading order group.
|
|
|
Only the nested regions should be allowed as group members.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="index" type="int" use="required">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Position (order number) of this item within the
|
|
|
current hierarchy level.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="caption" type="string"/>
|
|
|
<attribute name="type" type="pc:GroupTypeSimpleType"/>
|
|
|
<attribute name="continuation" type="boolean">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Is this group a continuation of another group
|
|
|
(from previous column or page, for example)?
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="custom" type="string">
|
|
|
<annotation>
|
|
|
<documentation>For generic use</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="comments" type="string"/>
|
|
|
</complexType>
|
|
|
<complexType name="RegionRefType">
|
|
|
<attribute name="regionRef" type="IDREF" use="required"/>
|
|
|
</complexType>
|
|
|
<complexType name="OrderedGroupType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Numbered group (contains ordered elements)
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
<sequence>
|
|
|
<element name="UserDefined" type="pc:UserDefinedType"
|
|
|
minOccurs="0" maxOccurs="1">
|
|
|
</element>
|
|
|
<element name="Labels" type="pc:LabelsType"
|
|
|
minOccurs="0" maxOccurs="unbounded">
|
|
|
<annotation>
|
|
|
<documentation>Semantic labels / tags</documentation>
|
|
|
</annotation>
|
|
|
</element>
|
|
|
<choice minOccurs="1" maxOccurs="unbounded">
|
|
|
<element name="RegionRefIndexed" type="pc:RegionRefIndexedType"/>
|
|
|
<element name="OrderedGroupIndexed" type="pc:OrderedGroupIndexedType"/>
|
|
|
<element name="UnorderedGroupIndexed" type="pc:UnorderedGroupIndexedType"/>
|
|
|
</choice>
|
|
|
</sequence>
|
|
|
<attribute name="id" type="ID" use="required"/>
|
|
|
<attribute name="regionRef" type="IDREF">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Optional link to a parent region of nested regions.
|
|
|
The parent region doubles as reading order group.
|
|
|
Only the nested regions should be allowed as group members.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="caption" type="string"/>
|
|
|
<attribute name="type" type="pc:GroupTypeSimpleType"/>
|
|
|
<attribute name="continuation" type="boolean">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Is this group a continuation of another group
|
|
|
(from previous column or page, for example)?
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="custom" type="string">
|
|
|
<annotation>
|
|
|
<documentation>For generic use</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="comments" type="string"/>
|
|
|
</complexType>
|
|
|
<complexType name="UnorderedGroupType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Numbered group (contains unordered elements)
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
<sequence>
|
|
|
<element name="UserDefined" type="pc:UserDefinedType"
|
|
|
minOccurs="0" maxOccurs="1">
|
|
|
</element>
|
|
|
<element name="Labels" type="pc:LabelsType"
|
|
|
minOccurs="0" maxOccurs="unbounded">
|
|
|
<annotation>
|
|
|
<documentation>Semantic labels / tags</documentation>
|
|
|
</annotation>
|
|
|
</element>
|
|
|
<choice minOccurs="1" maxOccurs="unbounded">
|
|
|
<element name="RegionRef" type="pc:RegionRefType"/>
|
|
|
<element name="OrderedGroup" type="pc:OrderedGroupType"/>
|
|
|
<element name="UnorderedGroup" type="pc:UnorderedGroupType"/>
|
|
|
</choice>
|
|
|
</sequence>
|
|
|
<attribute name="id" type="ID" use="required"/>
|
|
|
<attribute name="regionRef" type="IDREF">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Optional link to a parent region of nested regions.
|
|
|
The parent region doubles as reading order group.
|
|
|
Only the nested regions should be allowed as group members.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="caption" type="string"/>
|
|
|
<attribute name="type" type="pc:GroupTypeSimpleType"/>
|
|
|
<attribute name="continuation" type="boolean">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Is this group a continuation of another group
|
|
|
(from previous column or page, for example)?
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="custom" type="string">
|
|
|
<annotation>
|
|
|
<documentation>For generic use</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="comments" type="string"/>
|
|
|
</complexType>
|
|
|
<complexType name="BorderType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Border of the actual page (if the scanned image
|
|
|
contains parts not belonging to the page).
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
<sequence>
|
|
|
<element name="Coords" type="pc:CoordsType"></element>
|
|
|
</sequence>
|
|
|
</complexType>
|
|
|
<simpleType name="ColourSimpleType">
|
|
|
<restriction base="string">
|
|
|
<enumeration value="black"/>
|
|
|
<enumeration value="blue"/>
|
|
|
<enumeration value="brown"/>
|
|
|
<enumeration value="cyan"/>
|
|
|
<enumeration value="green"/>
|
|
|
<enumeration value="grey"/>
|
|
|
<enumeration value="indigo"/>
|
|
|
<enumeration value="magenta"/>
|
|
|
<enumeration value="orange"/>
|
|
|
<enumeration value="pink"/>
|
|
|
<enumeration value="red"/>
|
|
|
<enumeration value="turquoise"/>
|
|
|
<enumeration value="violet"/>
|
|
|
<enumeration value="white"/>
|
|
|
<enumeration value="yellow"/>
|
|
|
<enumeration value="other"/>
|
|
|
</restriction>
|
|
|
</simpleType>
|
|
|
<simpleType name="ReadingDirectionSimpleType">
|
|
|
<restriction base="string">
|
|
|
<enumeration value="left-to-right"/>
|
|
|
<enumeration value="right-to-left"/>
|
|
|
<enumeration value="top-to-bottom"/>
|
|
|
<enumeration value="bottom-to-top"/>
|
|
|
</restriction>
|
|
|
</simpleType>
|
|
|
<simpleType name="TextLineOrderSimpleType">
|
|
|
<restriction base="string">
|
|
|
<enumeration value="top-to-bottom"/>
|
|
|
<enumeration value="bottom-to-top"/>
|
|
|
<enumeration value="left-to-right"/>
|
|
|
<enumeration value="right-to-left"/>
|
|
|
</restriction>
|
|
|
</simpleType>
|
|
|
<simpleType name="TextTypeSimpleType">
|
|
|
<restriction base="string">
|
|
|
<enumeration value="paragraph"/>
|
|
|
<enumeration value="heading"/>
|
|
|
<enumeration value="caption"/>
|
|
|
<enumeration value="header"/>
|
|
|
<enumeration value="footer"/>
|
|
|
<enumeration value="page-number"/>
|
|
|
<enumeration value="drop-capital"/>
|
|
|
<enumeration value="credit"/>
|
|
|
<enumeration value="floating"/>
|
|
|
<enumeration value="signature-mark"/>
|
|
|
<enumeration value="catch-word"/>
|
|
|
<enumeration value="marginalia"/>
|
|
|
<enumeration value="footnote"/>
|
|
|
<enumeration value="footnote-continued"/>
|
|
|
<enumeration value="endnote"/>
|
|
|
<enumeration value="TOC-entry"/>
|
|
|
<enumeration value="list-label"/>
|
|
|
<enumeration value="other"/>
|
|
|
</restriction>
|
|
|
</simpleType>
|
|
|
<simpleType name="PageTypeSimpleType">
|
|
|
<restriction base="string">
|
|
|
<enumeration value="front-cover"/>
|
|
|
<enumeration value="back-cover"/>
|
|
|
<enumeration value="title"/>
|
|
|
<enumeration value="table-of-contents"/>
|
|
|
<enumeration value="index"/>
|
|
|
<enumeration value="content"/>
|
|
|
<enumeration value="blank"/>
|
|
|
<enumeration value="other"/>
|
|
|
</restriction>
|
|
|
</simpleType>
|
|
|
<simpleType name="ConfSimpleType">
|
|
|
<restriction base="float">
|
|
|
<minInclusive value="0"></minInclusive>
|
|
|
<maxInclusive value="1"></maxInclusive>
|
|
|
</restriction>
|
|
|
</simpleType>
|
|
|
<simpleType name="LanguageSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>ISO 639.x 2016-07-14</documentation>
|
|
|
</annotation>
|
|
|
<restriction base="string">
|
|
|
<enumeration value="Abkhaz"/>
|
|
|
<enumeration value="Afar"/>
|
|
|
<enumeration value="Afrikaans"/>
|
|
|
<enumeration value="Akan"/>
|
|
|
<enumeration value="Albanian"/>
|
|
|
<enumeration value="Amharic"/>
|
|
|
<enumeration value="Arabic"/>
|
|
|
<enumeration value="Aragonese"/>
|
|
|
<enumeration value="Armenian"/>
|
|
|
<enumeration value="Assamese"/>
|
|
|
<enumeration value="Avaric"/>
|
|
|
<enumeration value="Avestan"/>
|
|
|
<enumeration value="Aymara"/>
|
|
|
<enumeration value="Azerbaijani"/>
|
|
|
<enumeration value="Bambara"/>
|
|
|
<enumeration value="Bashkir"/>
|
|
|
<enumeration value="Basque"/>
|
|
|
<enumeration value="Belarusian"/>
|
|
|
<enumeration value="Bengali"/>
|
|
|
<enumeration value="Bihari"/>
|
|
|
<enumeration value="Bislama"/>
|
|
|
<enumeration value="Bosnian"/>
|
|
|
<enumeration value="Breton"/>
|
|
|
<enumeration value="Bulgarian"/>
|
|
|
<enumeration value="Burmese"/>
|
|
|
<enumeration value="Cambodian"/>
|
|
|
<enumeration value="Cantonese"/>
|
|
|
<enumeration value="Catalan"/>
|
|
|
<enumeration value="Chamorro"/>
|
|
|
<enumeration value="Chechen"/>
|
|
|
<enumeration value="Chichewa"/>
|
|
|
<enumeration value="Chinese"/>
|
|
|
<enumeration value="Chuvash"/>
|
|
|
<enumeration value="Cornish"/>
|
|
|
<enumeration value="Corsican"/>
|
|
|
<enumeration value="Cree"/>
|
|
|
<enumeration value="Croatian"/>
|
|
|
<enumeration value="Czech"/>
|
|
|
<enumeration value="Danish"/>
|
|
|
<enumeration value="Divehi"/>
|
|
|
<enumeration value="Dutch"/>
|
|
|
<enumeration value="Dzongkha"/>
|
|
|
<enumeration value="English"/>
|
|
|
<enumeration value="Esperanto"/>
|
|
|
<enumeration value="Estonian"/>
|
|
|
<enumeration value="Ewe"/>
|
|
|
<enumeration value="Faroese"/>
|
|
|
<enumeration value="Fijian"/>
|
|
|
<enumeration value="Finnish"/>
|
|
|
<enumeration value="French"/>
|
|
|
<enumeration value="Fula"/>
|
|
|
<enumeration value="Gaelic"/>
|
|
|
<enumeration value="Galician"/>
|
|
|
<enumeration value="Ganda"/>
|
|
|
<enumeration value="Georgian"/>
|
|
|
<enumeration value="German"/>
|
|
|
<enumeration value="Greek"/>
|
|
|
<enumeration value="Guaraní"/>
|
|
|
<enumeration value="Gujarati"/>
|
|
|
<enumeration value="Haitian"/>
|
|
|
<enumeration value="Hausa"/>
|
|
|
<enumeration value="Hebrew"/>
|
|
|
<enumeration value="Herero"/>
|
|
|
<enumeration value="Hindi"/>
|
|
|
<enumeration value="Hiri Motu"/>
|
|
|
<enumeration value="Hungarian"/>
|
|
|
<enumeration value="Icelandic"/>
|
|
|
<enumeration value="Ido"/>
|
|
|
<enumeration value="Igbo"/>
|
|
|
<enumeration value="Indonesian"/>
|
|
|
<enumeration value="Interlingua"/>
|
|
|
<enumeration value="Interlingue"/>
|
|
|
<enumeration value="Inuktitut"/>
|
|
|
<enumeration value="Inupiaq"/>
|
|
|
<enumeration value="Irish"/>
|
|
|
<enumeration value="Italian"/>
|
|
|
<enumeration value="Japanese"/>
|
|
|
<enumeration value="Javanese"/>
|
|
|
<enumeration value="Kalaallisut"/>
|
|
|
<enumeration value="Kannada"/>
|
|
|
<enumeration value="Kanuri"/>
|
|
|
<enumeration value="Kashmiri"/>
|
|
|
<enumeration value="Kazakh"/>
|
|
|
<enumeration value="Khmer"/>
|
|
|
<enumeration value="Kikuyu"/>
|
|
|
<enumeration value="Kinyarwanda"/>
|
|
|
<enumeration value="Kirundi"/>
|
|
|
<enumeration value="Komi"/>
|
|
|
<enumeration value="Kongo"/>
|
|
|
<enumeration value="Korean"/>
|
|
|
<enumeration value="Kurdish"/>
|
|
|
<enumeration value="Kwanyama"/>
|
|
|
<enumeration value="Kyrgyz"/>
|
|
|
<enumeration value="Lao"/>
|
|
|
<enumeration value="Latin"/>
|
|
|
<enumeration value="Latvian"/>
|
|
|
<enumeration value="Limburgish"/>
|
|
|
<enumeration value="Lingala"/>
|
|
|
<enumeration value="Lithuanian"/>
|
|
|
<enumeration value="Luba-Katanga"/>
|
|
|
<enumeration value="Luxembourgish"/>
|
|
|
<enumeration value="Macedonian"/>
|
|
|
<enumeration value="Malagasy"/>
|
|
|
<enumeration value="Malay"/>
|
|
|
<enumeration value="Malayalam"/>
|
|
|
<enumeration value="Maltese"/>
|
|
|
<enumeration value="Manx"/>
|
|
|
<enumeration value="Māori"/>
|
|
|
<enumeration value="Marathi"/>
|
|
|
<enumeration value="Marshallese"/>
|
|
|
<enumeration value="Mongolian"/>
|
|
|
<enumeration value="Nauru"/>
|
|
|
<enumeration value="Navajo"/>
|
|
|
<enumeration value="Ndonga"/>
|
|
|
<enumeration value="Nepali"/>
|
|
|
<enumeration value="North Ndebele"/>
|
|
|
<enumeration value="Northern Sami"/>
|
|
|
<enumeration value="Norwegian"/>
|
|
|
<enumeration value="Norwegian Bokmål"/>
|
|
|
<enumeration value="Norwegian Nynorsk"/>
|
|
|
<enumeration value="Nuosu"/>
|
|
|
<enumeration value="Occitan"/>
|
|
|
<enumeration value="Ojibwe"/>
|
|
|
<enumeration value="Old Church Slavonic"/>
|
|
|
<enumeration value="Oriya"/>
|
|
|
<enumeration value="Oromo"/>
|
|
|
<enumeration value="Ossetian"/>
|
|
|
<enumeration value="Pāli"/>
|
|
|
<enumeration value="Panjabi"/>
|
|
|
<enumeration value="Pashto"/>
|
|
|
<enumeration value="Persian"/>
|
|
|
<enumeration value="Polish"/>
|
|
|
<enumeration value="Portuguese"/>
|
|
|
<enumeration value="Punjabi"/>
|
|
|
<enumeration value="Quechua"/>
|
|
|
<enumeration value="Romanian"/>
|
|
|
<enumeration value="Romansh"/>
|
|
|
<enumeration value="Russian"/>
|
|
|
<enumeration value="Samoan"/>
|
|
|
<enumeration value="Sango"/>
|
|
|
<enumeration value="Sanskrit"/>
|
|
|
<enumeration value="Sardinian"/>
|
|
|
<enumeration value="Serbian"/>
|
|
|
<enumeration value="Shona"/>
|
|
|
<enumeration value="Sindhi"/>
|
|
|
<enumeration value="Sinhala"/>
|
|
|
<enumeration value="Slovak"/>
|
|
|
<enumeration value="Slovene"/>
|
|
|
<enumeration value="Somali"/>
|
|
|
<enumeration value="South Ndebele"/>
|
|
|
<enumeration value="Southern Sotho"/>
|
|
|
<enumeration value="Spanish"/>
|
|
|
<enumeration value="Sundanese"/>
|
|
|
<enumeration value="Swahili"/>
|
|
|
<enumeration value="Swati"/>
|
|
|
<enumeration value="Swedish"/>
|
|
|
<enumeration value="Tagalog"/>
|
|
|
<enumeration value="Tahitian"/>
|
|
|
<enumeration value="Tajik"/>
|
|
|
<enumeration value="Tamil"/>
|
|
|
<enumeration value="Tatar"/>
|
|
|
<enumeration value="Telugu"/>
|
|
|
<enumeration value="Thai"/>
|
|
|
<enumeration value="Tibetan"/>
|
|
|
<enumeration value="Tigrinya"/>
|
|
|
<enumeration value="Tonga"/>
|
|
|
<enumeration value="Tsonga"/>
|
|
|
<enumeration value="Tswana"/>
|
|
|
<enumeration value="Turkish"/>
|
|
|
<enumeration value="Turkmen"/>
|
|
|
<enumeration value="Twi"/>
|
|
|
<enumeration value="Uighur"/>
|
|
|
<enumeration value="Ukrainian"/>
|
|
|
<enumeration value="Urdu"/>
|
|
|
<enumeration value="Uzbek"/>
|
|
|
<enumeration value="Venda"/>
|
|
|
<enumeration value="Vietnamese"/>
|
|
|
<enumeration value="Volapük"/>
|
|
|
<enumeration value="Walloon"/>
|
|
|
<enumeration value="Welsh"/>
|
|
|
<enumeration value="Western Frisian"/>
|
|
|
<enumeration value="Wolof"/>
|
|
|
<enumeration value="Xhosa"/>
|
|
|
<enumeration value="Yiddish"/>
|
|
|
<enumeration value="Yoruba"/>
|
|
|
<enumeration value="Zhuang"/>
|
|
|
<enumeration value="Zulu"/>
|
|
|
<enumeration value="other"/>
|
|
|
</restriction>
|
|
|
</simpleType>
|
|
|
<simpleType name="ScriptSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>iso15924 2016-07-14</documentation>
|
|
|
</annotation>
|
|
|
<restriction base="string">
|
|
|
<enumeration value="Adlm - Adlam"/>
|
|
|
<enumeration value="Afak - Afaka"/>
|
|
|
<enumeration value="Aghb - Caucasian Albanian"/>
|
|
|
<enumeration value="Ahom - Ahom, Tai Ahom"/>
|
|
|
<enumeration value="Arab - Arabic"/>
|
|
|
<enumeration value="Aran - Arabic (Nastaliq variant)"/>
|
|
|
<enumeration value="Armi - Imperial Aramaic"/>
|
|
|
<enumeration value="Armn - Armenian"/>
|
|
|
<enumeration value="Avst - Avestan"/>
|
|
|
<enumeration value="Bali - Balinese"/>
|
|
|
<enumeration value="Bamu - Bamum"/>
|
|
|
<enumeration value="Bass - Bassa Vah"/>
|
|
|
<enumeration value="Batk - Batak"/>
|
|
|
<enumeration value="Beng - Bengali"/>
|
|
|
<enumeration value="Bhks - Bhaiksuki"/>
|
|
|
<enumeration value="Blis - Blissymbols"/>
|
|
|
<enumeration value="Bopo - Bopomofo"/>
|
|
|
<enumeration value="Brah - Brahmi"/>
|
|
|
<enumeration value="Brai - Braille"/>
|
|
|
<enumeration value="Bugi - Buginese"/>
|
|
|
<enumeration value="Buhd - Buhid"/>
|
|
|
<enumeration value="Cakm - Chakma"/>
|
|
|
<enumeration value="Cans - Unified Canadian Aboriginal Syllabics"/>
|
|
|
<enumeration value="Cari - Carian"/>
|
|
|
<enumeration value="Cham - Cham"/>
|
|
|
<enumeration value="Cher - Cherokee"/>
|
|
|
<enumeration value="Cirt - Cirth"/>
|
|
|
<enumeration value="Copt - Coptic"/>
|
|
|
<enumeration value="Cprt - Cypriot"/>
|
|
|
<enumeration value="Cyrl - Cyrillic"/>
|
|
|
<enumeration value="Cyrs - Cyrillic (Old Church Slavonic variant)"/>
|
|
|
<enumeration value="Deva - Devanagari (Nagari)"/>
|
|
|
<enumeration value="Dsrt - Deseret (Mormon)"/>
|
|
|
<enumeration value="Dupl - Duployan shorthand, Duployan stenography"/>
|
|
|
<enumeration value="Egyd - Egyptian demotic"/>
|
|
|
<enumeration value="Egyh - Egyptian hieratic"/>
|
|
|
<enumeration value="Egyp - Egyptian hieroglyphs"/>
|
|
|
<enumeration value="Elba - Elbasan"/>
|
|
|
<enumeration value="Ethi - Ethiopic"/>
|
|
|
<enumeration value="Geok - Khutsuri (Asomtavruli and Nuskhuri)"/>
|
|
|
<enumeration value="Geor - Georgian (Mkhedruli)"/>
|
|
|
<enumeration value="Glag - Glagolitic"/>
|
|
|
<enumeration value="Goth - Gothic"/>
|
|
|
<enumeration value="Gran - Grantha"/>
|
|
|
<enumeration value="Grek - Greek"/>
|
|
|
<enumeration value="Gujr - Gujarati"/>
|
|
|
<enumeration value="Guru - Gurmukhi"/>
|
|
|
<enumeration value="Hanb - Han with Bopomofo"/>
|
|
|
<enumeration value="Hang - Hangul"/>
|
|
|
<enumeration value="Hani - Han (Hanzi, Kanji, Hanja)"/>
|
|
|
<enumeration value="Hano - Hanunoo (Hanunóo)"/>
|
|
|
<enumeration value="Hans - Han (Simplified variant)"/>
|
|
|
<enumeration value="Hant - Han (Traditional variant)"/>
|
|
|
<enumeration value="Hatr - Hatran"/>
|
|
|
<enumeration value="Hebr - Hebrew"/>
|
|
|
<enumeration value="Hira - Hiragana"/>
|
|
|
<enumeration value="Hluw - Anatolian Hieroglyphs"/>
|
|
|
<enumeration value="Hmng - Pahawh Hmong"/>
|
|
|
<enumeration value="Hrkt - Japanese syllabaries"/>
|
|
|
<enumeration value="Hung - Old Hungarian (Hungarian Runic)"/>
|
|
|
<enumeration value="Inds - Indus (Harappan)"/>
|
|
|
<enumeration value="Ital - Old Italic (Etruscan, Oscan etc.)"/>
|
|
|
<enumeration value="Jamo - Jamo"/>
|
|
|
<enumeration value="Java - Javanese"/>
|
|
|
<enumeration value="Jpan - Japanese"/>
|
|
|
<enumeration value="Jurc - Jurchen"/>
|
|
|
<enumeration value="Kali - Kayah Li"/>
|
|
|
<enumeration value="Kana - Katakana"/>
|
|
|
<enumeration value="Khar - Kharoshthi"/>
|
|
|
<enumeration value="Khmr - Khmer"/>
|
|
|
<enumeration value="Khoj - Khojki"/>
|
|
|
<enumeration value="Kitl - Khitan large script"/>
|
|
|
<enumeration value="Kits - Khitan small script"/>
|
|
|
<enumeration value="Knda - Kannada"/>
|
|
|
<enumeration value="Kore - Korean (alias for Hangul + Han)"/>
|
|
|
<enumeration value="Kpel - Kpelle"/>
|
|
|
<enumeration value="Kthi - Kaithi"/>
|
|
|
<enumeration value="Lana - Tai Tham (Lanna)"/>
|
|
|
<enumeration value="Laoo - Lao"/>
|
|
|
<enumeration value="Latf - Latin (Fraktur variant)"/>
|
|
|
<enumeration value="Latg - Latin (Gaelic variant)"/>
|
|
|
<enumeration value="Latn - Latin"/>
|
|
|
<enumeration value="Leke - Leke"/>
|
|
|
<enumeration value="Lepc - Lepcha (Róng)"/>
|
|
|
<enumeration value="Limb - Limbu"/>
|
|
|
<enumeration value="Lina - Linear A"/>
|
|
|
<enumeration value="Linb - Linear B"/>
|
|
|
<enumeration value="Lisu - Lisu (Fraser)"/>
|
|
|
<enumeration value="Loma - Loma"/>
|
|
|
<enumeration value="Lyci - Lycian"/>
|
|
|
<enumeration value="Lydi - Lydian"/>
|
|
|
<enumeration value="Mahj - Mahajani"/>
|
|
|
<enumeration value="Mand - Mandaic, Mandaean"/>
|
|
|
<enumeration value="Mani - Manichaean"/>
|
|
|
<enumeration value="Marc - Marchen"/>
|
|
|
<enumeration value="Maya - Mayan hieroglyphs"/>
|
|
|
<enumeration value="Mend - Mende Kikakui"/>
|
|
|
<enumeration value="Merc - Meroitic Cursive"/>
|
|
|
<enumeration value="Mero - Meroitic Hieroglyphs"/>
|
|
|
<enumeration value="Mlym - Malayalam"/>
|
|
|
<enumeration value="Modi - Modi, Moḍī"/>
|
|
|
<enumeration value="Mong - Mongolian"/>
|
|
|
<enumeration value="Moon - Moon (Moon code, Moon script, Moon type)"/>
|
|
|
<enumeration value="Mroo - Mro, Mru"/>
|
|
|
<enumeration value="Mtei - Meitei Mayek (Meithei, Meetei)"/>
|
|
|
<enumeration value="Mult - Multani"/>
|
|
|
<enumeration value="Mymr - Myanmar (Burmese)"/>
|
|
|
<enumeration value="Narb - Old North Arabian (Ancient North Arabian)"/>
|
|
|
<enumeration value="Nbat - Nabataean"/>
|
|
|
<enumeration value="Newa - Newa, Newar, Newari"/>
|
|
|
<enumeration value="Nkgb - Nakhi Geba"/>
|
|
|
<enumeration value="Nkoo - N’Ko"/>
|
|
|
<enumeration value="Nshu - Nüshu"/>
|
|
|
<enumeration value="Ogam - Ogham"/>
|
|
|
<enumeration value="Olck - Ol Chiki (Ol Cemet’, Ol, Santali)"/>
|
|
|
<enumeration value="Orkh - Old Turkic, Orkhon Runic"/>
|
|
|
<enumeration value="Orya - Oriya"/>
|
|
|
<enumeration value="Osge - Osage"/>
|
|
|
<enumeration value="Osma - Osmanya"/>
|
|
|
<enumeration value="Palm - Palmyrene"/>
|
|
|
<enumeration value="Pauc - Pau Cin Hau"/>
|
|
|
<enumeration value="Perm - Old Permic"/>
|
|
|
<enumeration value="Phag - Phags-pa"/>
|
|
|
<enumeration value="Phli - Inscriptional Pahlavi"/>
|
|
|
<enumeration value="Phlp - Psalter Pahlavi"/>
|
|
|
<enumeration value="Phlv - Book Pahlavi"/>
|
|
|
<enumeration value="Phnx - Phoenician"/>
|
|
|
<enumeration value="Piqd - Klingon (KLI pIqaD)"/>
|
|
|
<enumeration value="Plrd - Miao (Pollard)"/>
|
|
|
<enumeration value="Prti - Inscriptional Parthian"/>
|
|
|
<enumeration value="Rjng - Rejang (Redjang, Kaganga)"/>
|
|
|
<enumeration value="Roro - Rongorongo"/>
|
|
|
<enumeration value="Runr - Runic"/>
|
|
|
<enumeration value="Samr - Samaritan"/>
|
|
|
<enumeration value="Sara - Sarati"/>
|
|
|
<enumeration value="Sarb - Old South Arabian"/>
|
|
|
<enumeration value="Saur - Saurashtra"/>
|
|
|
<enumeration value="Sgnw - SignWriting"/>
|
|
|
<enumeration value="Shaw - Shavian (Shaw)"/>
|
|
|
<enumeration value="Shrd - Sharada, Śāradā"/>
|
|
|
<enumeration value="Sidd - Siddham"/>
|
|
|
<enumeration value="Sind - Khudawadi, Sindhi"/>
|
|
|
<enumeration value="Sinh - Sinhala"/>
|
|
|
<enumeration value="Sora - Sora Sompeng"/>
|
|
|
<enumeration value="Sund - Sundanese"/>
|
|
|
<enumeration value="Sylo - Syloti Nagri"/>
|
|
|
<enumeration value="Syrc - Syriac"/>
|
|
|
<enumeration value="Syre - Syriac (Estrangelo variant)"/>
|
|
|
<enumeration value="Syrj - Syriac (Western variant)"/>
|
|
|
<enumeration value="Syrn - Syriac (Eastern variant)"/>
|
|
|
<enumeration value="Tagb - Tagbanwa"/>
|
|
|
<enumeration value="Takr - Takri"/>
|
|
|
<enumeration value="Tale - Tai Le"/>
|
|
|
<enumeration value="Talu - New Tai Lue"/>
|
|
|
<enumeration value="Taml - Tamil"/>
|
|
|
<enumeration value="Tang - Tangut"/>
|
|
|
<enumeration value="Tavt - Tai Viet"/>
|
|
|
<enumeration value="Telu - Telugu"/>
|
|
|
<enumeration value="Teng - Tengwar"/>
|
|
|
<enumeration value="Tfng - Tifinagh (Berber)"/>
|
|
|
<enumeration value="Tglg - Tagalog (Baybayin, Alibata)"/>
|
|
|
<enumeration value="Thaa - Thaana"/>
|
|
|
<enumeration value="Thai - Thai"/>
|
|
|
<enumeration value="Tibt - Tibetan"/>
|
|
|
<enumeration value="Tirh - Tirhuta"/>
|
|
|
<enumeration value="Ugar - Ugaritic"/>
|
|
|
<enumeration value="Vaii - Vai"/>
|
|
|
<enumeration value="Visp - Visible Speech"/>
|
|
|
<enumeration value="Wara - Warang Citi (Varang Kshiti)"/>
|
|
|
<enumeration value="Wole - Woleai"/>
|
|
|
<enumeration value="Xpeo - Old Persian"/>
|
|
|
<enumeration value="Xsux - Cuneiform, Sumero-Akkadian"/>
|
|
|
<enumeration value="Yiii - Yi"/>
|
|
|
<enumeration value="Zinh - Code for inherited script"/>
|
|
|
<enumeration value="Zmth - Mathematical notation"/>
|
|
|
<enumeration value="Zsye - Symbols (Emoji variant)"/>
|
|
|
<enumeration value="Zsym - Symbols"/>
|
|
|
<enumeration value="Zxxx - Code for unwritten documents"/>
|
|
|
<enumeration value="Zyyy - Code for undetermined script"/>
|
|
|
<enumeration value="Zzzz - Code for uncoded script"/>
|
|
|
<enumeration value="other"/>
|
|
|
</restriction>
|
|
|
</simpleType>
|
|
|
<simpleType name="ColourDepthSimpleType">
|
|
|
<restriction base="string">
|
|
|
<enumeration value="bilevel"/>
|
|
|
<enumeration value="greyscale"/>
|
|
|
<enumeration value="colour"/>
|
|
|
<enumeration value="other"/>
|
|
|
</restriction>
|
|
|
</simpleType>
|
|
|
<simpleType name="GraphicsTypeSimpleType">
|
|
|
<restriction base="string">
|
|
|
<enumeration value="logo"/>
|
|
|
<enumeration value="letterhead"/>
|
|
|
<enumeration value="decoration"/>
|
|
|
<enumeration value="frame"/>
|
|
|
<enumeration value="handwritten-annotation"/>
|
|
|
<enumeration value="stamp"/>
|
|
|
<enumeration value="signature"/>
|
|
|
<enumeration value="barcode"/>
|
|
|
<enumeration value="paper-grow"/>
|
|
|
<enumeration value="punch-hole"/>
|
|
|
<enumeration value="other"/>
|
|
|
</restriction>
|
|
|
</simpleType>
|
|
|
<simpleType name="ChartTypeSimpleType">
|
|
|
<restriction base="string">
|
|
|
<enumeration value="bar"/>
|
|
|
<enumeration value="line"/>
|
|
|
<enumeration value="pie"/>
|
|
|
<enumeration value="scatter"/>
|
|
|
<enumeration value="surface"/>
|
|
|
<enumeration value="other"/>
|
|
|
</restriction>
|
|
|
</simpleType>
|
|
|
<complexType name="LayersType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Can be used to express the z-index of overlapping
|
|
|
regions. An element with a greater z-index is always in
|
|
|
front of another element with lower z-index.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
<sequence minOccurs="1" maxOccurs="unbounded">
|
|
|
<element name="Layer" type="pc:LayerType"></element>
|
|
|
</sequence>
|
|
|
</complexType>
|
|
|
|
|
|
<complexType name="LayerType">
|
|
|
<sequence minOccurs="1" maxOccurs="unbounded">
|
|
|
<element name="RegionRef" type="pc:RegionRefType"/>
|
|
|
</sequence>
|
|
|
<attribute name="id" type="ID" use="required"/>
|
|
|
<attribute name="zIndex" type="int" use="required"/>
|
|
|
<attribute name="caption" type="string"/>
|
|
|
</complexType>
|
|
|
|
|
|
<complexType name="BaselineType">
|
|
|
<attribute name="points" type="pc:PointsType"
|
|
|
use="required">
|
|
|
</attribute>
|
|
|
<attribute name="conf" type="pc:ConfSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>Confidence value (between 0 and 1)</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
</complexType>
|
|
|
|
|
|
<simpleType name="PointsType">
|
|
|
<annotation>
|
|
|
<documentation>Point list with format "x1,y1 x2,y2 ..."</documentation>
|
|
|
</annotation>
|
|
|
<restriction base="string">
|
|
|
<pattern value="([0-9]+,[0-9]+ )+([0-9]+,[0-9]+)"></pattern>
|
|
|
</restriction>
|
|
|
</simpleType>
|
|
|
|
|
|
<complexType name="RelationsType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Container for one-to-one relations between layout
|
|
|
objects (for example: DropCap - paragraph, caption -
|
|
|
image).
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
<sequence minOccurs="1" maxOccurs="unbounded">
|
|
|
<element name="Relation" type="pc:RelationType"/>
|
|
|
</sequence>
|
|
|
</complexType>
|
|
|
|
|
|
<complexType name="RelationType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
One-to-one relation between to layout object. Use 'link'
|
|
|
for loose relations and 'join' for strong relations
|
|
|
(where something is fragmented for instance).
|
|
|
|
|
|
Examples for 'link': caption - image floating -
|
|
|
paragraph paragraph - paragraph (when a paragraph is
|
|
|
split across columns and the last word of the first
|
|
|
paragraph DOES NOT continue in the second paragraph)
|
|
|
drop-cap - paragraph (when the drop-cap is a whole word)
|
|
|
|
|
|
Examples for 'join': word - word (separated word at the
|
|
|
end of a line) drop-cap - paragraph (when the drop-cap
|
|
|
is not a whole word) paragraph - paragraph (when a
|
|
|
pragraph is split across columns and the last word of
|
|
|
the first paragraph DOES continue in the second
|
|
|
paragraph)
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
<sequence>
|
|
|
<element name="Labels" type="pc:LabelsType"
|
|
|
minOccurs="0" maxOccurs="unbounded">
|
|
|
<annotation>
|
|
|
<documentation>Semantic labels / tags</documentation>
|
|
|
</annotation>
|
|
|
</element>
|
|
|
<element name="SourceRegionRef" type="pc:RegionRefType"
|
|
|
minOccurs="1" maxOccurs="1">
|
|
|
</element>
|
|
|
<element name="TargetRegionRef" type="pc:RegionRefType"
|
|
|
minOccurs="1" maxOccurs="1">
|
|
|
</element>
|
|
|
</sequence>
|
|
|
<attribute name="id" type="ID" use="required"/>
|
|
|
<attribute name="type">
|
|
|
<simpleType>
|
|
|
<restriction base="string">
|
|
|
<enumeration value="link"/>
|
|
|
<enumeration value="join"/>
|
|
|
</restriction>
|
|
|
</simpleType>
|
|
|
</attribute>
|
|
|
<attribute name="custom" type="string">
|
|
|
<annotation>
|
|
|
<documentation>For generic use</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="comments" type="string"/>
|
|
|
</complexType>
|
|
|
|
|
|
<simpleType name="ProductionSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>Text production type</documentation>
|
|
|
</annotation>
|
|
|
<restriction base="string">
|
|
|
<enumeration value="printed"/>
|
|
|
<enumeration value="typewritten"/>
|
|
|
<enumeration value="handwritten-cursive"/>
|
|
|
<enumeration value="handwritten-printscript"/>
|
|
|
<enumeration value="medieval-manuscript"/>
|
|
|
<enumeration value="other"/>
|
|
|
</restriction>
|
|
|
</simpleType>
|
|
|
|
|
|
<complexType name="TextStyleType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Monospace (fixed-pitch, non-proportional) or
|
|
|
proportional font.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
<attribute name="fontFamily" type="string">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
For instance: Arial, Times New Roman.
|
|
|
Add more information if necessary
|
|
|
(e.g. blackletter, antiqua).
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="serif" type="boolean">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Serif or sans-serif typeface.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="monospace" type="boolean"/>
|
|
|
<attribute name="fontSize" type="float">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The size of the characters in points.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="xHeight" type="integer">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The x-height or corpus size refers to the distance
|
|
|
between the baseline and the mean line of
|
|
|
lower-case letters in a typeface.
|
|
|
The unit is assumed to be pixels.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="kerning" type="int">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
The degree of space (in points) between
|
|
|
the characters in a string of text.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="textColour" type="pc:ColourSimpleType"/>
|
|
|
<attribute name="textColourRgb" type="integer">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Text colour in RGB encoded format
|
|
|
(red value) + (256 x green value) + (65536 x blue value).
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="bgColour" type="pc:ColourSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>Background colour</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="bgColourRgb" type="integer">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Background colour in RGB encoded format
|
|
|
(red value) + (256 x green value) + (65536 x blue value).
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="reverseVideo" type="boolean">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Specifies whether the colour of the text appears
|
|
|
reversed against a background colour.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="bold" type="boolean"/>
|
|
|
<attribute name="italic" type="boolean"/>
|
|
|
<attribute name="underlined" type="boolean"/>
|
|
|
<attribute name="underlineStyle"
|
|
|
type="pc:underlineStyleType" use="optional">
|
|
|
<annotation>
|
|
|
<documentation>Line style details if "underlined" is TRUE
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="doubleUnderlined" type="boolean"/>
|
|
|
<attribute name="subscript" type="boolean"/>
|
|
|
<attribute name="superscript" type="boolean"/>
|
|
|
<attribute name="strikethrough" type="boolean"/>
|
|
|
<attribute name="smallCaps" type="boolean"/>
|
|
|
<attribute name="letterSpaced" type="boolean"/>
|
|
|
</complexType>
|
|
|
|
|
|
<complexType name="RegionType" abstract="true">
|
|
|
<sequence>
|
|
|
<element name="AlternativeImage" type="pc:AlternativeImageType"
|
|
|
minOccurs="0" maxOccurs="unbounded">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Alternative region images
|
|
|
(e.g. black-and-white).
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</element>
|
|
|
<element name="Coords" type="pc:CoordsType"/>
|
|
|
<element name="UserDefined" type="pc:UserDefinedType"
|
|
|
minOccurs="0" maxOccurs="1">
|
|
|
</element>
|
|
|
<element name="Labels" type="pc:LabelsType"
|
|
|
minOccurs="0" maxOccurs="unbounded">
|
|
|
<annotation>
|
|
|
<documentation>Semantic labels / tags</documentation>
|
|
|
</annotation>
|
|
|
</element>
|
|
|
<element name="Roles" type="pc:RolesType"
|
|
|
minOccurs="0" maxOccurs="1">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Roles the region takes
|
|
|
(e.g. in context of a parent region).
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</element>
|
|
|
<choice minOccurs="0" maxOccurs="unbounded">
|
|
|
<element name="TextRegion" type="pc:TextRegionType"/>
|
|
|
<element name="ImageRegion" type="pc:ImageRegionType"/>
|
|
|
<element name="LineDrawingRegion" type="pc:LineDrawingRegionType"/>
|
|
|
<element name="GraphicRegion" type="pc:GraphicRegionType"/>
|
|
|
<element name="TableRegion" type="pc:TableRegionType"/>
|
|
|
<element name="ChartRegion" type="pc:ChartRegionType"/>
|
|
|
<element name="SeparatorRegion" type="pc:SeparatorRegionType"/>
|
|
|
<element name="MathsRegion" type="pc:MathsRegionType"/>
|
|
|
<element name="ChemRegion" type="pc:ChemRegionType"/>
|
|
|
<element name="MusicRegion" type="pc:MusicRegionType"/>
|
|
|
<element name="AdvertRegion" type="pc:AdvertRegionType"/>
|
|
|
<element name="NoiseRegion" type="pc:NoiseRegionType"/>
|
|
|
<element name="UnknownRegion" type="pc:UnknownRegionType"/>
|
|
|
<element name="CustomRegion" type="pc:CustomRegionType"/>
|
|
|
</choice>
|
|
|
</sequence>
|
|
|
<attribute name="id" type="ID" use="required"/>
|
|
|
<attribute name="custom" type="string">
|
|
|
<annotation>
|
|
|
<documentation>For generic use</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="comments" type="string"/>
|
|
|
<attribute name="continuation" type="boolean">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Is this region a continuation of another region
|
|
|
(in previous column or page, for example)?
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
</complexType>
|
|
|
|
|
|
<complexType name="AlternativeImageType">
|
|
|
<attribute name="filename" type="string" use="required"/>
|
|
|
<attribute name="comments" type="string"/>
|
|
|
<attribute name="conf" type="pc:ConfSimpleType">
|
|
|
<annotation>
|
|
|
<documentation>Confidence value (between 0 and 1)</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
</complexType>
|
|
|
|
|
|
<simpleType name="AlignSimpleType">
|
|
|
<restriction base="string">
|
|
|
<enumeration value="left"/>
|
|
|
<enumeration value="centre"/>
|
|
|
<enumeration value="right"/>
|
|
|
<enumeration value="justify"/>
|
|
|
</restriction>
|
|
|
</simpleType>
|
|
|
<simpleType name="GroupTypeSimpleType">
|
|
|
<restriction base="string">
|
|
|
<enumeration value="paragraph"/>
|
|
|
<enumeration value="list"/>
|
|
|
<enumeration value="list-item"/>
|
|
|
<enumeration value="figure"/>
|
|
|
<enumeration value="article"/>
|
|
|
<enumeration value="div"/>
|
|
|
<enumeration value="other"/>
|
|
|
</restriction>
|
|
|
</simpleType>
|
|
|
<simpleType name="TextDataTypeSimpleType">
|
|
|
<restriction base="string">
|
|
|
<enumeration value="xsd:decimal">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Examples:
|
|
|
"123.456", "+1234.456",
|
|
|
"-1234.456", "-.456", "-456"
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</enumeration>
|
|
|
<enumeration value="xsd:float">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Examples:
|
|
|
"123.456", "+1234.456", "-1.2344e56",
|
|
|
"-.45E-6", "INF", "-INF", "NaN"
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</enumeration>
|
|
|
<enumeration value="xsd:integer">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Examples:
|
|
|
"123456", "+00000012", "-1", "-456"
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</enumeration>
|
|
|
<enumeration value="xsd:boolean">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Examples: "true", "false", "1", "0"
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</enumeration>
|
|
|
<enumeration value="xsd:date">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Examples:
|
|
|
"2001-10-26", "2001-10-26+02:00",
|
|
|
"2001-10-26Z", "2001-10-26+00:00",
|
|
|
"-2001-10-26", "-20000-04-01"
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</enumeration>
|
|
|
<enumeration value="xsd:time">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Examples:
|
|
|
"21:32:52", "21:32:52+02:00", "19:32:52Z",
|
|
|
"19:32:52+00:00", "21:32:52.12679"
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</enumeration>
|
|
|
<enumeration value="xsd:dateTime">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Examples:
|
|
|
"2001-10-26T21:32:52", "2001-10-26T21:32:52+02:00",
|
|
|
"2001-10-26T19:32:52Z", "2001-10-26T19:32:52+00:00",
|
|
|
"-2001-10-26T21:32:52", "2001-10-26T21:32:52.12679"
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</enumeration>
|
|
|
<enumeration value="xsd:string">
|
|
|
<annotation>
|
|
|
<documentation>Generic text string</documentation>
|
|
|
</annotation>
|
|
|
</enumeration>
|
|
|
<enumeration value="other">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
An XSD type that is not listed or a custom type
|
|
|
(use dataTypeDetails attribute).
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</enumeration>
|
|
|
</restriction>
|
|
|
</simpleType>
|
|
|
|
|
|
<!-- <simpleType name="TextMergeRuleSimpleType">
|
|
|
<restriction base="string">
|
|
|
<enumeration value="no-whitespace">
|
|
|
<annotation>
|
|
|
<documentation>Concatenate text without adding anything</documentation>
|
|
|
</annotation></enumeration>
|
|
|
<enumeration value="add-space">
|
|
|
<annotation>
|
|
|
<documentation>Adds a single space character between the text two parts</documentation>
|
|
|
</annotation></enumeration>
|
|
|
<enumeration value="add-tab">
|
|
|
<annotation>
|
|
|
<documentation>Adds a single tab character between the text two parts</documentation>
|
|
|
</annotation></enumeration>
|
|
|
<enumeration value="add-line-break">
|
|
|
<annotation>
|
|
|
<documentation>Adds a single line break character between the text two parts</documentation>
|
|
|
</annotation></enumeration>
|
|
|
<enumeration value="add-custom">
|
|
|
<annotation>
|
|
|
<documentation>Adds a user-defined string between the two text parts (specify in merge rule data)</documentation>
|
|
|
</annotation></enumeration>
|
|
|
<enumeration value="remove-last">
|
|
|
<annotation>
|
|
|
<documentation>Removes the last character of the first text part (e.g. remove hyphen)</documentation>
|
|
|
</annotation></enumeration>
|
|
|
<enumeration value="remove-multiple">
|
|
|
<annotation>
|
|
|
<documentation>Removes the last x characters from the first text part (specify number in merge rule data)</documentation>
|
|
|
</annotation></enumeration>
|
|
|
<enumeration value="remove-all">
|
|
|
<annotation>
|
|
|
<documentation>Ignores the first text part completely (e.g. for drop caps that are repeated in the following paragraph)</documentation>
|
|
|
</annotation></enumeration>
|
|
|
</restriction>
|
|
|
</simpleType> -->
|
|
|
|
|
|
<complexType name="GraphemesType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Container for graphemes, grapheme groups and
|
|
|
non-printing characters.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
<choice minOccurs="1" maxOccurs="unbounded">
|
|
|
<element name="Grapheme" type="pc:GraphemeType"/>
|
|
|
<element name="NonPrintingChar" type="pc:NonPrintingCharType"/>
|
|
|
<element name="GraphemeGroup" type="pc:GraphemeGroupType"/>
|
|
|
</choice>
|
|
|
</complexType>
|
|
|
<complexType name="GraphemeBaseType" abstract="true">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Base type for graphemes, grapheme groups and non-printing characters.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
<sequence>
|
|
|
<element name="TextEquiv" type="pc:TextEquivType"
|
|
|
minOccurs="0" maxOccurs="unbounded"/>
|
|
|
</sequence>
|
|
|
<attribute name="id" type="ID" use="required"/>
|
|
|
<attribute name="index" use="required">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Order index of grapheme, group, or non-printing character
|
|
|
within the parent container (graphemes or glyph or grapheme group).
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
<simpleType>
|
|
|
<restriction base="int">
|
|
|
<minInclusive value="0"></minInclusive>
|
|
|
</restriction>
|
|
|
</simpleType>
|
|
|
</attribute>
|
|
|
<attribute name="ligature" type="boolean"/>
|
|
|
<attribute name="charType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Type of character represented by the
|
|
|
grapheme, group, or non-printing character element.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
<simpleType>
|
|
|
<restriction base="string">
|
|
|
<enumeration value="base"/>
|
|
|
<enumeration value="combining"/>
|
|
|
</restriction>
|
|
|
</simpleType>
|
|
|
</attribute>
|
|
|
<attribute name="custom" type="string">
|
|
|
<annotation>
|
|
|
<documentation>For generic use</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="comments" type="string">
|
|
|
<annotation>
|
|
|
<documentation>For generic use</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
</complexType>
|
|
|
<complexType name="GraphemeType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Represents a sub-element of a glyph.
|
|
|
Smallest graphical unit that can be
|
|
|
assigned a Unicode code point.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
<complexContent>
|
|
|
<extension base="pc:GraphemeBaseType">
|
|
|
<sequence>
|
|
|
<element name="Coords" type="pc:CoordsType"/>
|
|
|
</sequence>
|
|
|
</extension>
|
|
|
</complexContent>
|
|
|
</complexType>
|
|
|
<complexType name="NonPrintingCharType">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
A glyph component without visual representation
|
|
|
but with Unicode code point.
|
|
|
Non-visual / non-printing / control character.
|
|
|
Part of grapheme container (of glyph) or grapheme sub group.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
<complexContent>
|
|
|
<extension base="pc:GraphemeBaseType">
|
|
|
</extension>
|
|
|
</complexContent>
|
|
|
</complexType>
|
|
|
<complexType name="GraphemeGroupType">
|
|
|
<complexContent>
|
|
|
<extension base="pc:GraphemeBaseType">
|
|
|
<choice minOccurs="0" maxOccurs="unbounded">
|
|
|
<element name="Grapheme" type="pc:GraphemeType"/>
|
|
|
<element name="NonPrintingChar" type="pc:NonPrintingCharType"/>
|
|
|
</choice>
|
|
|
</extension>
|
|
|
</complexContent>
|
|
|
</complexType>
|
|
|
|
|
|
<complexType name="UserDefinedType">
|
|
|
<annotation>
|
|
|
<documentation>Container for user-defined attributes</documentation>
|
|
|
</annotation>
|
|
|
<sequence>
|
|
|
<element name="UserAttribute" type="pc:UserAttributeType"
|
|
|
minOccurs="1" maxOccurs="unbounded">
|
|
|
</element>
|
|
|
</sequence>
|
|
|
</complexType>
|
|
|
<complexType name="UserAttributeType">
|
|
|
<annotation>
|
|
|
<documentation>Structured custom data defined by name, type and value.</documentation>
|
|
|
</annotation>
|
|
|
<attribute name="name" type="string"/>
|
|
|
<attribute name="description" type="string"/>
|
|
|
<attribute name="type">
|
|
|
<simpleType>
|
|
|
<restriction base="string">
|
|
|
<enumeration value="xsd:string"/>
|
|
|
<enumeration value="xsd:integer"/>
|
|
|
<enumeration value="xsd:boolean"/>
|
|
|
<enumeration value="xsd:float"/>
|
|
|
</restriction>
|
|
|
</simpleType>
|
|
|
</attribute>
|
|
|
<attribute name="value" type="string"/>
|
|
|
</complexType>
|
|
|
|
|
|
<complexType name="TableCellRoleType">
|
|
|
<attribute name="rowIndex" type="int" use="required">
|
|
|
<annotation>
|
|
|
<documentation>Cell position in table starting with row 0</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="columnIndex" type="int" use="required">
|
|
|
<annotation>
|
|
|
<documentation>Cell position in table starting with column 0</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="rowSpan" type="int">
|
|
|
<annotation>
|
|
|
<documentation>Number of rows the cell spans (optional; default is 1)</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="colSpan" type="int">
|
|
|
<annotation>
|
|
|
<documentation>Number of columns the cell spans (optional; default is 1)</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
<attribute name="header" type="boolean">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Is the cell a column or row header?
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</attribute>
|
|
|
</complexType>
|
|
|
<complexType name="RolesType">
|
|
|
<sequence>
|
|
|
<element name="TableCellRole" type="pc:TableCellRoleType"
|
|
|
minOccurs="0" maxOccurs="1">
|
|
|
<annotation>
|
|
|
<documentation>
|
|
|
Data for a region that takes on the role
|
|
|
of a table cell within a parent table region.
|
|
|
</documentation>
|
|
|
</annotation>
|
|
|
</element>
|
|
|
</sequence>
|
|
|
</complexType>
|
|
|
<simpleType name="underlineStyleType">
|
|
|
<restriction base="string">
|
|
|
<enumeration value="singleLine" />
|
|
|
<enumeration value="doubleLine" />
|
|
|
<enumeration value="other" />
|
|
|
</restriction>
|
|
|
</simpleType>
|
|
|
</schema>
|