KorAP_schemas/text.rng

<?xml version="1.0" encoding="UTF-8"?>
<grammar xmlns="http://relaxng.org/ns/structure/1.0" datatypeLibrary="http://www.w3.org/2001/XMLSchema-datatypes" ns="http://ids-mannheim.de/ns/KorAP">
  <!-- $Id$ -->
  <start>
    <element name="raw_text">
      <attribute name="docid">
        <data type="NCName"/>
      </attribute>
      <element name="metadata">
        <attribute name="file">
          <data type="anyURI"/>
        </attribute>
      </element>
      <element name="text">
        <optional>
          <attribute name="mimeType">
            <documentation xmlns="http://relaxng.org/ns/compatibility/annotations/1.0">By default,
              we assume "text/plain; charset=UTF-8", but we should be prepared for other
              encodings</documentation>
            <documentation xmlns="http://relaxng.org/ns/compatibility/annotations/1.0">Or maybe
              @charset would be enough here?</documentation>
            <data type="string"/>
          </attribute>
        </optional>        <text/>
      </element>
      <zeroOrMore>
        <element name="binary">
          <attribute name="mimeType">
            <data type="string"/>
          </attribute>
          <attribute name="file">
            <data type="anyURI"/>
          </attribute>
        </element>
      </zeroOrMore>
    </element>
  </start>
</grammar>