ParlaMint-teiCorpus.rnc

default namespace = "http://www.tei-c.org/ns/1.0"
namespace a = "http://relaxng.org/ns/compatibility/annotations/1.0"
namespace xi = "http://www.w3.org/2001/XInclude"

a:documentation [
  "RelaxNG XML schema for ParlaMint corpora: schema\x{a}" ~
  '  for "plain-text" root corpus file.'
]
include "ParlaMint-TEI.rnc" {
  
  ## Root element of a ParlaMint corpus root file.
  start =
    element teiCorpus {
      id.att >> a:documentation [ "Obligatory @xml:lang." ],
      attribute xml:lang { lang.val }
      >> a:documentation [ "Optional @corresp for MTed corpora." ],
      corresp.att,
      teiHeader,
      XInclude+
    }
  
  ## Metadata of the "plain-text" corpus.
  teiHeader =
    element teiHeader {
      global.atts, fileDesc, encodingDesc, profileDesc, revisionDesc?
    }
  
  ## The title statement.
  titleStmt =
    element titleStmt { title-header+, meeting*, respStmt+, funder+ }
  
  ## The encoding description.
  encodingDesc =
    element encodingDesc {
      global.atts,
      element projectDesc { paras }+,
      editorialDecl,
      tagsDecl,
      element classDecl { (taxonomy | XInclude)+ }
    }
  
  ## The profile description.
  profileDesc =
    element profileDesc {
      global.atts, settingDesc, textClass?, particDesc, langUsage
    }
}

## The editorial declaration.
editorialDecl =
  element editorialDecl {
    element correction { paras },
    element normalization { paras },
    element hyphenation { paras },
    element quotation { paras },
    element segmentation { paras }
  }

## The text class.
textClass =
  element textClass {
    element catRef {
      attribute scheme { xsd:anyURI }?,
      attribute target { anyURIs }
    }+
  }

## Metadata on the organisations (esp. political parties) and speakers.
##     There can be only one list of organisations and one of speakers, and they can be
##     XIncluded.
particDesc =
  element particDesc { (listOrg | XInclude), (listPerson | XInclude) }

## Used languages.
langUsage =
  element langUsage {
    element language {
      lang.att,
      attribute ident { lang.val },
      attribute usage { xsd:nonNegativeInteger }?,
      normalized-space.string
    }+
  }