ParlaMint-TEI.rnc

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

a:documentation [
  "RelaxNG XML schema for ParlaMint corpora: schema\x{a}" ~
  '  for "plain text" corpus component file.'
]
include "ParlaMint.rnc"

## Root element of a ParlaMint component file.
start =
  
  ## Obligatory @xml:id.
  element TEI {
    id.att >> a:documentation [ "Obligatory @xml:lang." ],
    attribute xml:lang { lang.val }
    >> a:documentation [
         "Obligatory @ana determining what scope this document has."
       ],
    attribute ana { anyURIs },
    teiHeader,
    text-body
  }
# METADATA

## Metadata of the component.
teiHeader =
  element teiHeader {
    global.atts, fileDesc, encodingDesc, profileDesc, revisionDesc?
  }
fileDesc =
  
  ## File description.
  element fileDesc {
    global.atts,
    titleStmt,
    editionStmt,
    extent,
    publicationStmt,
    sourceDesc
  }

## Title statement.
titleStmt =
  element titleStmt { title-header+, meeting+, respStmt*, funder* }

## Responsibility statement.
respStmt =
  element respStmt {
    (element persName {
       global.atts, ref.att, normalized-space.string
     }+,
     element resp { lang.att, normalized-space.string }+)+
  }

## Edition statement.
editionStmt =
  element editionStmt {
    element edition { normalized-space.string }
  }

## Extent of the document.
extent =
  element extent {
    element measure {
      attribute unit {
        # data type="NCName"/
        "speeches" | "words" | "tokens"
      },
      attribute quantity { xsd:integer },
      lang.att,
      text
    }+
  }

## Publications statement.
publicationStmt =
  element publicationStmt {
    publisher,
    idno,
    element availability {
      attribute status { "free" },
      [
        # data type="anyURI"/
        
        a:documentation [ "Fixed licence CC BY." ]
      ]
      element licence { "http://creativecommons.org/licenses/by/4.0/" },
      paras
    },
    date-simple
  }

## Source description.
sourceDesc =
  element sourceDesc {
    element bibl {
      title+,
      element edition { lang.att, normalized-space.string }*,
      publisher?,
      idno+,
      date*
    }+,
    element recordingStmt {
      element recording {
        attribute type { "audio" | "video" },
        attribute dur { xsd:duration }?,
        media+,
        respStmt*,
        time?,
        element equipment { abs }?
      }+
    }?
  }

## Encoding description.
encodingDesc =
  element encodingDesc {
    global.atts,
    element projectDesc { paras },
    tagsDecl
  }

## Profile description.
profileDesc = element profileDesc { global.atts, settingDesc }

## Revision description.
revisionDesc =
  element revisionDesc {
    global.atts,
    element change {
      attribute when { xsd:date },
      mixed { name-header* }
    }+
  }

## Tags declaration.
tagsDecl =
  element tagsDecl {
    element namespace {
      attribute name { xsd:anyURI },
      element tagUsage {
        attribute gi { xsd:Name },
        attribute occurs { xsd:nonNegativeInteger },
        text
      }+
    }
  }

## Setting description.
settingDesc =
  element settingDesc {
    element setting { name-header+, date }
  }

## A title in the teiHeader. Must have a @type.
title-header =
  element title {
    lang.att >> a:documentation [ "@type is here obligatory:" ],
    attribute type { "main" | "sub" },
    normalized-space.string
  }

## A title in the text.
title =
  element title {
    lang.att,
    
    ## @type is optional:
    (attribute type { "main" | "sub" }?),
    normalized-space.string
  }

## A "meeting" (e.g. session) of the parliament.
meeting = element meeting { global.atts, ana.att, corresp.att, text }

## Publisher of the corpus.
publisher =
  element publisher {
    lang.att,
    ((element orgName { lang.att, normalized-space.string }+,
      ref?)
     | text)
  }

## Funder of the corpus compilation.
funder =
  element funder {
    element orgName { lang.att, normalized-space.string }+,
    ref?
  }

## A name as it appears in the teiHeader (rather than in the
##     transcription).
name-header =
  element name {
    attribute type { xsd:Name }?,
    attribute key { xsd:Name }?,
    lang.att,
    normalized-space.string
  }

## A series of paragraphs. Paragrapgs are only used in the
##     teiHeader.
paras = element p { global.atts, annotated.text }+

## A series of anonymous blocks. Anonymous blocks are only used in the
##     teiHeader.
abs = element ab { global.atts, normalized-space.string }+
a:documentation [ "Start of definitions for text" ]

## The definition of the text and body elements.
text-body =
  element text {
    global.atts,
    ana.att,
    element body { \div+ }
  }

## The definition of the div element.
\div =
  element div {
    global.atts,
    
    ## A standard div with utterances.
    ((attribute type { "debateSection" },
      (head*)
      >> a:documentation [
           "We want at least one u in the debateSection."
         ],
      (comment | pb)*,
      u,
      (comment | pb | u)*)
     >> a:documentation [ "A div with transcriber comments only." ]
     | (attribute type { "commentSection" },
        head*,
        (comment | pb)+))
  }

## One speech (i.e. 'utterance').
u =
  element u {
    global.atts
    >> a:documentation [ "Attributes to link split utterances." ],
    attribute prev { xsd:anyURI }?,
    attribute next { xsd:anyURI }?,
    (who.att?)
    >> a:documentation [
         "Utterance has obligatory @ana, for the role of the speaker."
       ],
    ana.att,
    attribute source { anyURIs }?,
    (seg | comment | pb)+
  }

## A segment (i.e paragraph) inside a speech.
seg =
  element seg {
    global.atts,
    (comment
     | pb
     | text
       # ref name="normalized-space.string"/
       )+
  }

## A transcriber note.
note =
  element note {
    global.atts,
    attribute type { xsd:NCName }?,
    (time | pb | text)+
  }

## A page break, possible with its source URI.
pb = element pb { global.atts, corresp.att, source.att, empty }

## Any type of transcriber comment.
comment =
  note
  | element gap {
      attribute reason { "editorial" | "inaudible" | "foreign" },
      incidentDesc+
    }
  | element kinesic {
      attribute type { kinesicType.val }?,
      who.att?,
      incidentDesc+
    }
  | element incident {
      attribute type { incidentType.val }?,
      who.att?,
      incidentDesc+
    }
  | element vocal {
      attribute type { vocalType.val },
      who.att?,
      incidentDesc+
    }

## A description of an incident in a speech.
incidentDesc = element desc { lang.att, normalized-space.string }

## The media element
media =
  element media {
    global.atts,
    attribute mimeType { xsd:token },
    source.att,
    attribute url { anyURIs }
  }