<!-- corpus.dtd -->
<!-- $Id:$ -->

<!--+ Common attributes
    |    Every element has an xml:lang attribute.
    +-->
<!ENTITY % common.att '
  id CDATA #IMPLIED
  xml:lang NMTOKEN #IMPLIED'>

<!--+ Error entity
    |    All error types listed here
    +-->
<!ENTITY % error.ent "error | errorort | errorortreal | errorlex | errormorphsyn | errorsyn" >

<!--+ p continuation
    |    Every p element can contain the following:
    +-->
<!ENTITY % p_cont '
#PCDATA | em | hyph | %error.ent; | span | list'>

<!-- =============================================================== -->
<!-- Document -->
<!-- =============================================================== -->
<!ELEMENT document (header, body)>
<!ATTLIST document 
  %common.att; >


<!ELEMENT header (title,
                  genre+,
                  author+, 
                  translator*,
                  translated_from*,
                  year?,
                  place?,
                  publChannel?,
                  collection?,
                  wordcount?,
                  availability?,
                  submitter?,
                  multilingual?,
                  origFileName?, 
                  parallel_text*,
                  metadata,
                  version?,
                  note? )>

<!ELEMENT title (#PCDATA)>

<!ELEMENT translated_from EMPTY>
<!ATTLIST translated_from  
%common.att; >


<!-- scheme="dewey" code="44444" -->
<!-- scheme should be dewey or uit or whatever -->
<!ELEMENT genre EMPTY >
<!ATTLIST genre
            scheme  CDATA         #IMPLIED
            code    CDATA         #IMPLIED
    >

<!ELEMENT author (person | unknown)>
<!ELEMENT translator (person | unknown)>

<!ELEMENT person EMPTY >
<!ATTLIST person
            firstname CDATA         #REQUIRED
            lastname  CDATA         #REQUIRED
            born      CDATA         #IMPLIED
            sex       (m|f|unknown)   "unknown"
            nationality CDATA  #IMPLIED
    >

<!ELEMENT unknown EMPTY>

<!-- Whether the document is part of some journal, -->
<!-- book or article collection -->
<!ELEMENT collection (#PCDATA) >

<!ELEMENT year  (#PCDATA)>
<!ELEMENT place (#PCDATA)>
<!ELEMENT publChannel (publication|unpublished)>
<!ELEMENT publication (publisher, ISBN?, ISSN? )> <!-- etc -->

<!ELEMENT unpublished EMPTY>
<!ELEMENT publisher (#PCDATA)>
<!ELEMENT ISBN (#PCDATA)>
<!ELEMENT ISSN (#PCDATA)>

<!ELEMENT wordcount (#PCDATA)>


<!-- License information -->
<!ELEMENT availability (free|license)>
<!ELEMENT free EMPTY>
<!ELEMENT license EMPTY>
<!ATTLIST license
    type  (standard|other) "standard"
    contract_id CDATA #IMPLIED
>

<!-- indicates whether all relevant metadata is filled -->
<!ELEMENT metadata (complete|uncomplete)>
<!ELEMENT complete EMPTY>
<!ELEMENT uncomplete EMPTY>

<!-- The original name of the file -->
<!-- The name may change during the conversion, -->
<!-- although not yet implemented -->
<!ELEMENT origFileName (#PCDATA)>

<!ELEMENT parallel_text EMPTY>
<!ATTLIST parallel_text
           location CDATA #REQUIRED
           %common.att; >

<!-- List of languages that appear in the documente -->
<!-- The list is used in language recognition -->
<!ELEMENT multilingual (language)*>
<!ELEMENT language EMPTY>
<!ATTLIST language
%common.att; >

<!-- Version information of the different conversion tools -->
<!ELEMENT version (#PCDATA)>

<!-- The person who stored the file in the database -->
<!-- In case of web-upload, the one who did the upload -->
<!ELEMENT submitter EMPTY >
<!ATTLIST submitter
            name   CDATA     #REQUIRED
            email  CDATA     #REQUIRED>

<!ELEMENT note (#PCDATA) >

<!ENTITY % text.ent "list|table|p|pre">
<!ELEMENT body (section|(%text.ent;))*>
<!ELEMENT section (section|(%text.ent;))*>

<!ELEMENT list (p*) > <!-- type="listitem" -->
<!ELEMENT table (row*)>
<!ELEMENT row (p+)> <!-- type="tablecell" -->

<!ELEMENT p ( %p_cont; | s )* >
<!ATTLIST p type ( title | listitem | text | tablecell ) "text"
  %common.att; >

<!ELEMENT em (#PCDATA | hyph | %error.ent; )* >
<!ATTLIST em type ( bold | italic | underline | delimited ) "italic" >

<!ELEMENT span (#PCDATA | em | hyph | %error.ent; )* >
<!ATTLIST span 
  type ( quote ) "quote"
  %common.att; >

<!ELEMENT hyph EMPTY>
<!ELEMENT pre (#PCDATA)>

<!-- Error markup -->
<!-- this is <error correct="text">tekst</error> ... -->
<!-- Errors can be nested -->

<!-- Unclassified errors: -->
<!ELEMENT error (#PCDATA | %error.ent; | hyph | em)*>
<!ATTLIST error correct CDATA #IMPLIED >

<!-- Orthographical errors, non-words: -->
<!ELEMENT errorort (#PCDATA | hyph | errorlex)*>
<!ATTLIST errorort pos     CDATA #IMPLIED >
<!ATTLIST errorort errtype CDATA #IMPLIED >
<!ATTLIST errorort teacher CDATA #IMPLIED  >
<!ATTLIST errorort correct CDATA #IMPLIED >

<!-- Orthographical errors, real words: -->
<!ELEMENT errorortreal (#PCDATA | hyph | errorlex | errorort)*>
<!ATTLIST errorortreal pos     CDATA #IMPLIED >
<!ATTLIST errorortreal errtype CDATA #IMPLIED >
<!ATTLIST errorortreal teacher CDATA #IMPLIED  >
<!ATTLIST errorortreal correct CDATA #IMPLIED >

<!--
<!ATTLIST errorort
	pos ( noun | verb | adj | adv | num | interj | pp | cc | cs | pers | refl | dem | resip | indef | pcle | prop | acr | abbr | x ) "noun"
	errtype ( a | á | conc | svow | vowc | vow | con | mono | diph | lime | meta | cap | min | suf | mix | ascii | cmp ) "svow"
	teacher ( yes | no ) "no"
	correct CDATA #IMPLIED
>
-->

<!-- Lexical errors: -->
<!ELEMENT errorlex (#PCDATA | error | errorort | errorortreal | errormorphsyn | hyph )*>
<!ATTLIST errorlex correct CDATA #IMPLIED >
<!ATTLIST errorlex pos CDATA #IMPLIED >
<!ATTLIST errorlex origpos CDATA #IMPLIED >
<!ATTLIST errorlex errtype CDATA #IMPLIED >
<!ATTLIST errorlex teacher CDATA #IMPLIED  >

<!-- <!ATTLIST errorlex -->
<!--           correct CDATA #IMPLIED -->
<!--           pos ( noun | verb | adj | adv | num | interj | pp | cc | cs | pers | -->
<!--                 refl | dem | resip | indef | pcle | prop | x ) -->
<!--       origpos ( noun | verb | adj | adv | num | interj | pp | cc | cs | pers | -->
<!--                 refl | dem | resip | indef | pcle | prop | foreign | x ) -->
<!--           errtype ( der | w | x  ) -->
<!--           teacher ( yes | no ) -->
<!--           > -->

<!-- Morphosyntactic errors: -->
<!ELEMENT errormorphsyn ( %p_cont; )*>
<!ATTLIST errormorphsyn correct CDATA #IMPLIED >
<!ATTLIST errormorphsyn pos CDATA #IMPLIED >
<!ATTLIST errormorphsyn const CDATA #IMPLIED >
<!ATTLIST errormorphsyn cat CDATA #IMPLIED >
<!ATTLIST errormorphsyn orig CDATA #IMPLIED >
<!ATTLIST errormorphsyn errtype CDATA #IMPLIED >
<!ATTLIST errormorphsyn teacher CDATA #IMPLIED  >

<!-- <!ATTLIST errormorphsyn -->
<!--           correct CDATA #IMPLIED -->
<!--           pos ( noun | verb | adj | adv | num | interj | pp | cc | cs | pers | -->
<!--                 refl | dem | resip | indef | pcle | prop | x ) -->
<!--           const ( subj | obj | advl | v | spred | opred  | pcle  | interj | app | -->
<!--                   conj | pph | attr | x ) -->
<!--           cat ( nomsg | nompl | gensg | genpl | illsg | illpl | locsg | locpl | -->
<!--                 comsg | compl | ess | sg1prt | sg2prt | sg3prt | du1prt | du2prt | -->
<!--                 du3prt | pl1prt | pl2prt | pl3prt | sg1prs | sg2prs | sg3prs | -->
<!--                 du1prs | du2prs | du3prs | pl1prs | pl2prs | pl3prs | attr | -->
<!--                 pred | word | comp | superl | cmp | imprt | pot | infinite | -->
<!--                 cond | conneg | ger | vgen | x ) -->
<!--           orig ( nomsg | nompl | gensg | genpl | illsg | illpl | locsg | locpl | -->
<!--                  comsg | compl | ess | sg1prt | sg2prt | sg3prt | du1prt | -->
<!--                  du2prt | du3prt | pl1prt | pl2prt | pl3prt | sg1prs | sg2prs | -->
<!--                  sg3prs | du1prs | du2prs | du3prs | pl1prs | pl2prs | pl3prs | -->
<!--                  attr | pred | word | comp | superl | cmp | imprt | pot | -->
<!--                  infinite | cond | conneg | ger | vgen | x ) -->
<!--           errtype ( agr | case | tense | mode | nump | number | x  ) -->
<!--           teacher ( yes | no ) -->
<!--           > -->

<!-- Syntactic errors: -->
<!ELEMENT errorsyn ( %p_cont; )*>
<!ATTLIST errorsyn correct CDATA #IMPLIED >
<!ATTLIST errorsyn pos CDATA #IMPLIED >
<!ATTLIST errorsyn errtype CDATA #IMPLIED >
<!ATTLIST errorsyn teacher CDATA #IMPLIED  >
<!-- <!ATTLIST errorsyn -->
<!--           correct CDATA #IMPLIED -->
<!--           pos ( noun | verb | adj | adv | num | interj | pp | cc | cs | pers | -->
<!--                 refl | dem | resip | indef | pcle | prop | punct | x ) -->
<!--           errtype ( wo | pph | redun | missing | cmp  | x ) -->
<!--           teacher ( yes | no ) -->
<!--           > -->

<!-- Analyzed corpus -->

<!ELEMENT s (#PCDATA | w)*>
<!ATTLIST s
  %common.att; >

<!ELEMENT w (reading*)>
<!ATTLIST w
          form CDATA #REQUIRED
		  %common.att; >		  

<!ELEMENT reading EMPTY>
<!ATTLIST reading
    lemma CDATA #IMPLIED 
    Wordclass CDATA #IMPLIED
	Subclass CDATA #IMPLIED
	Number CDATA #IMPLIED
	Case CDATA #IMPLIED
	Attributive CDATA #IMPLIED
	Grade CDATA #IMPLIED
	Possessive CDATA #IMPLIED
	Tense CDATA #IMPLIED
	Mood CDATA #IMPLIED
	Person-Number CDATA #IMPLIED
	Transitivity CDATA #IMPLIED
	Polarity CDATA #IMPLIED
	Diathesis CDATA #IMPLIED
	Infinite CDATA #IMPLIED
	Clitic CDATA #IMPLIED
	Punctuation CDATA #IMPLIED
	Nametype CDATA #IMPLIED
	Numeraltype CDATA #IMPLIED
	Syntax CDATA #IMPLIED
	Derivation CDATA #IMPLIED
	DerivedRoot CDATA #IMPLIED
>