PLATEAU.DTD
used to encode the e-MED
<!-- May 1999 revision of the TEI-like 'TEIish' dtd -->
<!-- designed to take e-MED coding to a new plateau -->
<!-- with further revisions as noted in comments -->
<!-- FIRST DRAFT 4May1999 / pfs -->
<!-- Revised Oct 1999 / pfs -->
<!ELEMENT text - - (body) +(lb|hi)>
<!ELEMENT body - - (entryfree)+ >
<!ELEMENT entryfree - - (#PCDATA | form | etym | sense | sensegrp | usg | note | xr)* >
<!ATTLIST entryfree id ID #IMPLIED
seq CDATA #IMPLIED
dub (y|n) #IMPLIED
type (xref|affx|main) #IMPLIED
new CDATA #IMPLIED
rev CDATA #IMPLIED >
<!-- entryfree: name inherited from TEI dictionaries module. Each entryfree contains either -->
<!-- a 'real' entry or a block of cross-references (the latter inherited from legacy data -->
<!-- and stripped before indexing) -->
<!-- dwr added usg to content model to handle syntactic usage information living between the -->
<!-- etym and first sense and applying to entire entry, 1997-07-21 -->
<!-- pfs added note to content model, following MED, to handle "Cp...." notes after (or in) -->
<!-- form sections. TEI equivalence dictates replacing this eventually with XR element -->
<!-- plateau added dub (dubious=queried entries), type (following TEI), and new/rev -->
<!-- (added/revision date) attributes, though none of these is yet used -->
<!-- pfs added seq attribute to allow automatic assignment of sequence to MED entries Oct 99 -->
<!-- ======================================================================================= -->
<!ELEMENT form - - (#pcdata | hdorth | orth | pos | lbl)* >
<!-- form: contains the MED 'form section' that runs from headword through spelling variants -->
<!-- plateau added hdorth and lbl to content model -->
<!-- ======================================================================================= -->
<!ELEMENT orth - - (#pcdata) >
<!ATTLIST orth id ID #IMPLIED
norm CDATA #IMPLIED
split CDATA #IMPLIED
exp CDATA #IMPLIED >
<!-- orth: contains spelling form (as well as some layered phonological info, which is -->
<!-- not separately tagged at present and probably never will be). -->
<!-- plateau added last three attributes to handle possible form expansion in attributes, -->
<!-- using TEI 'editorial view' to expand compressed information. -->
<!-- ======================================================================================= -->
<!ELEMENT hdorth - - (#pcdata) >
<!ATTLIST hdorth id ID #IMPLIED
norm CDATA #IMPLIED
split CDATA #IMPLIED
exp CDATA #IMPLIED >
<!-- hdorth: added by plateau as optional way to handle headwords. Currently tagged simply -->
<!-- as orth. -->
<!-- plateau added last three attributes to handle possible form expansion in attributes, -->
<!-- using TEI 'editorial view' to expand compressed information -->
<!-- ======================================================================================= -->
<!ELEMENT pos - - (#pcdata) >
<!ATTLIST pos exp CDATA #IMPLIED
n CDATA #IMPLIED >
<!-- pos: part of speech label, borrowed from TEI but not really equivalent. -->
<!-- plateau added attributes exp (to expand abbreviated POS) and n -->
<!-- (to code numerators, e.g. the "2" in "n.(2)") if desired. -->
<!-- note that MEC uses pos as rough equivalent to TEI gramgrp element, i.e. groups -->
<!-- grammatical information related to lemma -->
<!-- ======================================================================================= -->
<!ELEMENT sensegrp - - (def,sense+) >
<!ATTLIST sensegrp n CDATA #IMPLIED id ID #IMPLIED>
<!-- sensegrp added to cover roman-numeral sense groups in E and F vols., pfs 15 Oct 98 -->
<!-- Occasionally used also to code groups of subsenses that begin with an introductory -->
<!-- phrase ("in prepositional phrases:"), though the latter *may* also be found tagged -->
<!-- as a usg or usg type="def" -->
<!-- ======================================================================================= -->
<!ELEMENT sense - - ((def)+,(eg)+) >
<!ATTLIST sense n CDATA #IMPLIED id ID #IMPLIED>
<!-- nigelk and dwr changed eg+ to eg*, 1997-07-18, for affix entries without attached -->
<!-- quotations; changed back temporarily to catch affix entries, dwr 1997-07-24 -->
<!-- left in place for now by plateau -->
<!-- ======================================================================================= -->
<!ELEMENT def - - ((#pcdata | date | hi | title | bibl | note | usg | def | p | pos | lang | tax | xref | word | re)*) >
<!ATTLIST def n CDATA #IMPLIED>
<!-- p added to def content model by nigelk, 1997-07-23, to preserve paragraphing in affix -->
<!-- entries; plateau added pos lang tax xref word -->
<!-- ======================================================================================= -->
<!ELEMENT p - - ((#pcdata | date | hi | title | bibl | note | usg | pos | lang | tax | xref | word)*) >
<!ATTLIST p n CDATA #IMPLIED>
<!-- p: paragraph -->
<!-- added p to the dtd, for them durn affixes, nigelk 1997-07-24 -->
<!-- plateau removed p from itself, added other elements to match def -->
<!-- ======================================================================================= -->
<!ELEMENT eg - - (cit)+ +(etym|form|usg)>
<!ATTLIST eg n CDATA #IMPLIED id ID #IMPLIED>
<!-- eg: examples. groups quotation/citation blocks under each sense -->
<!-- the inclusion is inherited from previous versions but has no obvious justification. -->
<!-- need to check if it is being used or can safely be removed (pfs/Oct 99) -->
<!-- ======================================================================================= -->
<!ELEMENT cit - - (bibl,q) >
<!ATTLIST cit
n CDATA #IMPLIED
id ID #IMPLIED
type (n|b) n>
<!-- cit: citation: groups bibliographical information and quotation. -->
<!-- type added to attributes of cit by pfs 27May98 to handle bracketed quots. (those -->
<!-- regarded as not Middle English and provided only as illustrative of the word's history -->
<!-- ======================================================================================= -->
<!ELEMENT bibl - - (#pcdata | STNCL)* >
<!ATTLIST bibl id ID #IMPLIED
ibid (y|n) n>
<!-- bibl: contains bibliographic information (stencil and page reference) -->
<!-- ibid added as attribute of bibl by pfs 8 Oct 99 to flag ibids removed automatically by -->
<!-- de-ibidizing script; manually treated ibids not so flagged, on the assumption that -->
<!-- machines may make mistakes but people never do :-) -->
<!-- ======================================================================================= -->
<!ELEMENT q - - (#pcdata | title | added | ovar)* >
<!ATTLIST q n CDATA #IMPLIED id ID #IMPLIED >
<!-- q: quotation -->
<!-- nigelk and dwr added TITLE to content model of q, to handle variant readings 1997-07-21 -->
<!-- Plateau left title in place to handle legacy coding, but all instances really belong -->
<!-- in "added" tags; if that element is used, title should be removed from this content -->
<!-- Plateau added ovar to content to allow tagging of headword in quotations, if desired -->
<!-- ======================================================================================= -->
<!ELEMENT STNCL - - (#PCDATA | date | hi | title | author | ms)* >
<!ATTLIST STNCL id ID #IMPLIED
rid CDATA #IMPLIED>
<!-- stncl: stencil (MED bibliographic short form) -->
<!-- rid attribute added Nov 98/pfs to accommodate pointers to hyperbib -->
<!-- (R)IDs generated by merger process of bibliographic harmonization with hb -->
<!-- ======================================================================================= -->
<!ELEMENT ms - - (#pcdata) >
<!ATTLIST ms exp CDATA #IMPLIED
rid CDATA #IMPLIED >
<!-- ms element added by pfs to allow tagging of ms abbreviations during -->
<!-- stencil merger, May 98. HB's "abbr" used otherwise in plateau.dtd -->
<!-- exp and rid added since this element may also be used eventually to tag variant -->
<!-- readings in quots. -->
<!-- ======================================================================================= -->
<!ELEMENT author - - (#pcdata) >
<!ATTLIST author id ID #IMPLIED
sort CDATA #IMPLIED >
<!-- ======================================================================================= -->
<!ELEMENT date - - (#pcdata) >
<!ATTLIST date cp NUMBER #IMPLIED
ms NUMBER #IMPLIED >
<!-- attributes added by plateau to simplify date searching by COMPOSITION and MS date -->
<!-- these attributes not currently used. -->
<!-- ======================================================================================= -->
<!ELEMENT title - - (#pcdata) >
<!ATTLIST title rid CDATA #IMPLIED
exp CDATA #IMPLIED >
<!-- title: used broadly within stencil to mark everything rendered in italics in MED. -->
<!-- also used elsewhere either to code real (book) titles or to refer to pieces of stencils -->
<!-- ======================================================================================= -->
<!ELEMENT usg - - (#pcdata | ovar)* >
<!ATTLIST usg TYPE (general|field|def|syntax) #IMPLIED
REND (norm|ital) ital
EXP CDATA #IMPLIED >
<!-- usg: usage labels -->
<!-- REND attribute experimentally added 3Apr98/pfs in order to handle the -->
<!-- non-italic usg labels being employed by current MED production. -->
<!-- "general" removed as default usg type in favor of #implied, 29July98/pfs, because -->
<!-- MED is using TYPE="general" very specifically -->
<!-- Plateau added exp attribute to handle attribute-expanded abbreviated usage labels -->
<!-- Plateau added ovar to content to allow tagging of bold participles as ovar instead -->
<!-- of hi within usg tags. Neither exp or ovar currently used -->
<!-- ======================================================================================= -->
<!ELEMENT etym - - (#pcdata | lang | ety | xref)* >
<!-- etym: etymology statement, rendered in square brackets -->
<!-- Plateau added ety and xref tags. all bold text in etymologies should be tagged either -->
<!-- as ety or within xref tags. eventually. there may be exceptions. -->
<!-- ======================================================================================= -->
<!ELEMENT hi - - (#pcdata) >
<!ATTLIST hi rend CDATA #IMPLIED>
<!-- ======================================================================================= -->
<!ELEMENT lang - - (#pcdata) >
<!ATTLIST lang exp CDATA #IMPLIED >
<!-- Plateau added exp attribute to expand abbreviated languages if desired. -->
<!-- ======================================================================================= -->
<!ELEMENT lb - O EMPTY >
<!-- ======================================================================================= -->
<!ELEMENT xr - - (#pcdata | pos | xref | word | abbr)* >
<!ELEMENT note - - (#pcdata | pos | xref | word | abbr | lang | ety | author)* >
<!ATTLIST note rend (b|p|n) #IMPLIED
type CDATA #IMPLIED >
<!-- xr: extended reference note -->
<!-- Plateau added xr alternative to match TEI semantics; also pos, xref, word. Most current -->
<!-- "notes" should be changed to XR -->
<!-- Note element may be used for more general purposes than cross-reference notes; plateau -->
<!-- enlarged content model of note. -->
<!-- Plateau added rend attrib to note to cover bracketed, parenthesised and unmarked notes. -->
<!-- Plateau added type attrib to note to allow searching specific types of notes for -->
<!-- specific types of info (e.g. etymological) -->
<!-- ======================================================================================= -->
<!ELEMENT ety - - (#pcdata) >
<!ATTLIST ety rid CDATA #IMPLIED
lang CDATA #IMPLIED
type CDATA #IMPLIED
src CDATA #IMPLIED >
<!-- ety: etymon -->
<!-- Plateau added this element to replace bold text in etymologies and etymological notes. -->
<!-- type attrib designed for head/variant distinction -->
<!-- Src and rid attribs are hooks to attach related dictionaries -->
<!-- ======================================================================================= -->
<!ELEMENT tax - - (#pcdata) >
<!ATTLIST tax par (y|n) y
exp CDATA #IMPLIED
type CDATA #IMPLIED >
<!-- Plateau added this element to tag botanical/zoological taxonomic names in definitions -->
<!-- ======================================================================================= -->
<!ELEMENT xref - - (#pcdata | word | pos | ety)* >
<!ATTLIST xref rid CDATA #IMPLIED
sense CDATA #IMPLIED >
<!-- Plateau added this element as optional way to group text elements constituting an -->
<!-- internal cross-reference. currently tagged as xref containing hi and pos -->
<!-- ======================================================================================= -->
<!ELEMENT lbl - - (#pcdata | dial | era | word | xref)* >
<!ATTLIST lbl exp CDATA #IMPLIED
rid CDATA #IMPLIED
par (y|n) y >
<!-- Plateau added this element to tag parenthesised (and nonpar, see par attrib) labels in -->
<!-- form sections, either free form or using specific kinds of labels (dial, era) -->
<!-- ======================================================================================= -->
<!ELEMENT (dial|era) - - (#pcdata) >
<!ATTLIST (dial|era) exp CDATA #IMPLIED
rid CDATA #IMPLIED >
<!-- Plateau added these elements to tag specific abbreviated dialect and period labels in -->
<!-- form sections -->
<!-- ======================================================================================= -->
<!ELEMENT phrase - - (#pcdata | word | ovar | oref)* >
<!ATTLIST phrase norm CDATA #IMPLIED
split CDATA #IMPLIED
exp CDATA #IMPLIED >
<!-- Plateau added this element to tag bold phrases and combinations in definitions -->
<!-- ======================================================================================= -->
<!ELEMENT abbr - - (#pcdata | title | author)* >
<!ATTLIST abbr exp CDATA #IMPLIED
rid CDATA #IMPLIED >
<!-- Plateau added this element to tag (optionally) abbreviations other than pos lbl usg etc -->
<!-- ======================================================================================= -->
<!ELEMENT word - - (#pcdata) >
<!ATTLIST word lang CDATA #IMPLIED
rend (i|b|n) #IMPLIED
rid CDATA #IMPLIED >
<!-- Plateau added this element to tag citations of words and implicit cross-references in -->
<!-- notes, etc. -->
<!-- lang attrib distinguishes ME words from others -->
<!-- rend attrib distinguishes italic, bold, and normal -->
<!-- ======================================================================================= -->
<!ELEMENT added - - (#pcdata | lang | title | stncl | abbr | ms | me)* >
<!-- Plateau added this element to tag bracketed non-ME material in quotations. Similar to -->
<!-- TEI add/supplied -->
<!-- ======================================================================================= -->
<!ELEMENT me - - (#pcdata) >
<!-- Plateau added this element to tag exceptions to added, i.e. ME stuff within bracketed -->
<!-- additions. None of these tags currently used or supported by middleware -->
<!-- ======================================================================================= -->
<!ELEMENT oref - o EMPTY >
<!ATTLIST oref exp CDATA #IMPLIED >
<!ELEMENT ovar - - (#pcdata) >
<!ATTLIST ovar type CDATA #IMPLIED
rend (b|i|n) #IMPLIED >
<!-- Plateau added ovar and oref, used more or less as in TEI -->
<!-- ======================================================================================= -->
<!ELEMENT re - - (phrase, (#pcdata | note | def)*) >
<!-- Plateau added re (related entry) to allow grouping of definitions with the phrases that -->
<!-- they define. Not used. -->
<!-- ======================================================================================= -->
<!-- The following entities file contains all the necessary MED character entities -->
<!ENTITY % MEDsel1 PUBLIC "MEC 1998-1//ENTITIES MED 1//EN">
%MEDsel1;
<!-- END -->