fragid, maintenance, typos, UML...

.github/workflows/main.yml

@@ -29,8 +29,10 @@ jobs:
       - uses: actions/checkout@v2
       - name: Install packages
         run: |
+          echo ttf-mscorefonts-installer msttcorefonts/accepted-mscorefonts-eula select true | sudo debconf-set-selections
           sudo apt-get update
-          sudo apt-get install xsltproc fop xmlstarlet
+          sudo apt-get install xsltproc fop xmlstarlet ttf-mscorefonts-installer graphviz
+          sudo fc-cache -f
       - name: Make XML, HTML and PDF
         run: make html pdf
       - name: Update Automatic Release

.github/workflows/pr.yml

@@ -29,7 +29,9 @@ jobs:
       - uses: actions/checkout@v2
       - name: Install packages
         run: |
+          echo ttf-mscorefonts-installer msttcorefonts/accepted-mscorefonts-eula select true | sudo debconf-set-selections
           sudo apt-get update
-          sudo apt-get install xsltproc fop xmlstarlet
+          sudo apt-get install xsltproc fop xmlstarlet ttf-mscorefonts-installer graphviz
+          sudo fc-cache -f
       - name: Make XML, HTML and PDF
         run: make html pdf

dmlex-v1.0/specification/Makefile

@@ -13,9 +13,24 @@ dmlex-v1.0-wd01.fo: dmlex-v1.0-wd01.xml
 dmlex-v1.0-wd01.html: dmlex-v1.0-wd01.xml
 	xsltproc --xinclude -o $@ stylesheets/oasis-specification-html.xsl $<
 
-dmlex-v1.0-wd01.xml: dmlex.xml $(SOURCE_XML)
+dmlex-v1.0-wd01.xml: dmlex.xml docbook/dbgenent.mod dmlex_uml.svg $(SOURCE_XML)
 	./merge.sh
 
+dmlex_uml.pdf: dmlex.dot
+	dot -Tpdf < $< > $@
+
+dmlex_uml.svg: dmlex.dot
+	dot -Tsvg < $< > $@
+
+dmlex.dot: dmlex.dot.m4 dmlex.dot.content
+	m4 $< > $@
+
+dmlex.dot.content: dmlex.nvh nvh2dot.py
+	awk '/<programlisting>/,/<\/programlisting>/' $< | tail -n+2 | head -n-1 | ./nvh2dot.py > $@
+
 Makefile.dep: dmlex.xml $(SOURCE_XML) makedep.sh
 	echo -n "SOURCE_XML=" > $@
 	./makedep.sh >> $@
+
+clean:
+	rm dmlex-v1.0-wd01.html dmlex-v1.0-wd01.pdf dmlex-v1.0-wd01.fo dmlex-v1.0-wd01.xml Makefile.dep dmlex_uml.svg dmlex_uml.pdf dmlex.dot dmlex.dot.content

dmlex-v1.0/specification/conformance/conformance.xml

@@ -1,6 +1,6 @@
 <?xml version="1.0" encoding="UTF-8"?>
 <!DOCTYPE section PUBLIC "-//OASIS//DTD DocBook XML V4.5//EN"
-                         "http://www.docbook.org/xml/4.5/docbookx.dtd" [
+                         "../docbook/docbookx.dtd" [
                          <!ENTITY % xinclude SYSTEM "../docbook/xinclude.mod" >
                          %xinclude;
                          <!ENTITY % local.common.attrib "xml:base CDATA #IMPLIED" >

dmlex-v1.0/specification/core/objectTypes/definition.xml

@@ -22,21 +22,9 @@
 
   <itemizedlist>
     <title>Properties</title>
-    <listitem>
-      <para><literal>id</literal>
-        <glossterm>optional</glossterm> (zero or one). A unique identifier.</para>
-      <formalpara>
-        <title>Uniqueness constraint</title>
-        <para>
-          No two instances of <literal>definition</literal>
-          in the same <literal>lexicographicResource</literal>
-          can have the same <literal>id</literal>.
-        </para>
-      </formalpara> 
-    </listitem>
     <listitem>
       <para><literal>text</literal>
-        <glossterm>required</glossterm> (exactly one). Non-empty string. A statement, in the same
+        <glossterm>required</glossterm> (exactly one) and <glossterm>UNIQUE</glossterm>. Non-empty string. A statement, in the same
         language as the headword, that describes and/or explains the meaning of a sense. In DMLex,
         the term definition encompasses not only formal definitions, but also less formal
         explanations.</para>

dmlex-v1.0/specification/core/objectTypes/entry.xml

@@ -22,34 +22,18 @@
 
   <itemizedlist>
     <title>Properties</title>
-    <listitem>
-      <para><literal>id</literal>
-        <glossterm>optional</glossterm> (zero or one). An unique identifier of the entry. Entries
-        which have identifiers are capable of being involved in relations created with the <olink
-          targetptr="linking">Linking module</olink>.</para>
-      <formalpara>
-        <title>Uniqueness constraint</title>
-        <para>
-          No two instances of <literal>entry</literal>
-          in the same <literal>lexicographicResource</literal>
-          (or in the same collection of lexicographic resources if cross-resource linking is required)
-          can have the same <literal>id</literal>.
-        </para>
-      </formalpara> 
-    </listitem>
     <listitem>
       <para><literal>headword</literal>
-        <glossterm>required</glossterm> (exactly one). Non-empty string. The entry's
+        <glossterm>required</glossterm> (exactly one) and <glossterm>unique</glossterm> (in combination with other unique properties if present). Non-empty string. The entry's
         headword.</para>
     </listitem>
     <listitem>
       <para><literal>homographNumber</literal>
-        <glossterm>optional</glossterm> (zero or one). The entry's homograph number, as a guide
-        to distinguish entries with the same headword.</para>
+        <glossterm>optional</glossterm> (zero or one) and <glossterm>unique</glossterm> (in combination with other unique properties if present). The entry's homograph number, as a guide to distinguish entries with the same headword.</para>
     </listitem>
     <listitem>
       <para><literal><olink targetptr="core_partOfSpeech">partOfSpeech</olink></literal>
-        <glossterm>optional</glossterm> (zero or more). </para>
+        <glossterm>optional</glossterm> (zero or more) and <glossterm>unique</glossterm> (in combination with other unique properties if present). </para>
     </listitem>
     <listitem>
       <para><literal><olink targetptr="core_label">label</olink></literal>
@@ -115,7 +99,7 @@
   <example>
     <title>XML</title>
     <programlisting>
-&lt;entry id="..." homographNumber="..."&gt;
+&lt;entry homographNumber="..."&gt;
     &lt;headword&gt;...&lt;/headword&gt;
     &lt;partOfSpeech.../&gt;
     &lt;label.../&gt;
@@ -130,7 +114,6 @@
     <title>JSON</title>
     <programlisting>
 {
-    "id": "...",
     "headword": "...",
     "homographNumber": "...",
     "partsOfSpeech": [...],

dmlex-v1.0/specification/core/objectTypes/example.xml

@@ -23,21 +23,9 @@
 
   <itemizedlist>
     <title>Properties</title>
-    <listitem>
-      <para><literal>id</literal>
-        <glossterm>optional</glossterm> (zero or one). A unique identifier.</para>
-      <formalpara>
-        <title>Uniqueness constraint</title>
-        <para>
-          No two instances of <literal>example</literal>
-          in the same <literal>lexicographicResource</literal>
-          can have the same <literal>id</literal>.
-        </para>
-      </formalpara> 
-    </listitem>
     <listitem>
       <para><literal>text</literal>
-        <glossterm>required</glossterm> (exactly one). Non-empty string. The example itself.</para>
+        <glossterm>required</glossterm> (exactly one) and <glossterm>unique</glossterm>. Non-empty string. The example itself.</para>
     </listitem>
     <listitem>
       <para><literal>sourceIdentity</literal>

dmlex-v1.0/specification/core/objectTypes/inflectedForm.xml

@@ -24,16 +24,9 @@
   <itemizedlist>
     <title>Properties</title>
     <listitem>
-      <para><literal>id</literal>
-        <glossterm>optional</glossterm> (zero or one). A unique identifier.</para>
-      <formalpara>
-        <title>Uniqueness constraint</title>
-        <para>
-          No two instances of <literal>inflectedForm</literal>
-          in the same <literal>lexicographicResource</literal>
-          can have the same <literal>id</literal>.
-        </para>
-      </formalpara> 
+      <para><literal>text</literal>
+        <glossterm>required</glossterm> (exactly one) and <glossterm>unique</glossterm>. Non-empty string. The text of the inflected
+        form.</para>
     </listitem>
     <listitem>
       <para><literal>tag</literal>
@@ -45,11 +38,6 @@
         inflection tags are allowed to occur in the lexicographic resource, and to map them onto
         external inventories and ontologies.</para>
     </listitem>
-    <listitem>
-      <para><literal>text</literal>
-        <glossterm>required</glossterm> (exactly one). Non-empty string. The text of the inflected
-        form.</para>
-    </listitem>
     <listitem>
       <para><literal><olink targetptr="core_label">label</olink></literal>
         <glossterm>optional</glossterm> (zero or more).</para>

dmlex-v1.0/specification/core/objectTypes/lexicographicResource.xml

@@ -47,10 +47,10 @@
     <title>Comments</title>
     <listitem>
       <para>
-        <code>language</code> identifies the language of headwords, definitions and examples in this
+        <code>langCode</code> identifies the language of headwords, definitions and examples in this
         dictionary. DMLex is based on the assumption that all headwords in a lexicographic resource
         are in the same language, and that definitions and examples, if any are included in the
-        lexicographic resource, are in that language too. The <code>language</code> property of
+        lexicographic resource, are in that language too. The <code>langCode</code> property of
           <code>lexicographicResource</code> informs potential users of the lexicographic resource
         which language that is.</para>
     </listitem>
@@ -60,11 +60,9 @@
         for the existence of lexicographic resources which are not yet complete. </para>
     </listitem>
     <listitem>
-      <para>The <code>lexicographicResource</code> data type does not contain properties for
-        detailed metadata about the lexicographic resource, such as author, editor, publisher,
-        copyright status or publication year. Describing these properties of lexicographic resources
-        is outside the scope of DMLex. DMLex is a formalism for modelling the internal structure of
-        a lexicographic resource, not its metadata.</para>
+      <para>The <code>lexicographicResource</code> data type does not prescribe properties for
+	  detailed metadata about the lexicographic resource, implementers are advised to refer to
+	  the properties in the <link linkend="DublinCore">Dublin Core <code>elements</code> namespace</link> as for basic metadata. </para>
     </listitem>
   </itemizedlist>
 

dmlex-v1.0/specification/core/objectTypes/pronunciation.xml

@@ -25,28 +25,16 @@
 
   <itemizedlist>
     <title>Properties</title>
-    <listitem>
-      <para><literal>id</literal>
-        <glossterm>optional</glossterm> (zero or one). A unique identifier.</para>
-      <formalpara>
-        <title>Uniqueness constraint</title>
-        <para>
-          No two instances of <literal>pronunciation</literal>
-          in the same <literal>lexicographicResource</literal>
-          can have the same <literal>id</literal>.
-        </para>
-      </formalpara> 
-    </listitem>
     <listitem>
       <para>At least one of:</para>
       <itemizedlist>
         <listitem>
           <para><literal>soundFile</literal>
-            <glossterm>optional</glossterm> (zero or one). A pointer to a file, such as a filename or a URI, containing a sound recording of the pronunciation</para>
+            <glossterm>optional</glossterm> (zero or one) and <glossterm>unique</glossterm>. A pointer to a file, such as a filename or a URI, containing a sound recording of the pronunciation</para>
         </listitem>
         <listitem>
           <para><literal><olink targetptr="core_transcription">transcription</olink></literal>
-            <glossterm>optional</glossterm> (zero or more).</para>
+            <glossterm>optional</glossterm> (zero or more) and <glossterm>unique</glossterm>.</para>
         </listitem>
       </itemizedlist>
     </listitem>

dmlex-v1.0/specification/core/objectTypes/sense.xml

@@ -22,29 +22,14 @@
 
   <itemizedlist>
     <title>Properties</title>
-    <listitem>
-      <para><literal>id</literal>
-        <glossterm>optional</glossterm> (zero or one). A unique identifier of the sense. Senses
-        which have identifiers can be involved in relations created with the <olink
-          targetptr="linking">Linking module</olink>.</para>
-      <formalpara>
-        <title>Uniqueness constraint</title>
-        <para>
-          No two instances of <literal>sense</literal>
-          in the same <literal>lexicographicResource</literal>
-          (or in the same collection of lexicographic resources if cross-resource linking is required)
-          can have the same <literal>id</literal>.
-        </para>
-      </formalpara> 
-    </listitem>
     <listitem>
       <para><literal>listingOrder</literal>
         <glossterm>required</glossterm> (exactly one). Number. The position of this sense among
         other senses of the same entry. Can be implicit from the serialization.</para>
     </listitem>
     <listitem>
       <para><literal>indicator</literal>
-        <glossterm>optional</glossterm> (zero or one). A short statement, in the same language as
+        <glossterm>optional</glossterm> (zero or one) and <glossterm>unique</glossterm>. A short statement, in the same language as
         the headword, that gives an indication of the meaning of a sense and permits its
         differentiation from other senses in the entry. Indicators are sometimes used in
         dictionaries instead of or in addition to definitions.</para>
@@ -55,7 +40,7 @@
     </listitem>
     <listitem>
       <para><literal><olink targetptr="core_definition">definition</olink></literal>
-        <glossterm>optional</glossterm> (zero or more). </para>
+        <glossterm>optional</glossterm> (zero or more) and <glossterm>unique</glossterm>. </para>
     </listitem>
     <listitem>
       <para><literal><olink targetptr="core_example">example</olink></literal>
@@ -81,7 +66,7 @@
   <example>
     <title>XML</title>
     <programlisting>
-&lt;sense id="..."&gt;
+&lt;sense&gt;
     &lt;indicator&gt;...&lt;/indicator&gt;
     &lt;label.../&gt;
     &lt;definition.../&gt;
@@ -94,7 +79,6 @@
     <title>JSON</title>
     <programlisting>
 {
-    "id": "...",
     "indicator": "...",
     "labels": [...],
     "definitions": [...],

dmlex-v1.0/specification/core/objectTypes/transcription.xml

@@ -22,21 +22,9 @@
 
   <itemizedlist>
     <title>Properties</title>
-    <listitem>
-      <para><literal>id</literal>
-        <glossterm>optional</glossterm> (zero or one). A unique identifier.</para>
-      <formalpara>
-        <title>Uniqueness constraint</title>
-        <para>
-          No two instances of <literal>transcription</literal>
-          in the same <literal>lexicographicResource</literal>
-          can have the same <literal>id</literal>.
-        </para>
-      </formalpara> 
-    </listitem>
     <listitem>
       <para><literal>text</literal>
-        <glossterm>required</glossterm> (exactly one). Non-empty string. The actual
+        <glossterm>required</glossterm> (exactly one) and <glossterm>unique</glossterm>. Non-empty string. The actual
         transcription.</para>
     </listitem>
     <listitem>

dmlex-v1.0/specification/core/specification.xml

@@ -41,12 +41,30 @@
 			<para><literal><olink targetptr="core_example">example</olink></literal></para>
     </listitem>
   </itemizedlist>
-  <para>
-    When exchanging data encoded in a DMLex serialization
-    which has the concept of a "root" or top-level object, such as XML and JSON,
-    then the object types <literal>lexicographicResource</literal> and <literal>entry</literal>
-    can serve as such roots.
-  </para>
+  <simplesect id="optionalroots">
+    <title>Optional roots</title>
+    <para>
+      When exchanging data encoded in a DMLex serialization
+      which has the concept of a "root" or top-level object, such as XML, JSON or NVH,
+      the object types <literal>lexicographicResource</literal> and <literal>entry</literal>
+      can serve as such roots.
+    </para>
+  </simplesect>
+  <simplesect id="fragid">
+    <title>Fragment identification</title>
+    <para>
+      Incomplete parts of DMLex objects represent valid fragments as long as it is possible to identify their complete source DMLex object.
+      For a particular serialization, this can be achieved by:
+      <itemizedlist>
+        <listitem>
+          <para>Making sure the fragment contains all <olink targetptr="uniqueids">unique properties</olink> of the DMLex parent-child hierarchy that are necessary to determine its source object.</para>
+        </listitem>
+        <listitem>
+          <para>Using a serialization-specific addressing mechanism (e.g. XPath for XML, JMESPath for JSON or an NVH query) to provide the location of the fragment in the source DMLex object.</para>
+        </listitem>
+     </itemizedlist>
+	</para>
+  </simplesect>
   <xi:include href="objectTypes/lexicographicResource.xml" xmlns:xi="http://www.w3.org/2001/XInclude"/>
   <xi:include href="objectTypes/entry.xml" xmlns:xi="http://www.w3.org/2001/XInclude"/>
   <xi:include href="objectTypes/partOfSpeech.xml" xmlns:xi="http://www.w3.org/2001/XInclude"/>

dmlex-v1.0/specification/dmlex.dot.m4

@@ -0,0 +1,7 @@
+digraph dmlex {
+rankdir="LR";
+splines="ortho";
+node[shape=record,style=filled,fillcolor=gray95]
+edge[arrowhead=empty]
+include(`dmlex.dot.content')
+}