MCR-3111 mycore-tei datamodel and tools

mycore-base/src/main/java/org/mycore/common/MCRConstants.java

@@ -119,6 +119,8 @@ public final class MCRConstants {
 
     public static final Namespace MCR_NAMESPACE = Namespace.getNamespace("mcr", MCR_URL);
 
+    public static final Namespace TEI_NAMESPACE = Namespace.getNamespace("tei", "http://www.tei-c.org/ns/1.0");
+
     private static final HashMap<String, Namespace> NAMESPACES_BY_PREFIX;
 
     static {

mycore-tei/pom.xml

@@ -33,6 +33,14 @@
       <groupId>junit</groupId>
       <artifactId>junit</artifactId>
     </dependency>
+    <dependency>
+      <groupId>org.mycore</groupId>
+      <artifactId>mycore-base</artifactId>
+    </dependency>
+    <dependency>
+      <groupId>org.apache.logging.log4j</groupId>
+      <artifactId>log4j-api</artifactId>
+    </dependency>
   </dependencies>
   <dependencyManagement>
     <dependencies>

mycore-tei/src/main/datamodel/def/tei.xml

@@ -0,0 +1,15 @@
+<objecttype name="tei" isChild="false" isParent="false" hasDerivates="true" xmlns:xs="http://www.w3.org/2001/XMLSchema"
+            xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
+            xmlns:tei="http://www.tei-c.org/ns/1.0"
+            xsi:noNamespaceSchemaLocation="datamodel.xsd">
+  <xsd>
+    <xs:import namespace="http://www.tei-c.org/ns/1.0" schemaLocation="schema/tei_all.xsd" />
+  </xsd>
+  <metadata>
+    <element name="teiContainer" type="xml" style="dontknow" notinherit="true" heritable="false">
+      <xs:sequence>
+        <xs:element ref="tei:teiHeader" />
+      </xs:sequence>
+    </element>
+  </metadata>
+</objecttype>

mycore-tei/src/main/java/org/mycore/tei/MCRTEISplitter.java

@@ -0,0 +1,166 @@
+/*
+ * This file is part of ***  M y C o R e  ***
+ * See http://www.mycore.de/ for details.
+ *
+ * MyCoRe is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation, either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * MyCoRe is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with MyCoRe.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+package org.mycore.tei;
+
+import java.util.ArrayList;
+import java.util.List;
+
+import org.jdom2.Attribute;
+import org.jdom2.Content;
+import org.jdom2.Document;
+import org.jdom2.Element;
+import org.jdom2.filter.Filters;
+import org.jdom2.xpath.XPathExpression;
+import org.jdom2.xpath.XPathFactory;
+import org.mycore.common.MCRConstants;
+
+/**
+ * Can split a TEI document into multiple documents based on the pb elements.
+ */
+public class MCRTEISplitter {
+
+    private final TeiFile original;
+
+    private Element copyTarget;
+
+    private List<TeiFile> splitDocumentList = new ArrayList<>();
+
+    private int size = -1;
+
+    public MCRTEISplitter(TeiFile original) {
+        this.original = original;
+    }
+
+    /**
+     * Checks if the document is splitable. A document is splitable if it contains pb elements.
+     * @return true if the document is splitable, false otherwise
+     */
+    public boolean isSplitable() {
+        XPathFactory xFactory = XPathFactory.instance();
+        XPathExpression<Element> expr = xFactory.compile("//tei:pb", Filters.element(), null,
+            MCRConstants.TEI_NAMESPACE);
+        List<Element> elementList = expr.evaluate(this.original.doc());
+        return !elementList.isEmpty();
+    }
+
+    /**
+     * Returns the estimated size of the split documents. This is the number of pb
+     * @return the estimated size of the split documents
+     */
+    public int getEstimatedSize() {
+        if (size == -1) {
+            XPathFactory xFactory = XPathFactory.instance();
+            XPathExpression<Element> expr = xFactory.compile("//tei:pb", Filters.element(), null,
+                MCRConstants.TEI_NAMESPACE);
+            List<Element> elementList = expr.evaluate(this.original.doc());
+            size = elementList.size();
+        }
+
+        return size;
+    }
+
+    private Stub copyAncestors(Element pbElement, String name) {
+        Element parent = pbElement;
+        Element lastClone = null;
+        Element firstClone = null;
+        while ((parent = parent.getParentElement()) != null) {
+            Element cloned = cloneElement(parent);
+            if (firstClone == null) {
+                firstClone = cloned;
+            }
+            if (lastClone != null) {
+                cloned.addContent(lastClone);
+            }
+            lastClone = cloned;
+        }
+
+        TeiFile teiFile = new TeiFile(name, new Document(lastClone));
+        this.splitDocumentList.add(teiFile);
+        return new Stub(firstClone, teiFile);
+    }
+
+    private void traverse(Element element) {
+        for (Content content : element.getContent()) {
+            if (content instanceof Element contentElement && contentElement.getName().equals("pb")) {
+                String facs = contentElement.getAttributeValue("facs");
+                if (facs.startsWith("images/")) {
+                    facs = facs.substring("images/".length());
+                }
+                Stub newStub = copyAncestors(element, facs);
+                copyTarget = newStub.newEl;
+                continue;
+            }
+
+            copyToNew(content);
+        }
+        copyTarget = copyTarget.getParentElement();
+    }
+
+    private void copyToNew(Content content) {
+        if (content instanceof Element elementContent) {
+            Element cloned = cloneElement(elementContent);
+            copyTarget.addContent(cloned);
+
+            copyTarget = cloned;
+            traverse(elementContent);
+        } else {
+            copyTarget.addContent(content.clone());
+        }
+
+    }
+
+    private Element cloneElement(Element elementContent) {
+        Element element = new Element(elementContent.getName(), elementContent.getNamespace());
+        elementContent.getAttributes()
+            .stream()
+            .map(Attribute::clone)
+            .forEach(element::setAttribute);
+
+        return element;
+    }
+
+    /**
+     * Splits the document into multiple documents based on the pb elements.
+     * @return a list of the split documents
+     */
+    public List<TeiFile> split() {
+        if (!splitDocumentList.isEmpty()) {
+            splitDocumentList = new ArrayList<>();
+        }
+
+        Element originalText = this.original.doc()
+            .getRootElement().getChild("text", MCRConstants.TEI_NAMESPACE);
+
+        Stub newStub = copyAncestors(originalText.getChildren().get(0), null);
+        copyTarget = newStub.newEl;
+
+        traverse(originalText);
+
+        return splitDocumentList;
+    }
+
+    /**
+     * Represents a TEI file.
+     */
+    public record TeiFile(String name, Document doc) {
+    }
+
+    private record Stub(Element newEl, TeiFile teiFile) {
+    }
+}

mycore-tei/src/main/java/org/mycore/tei/MCRTEIValidator.java

@@ -35,7 +35,9 @@
 
 /**
  * SAX Error handler for validating TEI XML
+ * @deprecated use the {@link org.mycore.datamodel.metadata.validator.MCREditorOutValidator} instead
  */
+@Deprecated
 public class MCRTEIValidator implements ErrorHandler {
 
     /**

mycore-tei/src/main/resources/META-INF/resources/tei/css/tei.css

@@ -0,0 +1,73 @@
+/*
+ * This file is part of ***  M y C o R e  ***
+ * See http://www.mycore.de/ for details.
+ *
+ * MyCoRe is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation, either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * MyCoRe is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with MyCoRe.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+.fileSection {
+    margin-top: 50px;
+}
+
+.teiDisplaySection {
+    padding: 15px;
+    background-color: #fff;
+    border: 1px solid #ddd;
+    border-radius: 4px;
+    margin-top: 15px;
+}
+
+.possibleMatch{
+    margin-top: 30px;
+}
+
+.possibleMatch a {
+    display: block;
+}
+
+.possibleMatch b {
+    margin-right: 5px;
+}
+
+.teiLine {
+    display: block;
+}
+
+.teiRow {
+    display: table-row;
+}
+
+.teiCell {
+    display: table-cell;
+    padding: 5px;
+}
+
+.teiTable {
+    display: table;
+}
+
+dt a {
+    vertical-align: super;
+    line-height: 2;
+}
+
+dt, dd {
+    display: inline;
+    font-weight: normal;
+    line-height: normal;
+}
+
+.popupTrigger {
+    cursor: pointer;
+}

mycore-tei/src/main/resources/META-INF/resources/tei/css/webfont-junicode.css

@@ -0,0 +1,52 @@
+/*
+ * This file is part of ***  M y C o R e  ***
+ * See http://www.mycore.de/ for details.
+ *
+ * MyCoRe is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation, either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * MyCoRe is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with MyCoRe.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+@font-face {
+    font-family: 'junicode';
+    src: url('../webfonts/junicode/Junicode.ttf.eot');
+    src: url('../webfonts/junicode/Junicode.ttf?iefix') format('eot'),
+    url('../webfonts/junicode/Junicode.ttf.woff') format('woff'),
+    url('../webfonts/junicode/Junicode.ttf.svg#webfont') format('svg');
+}
+
+.transcription {
+    font-family: 'junicode', serif;
+}
+
+a:link {
+    text-decoration: none;
+}
+
+a:visited {
+}
+
+a:hover {
+    color: #FFBF00;
+}
+
+a:active {
+    color: #FF0000;
+}
+
+.linebreak_none br {
+    display: none;
+}
+
+:target {
+    border: 3px solid #FFFF00;
+}