Merge pull request #1106 from kermitt2/bugfix/sent-seg-ack-fund

Add missing sentence segmentation in funding and acknowledgement

.github/workflows/ci-build-unstable.yml

@@ -25,6 +25,12 @@ jobs:
       - name: Test with Gradle Jacoco and Coveralls
         run: ./gradlew test jacocoTestReport coveralls --no-daemon
 
+      - name: Publish Test Report
+        uses: mikepenz/action-junit-report@v4
+        if: success() || failure() # always run even if the previous step fails
+        with:
+          report_paths: '**/build/test-results/test/TEST-*.xml'
+
       - name: Coveralls GitHub Action
         uses: coverallsapp/github-action@v2
         with:

Dockerfile.crf

@@ -78,7 +78,7 @@ WORKDIR /opt/grobid
 
 COPY --from=builder /opt/grobid .
 
-ENV GROBID_SERVICE_OPTS "-Djava.library.path=grobid-home/lib/lin-64:grobid-home/lib/lin-64/jep --add-opens java.base/java.lang=ALL-UNNAMED"
+ENV GROBID_SERVICE_OPTS "-Djava.library.path=grobid-home/lib/lin-64:grobid-home/lib/lin-64/jep --add-opens java.base/java.lang=ALL-UNNAMED --add-opens java.base/sun.nio.ch=ALL-UNNAMED --add-opens java.base/java.io=ALL-UNNAMED"
 
 CMD ["./grobid-service/bin/grobid-service"]
 

build.gradle

@@ -111,7 +111,8 @@ subprojects {
         testImplementation 'org.easymock:easymock:5.1.0'
         testImplementation "org.powermock:powermock-api-easymock:2.0.7"
         testImplementation "org.powermock:powermock-module-junit4:2.0.7"
-        testImplementation "xmlunit:xmlunit:1.6"
+        testImplementation "org.xmlunit:xmlunit-matchers:2.10.0"
+        testImplementation "org.xmlunit:xmlunit-legacy:2.10.0"
         testImplementation "org.hamcrest:hamcrest-all:1.3"
         testImplementation 'org.jetbrains.kotlin:kotlin-test'
         testImplementation "io.mockk:mockk:1.13.9"

grobid-core/src/main/java/org/grobid/core/data/AnnotatedXMLElement.java

@@ -0,0 +1,35 @@
+package org.grobid.core.data;
+
+import nu.xom.Element;
+import org.grobid.core.utilities.OffsetPosition;
+
+/**
+ * This class represent an annotation in an XML node.
+ * The annotation is composed by two information: the XML Element node and the offset position
+ */
+public class AnnotatedXMLElement {
+
+    private OffsetPosition offsetPosition;
+    private Element annotationNode;
+
+    public AnnotatedXMLElement(Element annotationNode, OffsetPosition offsetPosition) {
+        this.annotationNode = annotationNode;
+        this.offsetPosition = offsetPosition;
+    }
+
+    public OffsetPosition getOffsetPosition() {
+        return offsetPosition;
+    }
+
+    public void setOffsetPosition(OffsetPosition offsetPosition) {
+        this.offsetPosition = offsetPosition;
+    }
+
+    public Element getAnnotationNode() {
+        return annotationNode;
+    }
+
+    public void setAnnotationNode(Element annotationNode) {
+        this.annotationNode = annotationNode;
+    }
+}

grobid-core/src/main/java/org/grobid/core/data/Funder.java

@@ -60,6 +60,8 @@ public class Funder {
         prefixFounders.put("NSF", "National Science Foundation");
         prefixFounders.put("NIH", "National Institutes of Health");
         prefixFounders.put("ERC", "European Research Council");
+        //Japanese government
+        prefixFounders.put("MEXT", "Ministry of Education, Culture, Sports, Science and Technology");
     }
 
     public Funder() {

grobid-core/src/main/java/org/grobid/core/data/FundingAcknowledgmentParse.java

@@ -0,0 +1,46 @@
+package org.grobid.core.data;
+
+import java.util.ArrayList;
+import java.util.List;
+
+/**
+ * This class represent the funding / acknowledgement statement
+ */
+public class FundingAcknowledgmentParse {
+    List<Funding> fundingList = new ArrayList<>();
+    List<Person> personList = new ArrayList<>();
+    List<Affiliation> affiliations = new ArrayList<>();
+//    List<Pair<OffsetPosition, Element> statementAnnotations = new ArrayList<>();
+
+    public List<Funding> getFundings() {
+        return fundingList;
+    }
+
+    public void setFundings(List<Funding> fundingList) {
+        this.fundingList = fundingList;
+    }
+
+    public List<Person> getPersons() {
+        return personList;
+    }
+
+    public void setPersons(List<Person> personList) {
+        this.personList = personList;
+    }
+
+    public List<Affiliation> getAffiliations() {
+        return affiliations;
+    }
+
+    public void setAffiliations(List<Affiliation> fundingBodies) {
+        this.affiliations = fundingBodies;
+    }
+
+//    public List<GrobidAnnotation> getStatementAnnotations() {
+//        return statementAnnotations;
+//    }
+
+//    public void setStatementAnnotations(List<GrobidAnnotation> statementAnnotations) {
+//        this.statementAnnotations = statementAnnotations;
+//    }
+}

grobid-core/src/main/java/org/grobid/core/engines/Engine.java

@@ -1,18 +1,10 @@
 package org.grobid.core.engines;
 
-import org.apache.commons.lang3.tuple.Pair;
+import nu.xom.Element;
 import org.apache.commons.lang3.tuple.MutablePair;
 import org.apache.commons.lang3.tuple.MutableTriple;
-
-import nu.xom.Element;
-
-import org.grobid.core.data.Affiliation;
-import org.grobid.core.data.BibDataSet;
-import org.grobid.core.data.BiblioItem;
-import org.grobid.core.data.ChemicalEntity;
-import org.grobid.core.data.PatentItem;
-import org.grobid.core.data.Person;
-import org.grobid.core.data.Funding;
+import org.apache.commons.lang3.tuple.Pair;
+import org.grobid.core.data.*;
 import org.grobid.core.document.Document;
 import org.grobid.core.document.DocumentSource;
 import org.grobid.core.engines.config.GrobidAnalysisConfig;
@@ -24,14 +16,15 @@
 import org.grobid.core.utilities.Utilities;
 import org.grobid.core.utilities.counters.CntManager;
 import org.grobid.core.utilities.counters.impl.CntManagerFactory;
-
 import org.grobid.core.utilities.crossref.CrossrefClient;
 import org.slf4j.Logger;
 import org.slf4j.LoggerFactory;
 
 import java.io.*;
-import java.util.*;
 import java.nio.charset.StandardCharsets;
+import java.util.ArrayList;
+import java.util.List;
+import java.util.Map;
 
 /**
  * Class for managing the extraction of bibliographical information from PDF
@@ -1184,7 +1177,7 @@ public String processFundingAcknowledgement(String text, GrobidAnalysisConfig co
                 result.append(localResult.getLeft().toXML()); 
 
         } catch (final Exception exp) {
-            throw new GrobidException("An exception occured while running Grobid funding-acknowledgement model.", exp);
+            throw new GrobidException("An exception occurred while running Grobid funding-acknowledgement model.", exp);
         }
 
         return result.toString();