excludeBrackets

DOREMUS-ANR · Jan 27, 2020 · d60ef6a · d60ef6a
1 parent cf992cf
commit d60ef6a
Show file tree

Hide file tree

Showing 6 changed files with 74 additions and 17 deletions.
diff --git a/build.gradle b/build.gradle
@@ -1,5 +1,5 @@
 group 'org.doremus'
-version '0.4'
+version '0.5'
 
 apply plugin: 'java'
 apply plugin: 'application'

diff --git a/src/main/java/org/doremus/string2vocabulary/MODS.java b/src/main/java/org/doremus/string2vocabulary/MODS.java
@@ -24,22 +24,22 @@ public MODS(String name, Model model) {
   }
 
   @Override
-  public Resource findConcept(String text, boolean strict) {
+  public Resource findConcept(String text, boolean strict, boolean excludeBrackets) {
     return findModsResource(text, null);
   }
 
   public Resource findModsResource(String identifier, List<String> subjects) {
     if (identifier == null || identifier.isEmpty()) return null;
 
     String modsSearch =
-            "prefix modsrdf: <http://www.loc.gov/standards/mods/rdf/v1/#>\n" +
-                    "select distinct ?cat where {\n" +
-                    "  { ?cat modsrdf:identifier ?id}\n" +
-                    "  UNION {\n" +
-                    "    ?cat modsrdf:identifierGroup / modsrdf:identifierGroupValue ?id\n" +
-                    "  }\n" +
-                    "  FILTER (lcase(str(?id)) = \"" + identifier.toLowerCase() + "\")\n" +
-                    "}";
+      "prefix modsrdf: <http://www.loc.gov/standards/mods/rdf/v1/#>\n" +
+        "select distinct ?cat where {\n" +
+        "  { ?cat modsrdf:identifier ?id}\n" +
+        "  UNION {\n" +
+        "    ?cat modsrdf:identifierGroup / modsrdf:identifierGroupValue ?id\n" +
+        "  }\n" +
+        "  FILTER (lcase(str(?id)) = \"" + identifier.toLowerCase() + "\")\n" +
+        "}";
 
     // search all catalogs with that identifier
     QueryExecution qexec = QueryExecutionFactory.create(modsSearch, vocabulary);

diff --git a/src/main/java/org/doremus/string2vocabulary/SKOSVocabulary.java b/src/main/java/org/doremus/string2vocabulary/SKOSVocabulary.java
@@ -11,6 +11,7 @@
 
 public class SKOSVocabulary extends Vocabulary {
   private Map<String, List<Resource>> substitutionMap;
+  private Map<String, List<Resource>> substitutionMapNoBrackets;
 
   public SKOSVocabulary(String name, Model model) {
     super(name, model);
@@ -19,6 +20,7 @@ public SKOSVocabulary(String name, Model model) {
 
     // Build a map
     substitutionMap = new HashMap<>();
+    substitutionMapNoBrackets = new HashMap<>();
 
     // for each concept
     StmtIterator conceptIter =
@@ -37,40 +39,55 @@ public SKOSVocabulary(String name, Model model) {
       while (labelIterator.hasNext()) {
         Literal nx = labelIterator.nextStatement().getLiteral();
         String value = norm(nx.getLexicalForm());
+        String valueNb = normNb(nx.getLexicalForm());
         String lang = nx.getLanguage();
-        if (lang != null && !lang.isEmpty()) value += "@" + nx.getLanguage();
+        if (lang != null && !lang.isEmpty()){
+          value += "@" + nx.getLanguage();
+          valueNb += "@" + nx.getLanguage();
+        }
 
         // get the list or create a new one
         List<Resource> ls = substitutionMap.computeIfAbsent(value, k -> new ArrayList<>());
+        List<Resource> lsNb = substitutionMapNoBrackets.computeIfAbsent(valueNb, k -> new ArrayList<>());
         // add it to the list
         ls.add(resource);
+        lsNb.add(resource);
       }
 
       labelIterator = resource.listProperties(SKOS.altLabel);
       //for each label
       while (labelIterator.hasNext()) {
         Literal nx = labelIterator.nextStatement().getLiteral();
         String value = norm(nx.getLexicalForm());
+        String valueNb = normNb(nx.getLexicalForm());
         String lang = nx.getLanguage();
-        if (lang != null && !lang.isEmpty()) value += "@" + nx.getLanguage();
+        if (lang != null && !lang.isEmpty()){
+          value += "@" + nx.getLanguage();
+          valueNb += "@" + nx.getLanguage();
+        }
 
         // get the list or create a new one
         List<Resource> ls = substitutionMap.computeIfAbsent(value, k -> new ArrayList<>());
+        List<Resource> lsNb = substitutionMapNoBrackets.computeIfAbsent(valueNb, k -> new ArrayList<>());
         // add it to the list
         ls.add(resource);
+        lsNb.add(resource);
       }
     }
   }
 
 
   @Override
-  public Resource findConcept(String text, boolean strict) {
-    for (Map.Entry<String, List<Resource>> entry : substitutionMap.entrySet()) {
+  public Resource findConcept(String text, boolean strict, boolean excludeBrackets) {
+    String textOnly = text.replaceAll("@[a-z]{2,3}$", "");
+
+    Map<String, List<Resource>> map = excludeBrackets ? substitutionMapNoBrackets : substitutionMap;
+    for (Map.Entry<String, List<Resource>> entry : map.entrySet()) {
       String key = entry.getKey();
       String keyPlain = key.replaceAll("@[a-z]{2,3}$", "");
 
       boolean textLangMatch = text.equalsIgnoreCase(key);
-      boolean textOnlyMatch = !strict && text.replaceAll("@[a-z]{2,3}$", "").equalsIgnoreCase(keyPlain);
+      boolean textOnlyMatch = !strict && textOnly.equalsIgnoreCase(keyPlain);
 
       if (textLangMatch || textOnlyMatch) {
         List<Resource> matches = entry.getValue();

diff --git a/src/main/java/org/doremus/string2vocabulary/Vocabulary.java b/src/main/java/org/doremus/string2vocabulary/Vocabulary.java
@@ -57,7 +57,11 @@ public Resource getConcept(String code) {
     else return null;
   }
 
-  public abstract Resource findConcept(String text, boolean strict);
+  public Resource findConcept(String text, boolean strict) {
+    return findConcept(text, strict, false);
+  }
+
+  public abstract Resource findConcept(String text, boolean strict, boolean excludeBrackets);
 
 
   @Override
@@ -108,4 +112,10 @@ protected static String norm(String input) {
     // lowercase
     return seed.toLowerCase();
   }
+
+  protected static String normNb(String input) {
+    // remove brackets
+    input = input.replaceAll("\\([^)]+\\)", "").trim();
+    return norm(input);
+  }
 }
diff --git a/src/test/java/org/doremus/string2vocabulary/ModuleTest.java b/src/test/java/org/doremus/string2vocabulary/ModuleTest.java
@@ -1,10 +1,12 @@
 package org.doremus.string2vocabulary;
 
 import org.apache.jena.rdf.model.Model;
+import org.apache.jena.rdf.model.Resource;
 import org.apache.jena.riot.RDFDataMgr;
 import org.junit.Assert;
 import org.junit.Test;
 
+import java.io.File;
 import java.io.IOException;
 import java.io.StringWriter;
 
@@ -31,11 +33,26 @@ public void string2uri() {
     }
   }
 
+  @Test
+  public void matchNoBrackets() {
+    ClassLoader classLoader = getClass().getClassLoader();
+    String vocabularyFolder = classLoader.getResource("vocabulary").getPath();
+
+
+    Vocabulary v = Vocabulary.fromFile(new File(vocabularyFolder + "/test.ttl"));
+
+    Resource brackMatch = v.findConcept("test@en", true);
+    Resource noBrackMatch = v.findConcept("test@en", true, true);
+    System.out.println(noBrackMatch);
+    Assert.assertNull(brackMatch);
+    Assert.assertNotNull(noBrackMatch);
+  }
+
   private String toTtlString(Model m) {
     StringWriter sw = new StringWriter();
     m.write(sw, syntax);
     return sw.toString();
   }
 
 
-}
+}
diff --git a/src/test/resources/vocabulary/test.ttl b/src/test/resources/vocabulary/test.ttl
@@ -0,0 +1,13 @@
+@prefix dct:  <http://purl.org/dc/terms/> .
+@prefix rdf:  <http://www.w3.org/1999/02/22-rdf-syntax-ns#> .
+@prefix owl:  <http://www.w3.org/2002/07/owl#> .
+@prefix skos: <http://www.w3.org/2004/02/skos/core#> .
+@prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#> .
+
+<http://example.org/vocabulary/>
+    a            skos:ConceptScheme .
+
+<http://example.org/vocabulary/brackets>
+    a                 skos:Concept ;
+    skos:prefLabel    "test (to be ignored) "@en ;
+    skos:topConceptOf <http://example.org/vocabulary/> .