Add solr field with number of times a field is used in a record #342:…

… add subfield count
pkiraly · Dec 19, 2024 · 24ba0e6 · 24ba0e6
1 parent 9e228f0
commit 24ba0e6
Show file tree

Hide file tree

Showing 2 changed files with 18 additions and 2 deletions.
diff --git a/src/main/java/de/gwdg/metadataqa/marc/cli/MarcToSolr.java b/src/main/java/de/gwdg/metadataqa/marc/cli/MarcToSolr.java
@@ -28,6 +28,7 @@
 import java.io.Serializable;
 import java.nio.file.Path;
 import java.text.DecimalFormat;
+import java.util.ArrayList;
 import java.util.Collections;
 import java.util.HashMap;
 import java.util.List;
@@ -207,6 +208,7 @@ private void indexFieldCounts(BibliographicRecord bibliographicRecord,
                                 SolrInputDocument document) {
     Counter<String> counter = new Counter<>();
     boolean isPica = bibliographicRecord.getSchemaType().equals(SchemaType.PICA);
+    Map<String, List<Integer>> subfields = new HashMap<>();
     for (DataField field : bibliographicRecord.getDatafields()) {
       String tag;
       if (field.getDefinition() != null) {
@@ -221,9 +223,15 @@ private void indexFieldCounts(BibliographicRecord bibliographicRecord,
         counter.count(safeTag);
 
       if (parameters.isIndexSubfieldCounts()) {
+        Counter<String> subfieldCounter = new Counter<>();
         for (MarcSubfield subfield : field.getSubfields()) {
           String safeSubfieldCode = DataFieldKeyGenerator.escape(subfield.getCode());
-          counter.count(safeTag + safeSubfieldCode);
+          subfieldCounter.count(safeTag + safeSubfieldCode);
+        }
+        for (Map.Entry<String, Integer> entry : subfieldCounter.entrySet()) {
+          if (!subfields.containsKey(entry.getKey()))
+            subfields.put(entry.getKey(), new ArrayList<>());
+          subfields.get(entry.getKey()).add(entry.getValue());
         }
       }
     }
@@ -232,6 +240,14 @@ private void indexFieldCounts(BibliographicRecord bibliographicRecord,
         "%s%s_count_i",
         parameters.getFieldPrefix(), entry.getKey()), entry.getValue());
     }
+
+    if (parameters.isIndexSubfieldCounts()) {
+      for (Map.Entry<String, List<Integer>> entry : subfields.entrySet()) {
+        document.addField(String.format(
+          "%s%s_count_is",
+          parameters.getFieldPrefix(), entry.getKey()), entry.getValue());
+      }
+    }
   }
 
   private String escape(String tag) {

diff --git a/src/main/java/de/gwdg/metadataqa/marc/cli/parameters/MarcToSolrParameters.java b/src/main/java/de/gwdg/metadataqa/marc/cli/parameters/MarcToSolrParameters.java
@@ -123,7 +123,7 @@ public boolean isIndexSubfieldCounts() {
   }
 
   public String getFieldPrefix() {
-    return fieldPrefix;
+    return fieldPrefix != null ? fieldPrefix : "";
   }
 
   public void setFieldPrefix(String fieldPrefix) {