Jobergum/bolding control

vespa-cloud/cord-19-search/src/main/application/services.xml

@@ -14,6 +14,7 @@
           </chain>
           <chain id="default" inherits="vespa">
             <searcher id="ai.vespa.example.cord19.searcher.RelatedArticlesByNNSearcher" bundle="cord-19"/>
+            <searcher id="ai.vespa.example.cord19.searcher.BoldingSearcher" bundle="cord-19"/>
           </chain>
 
         </search>

vespa-cloud/cord-19-search/src/main/java/ai/vespa/example/cord19/searcher/BoldingSearcher.java

@@ -0,0 +1,42 @@
+package ai.vespa.example.cord19.searcher;
+import com.yahoo.prelude.query.CompositeItem;
+import com.yahoo.prelude.query.Item;
+import com.yahoo.prelude.query.PhraseItem;
+import com.yahoo.prelude.query.WordItem;
+import com.yahoo.search.Query;
+import com.yahoo.search.Result;
+import com.yahoo.search.Searcher;
+import com.yahoo.search.searchchain.Execution;
+
+/**
+ * This searcher traverse the query tree and looks for known stopwords, the stopwords are annotated with filter which will avoid
+ * bolding them in the search result
+ */
+
+public class BoldingSearcher extends Searcher {
+
+
+
+    @Override
+    public Result search(Query query, Execution execution) {
+        setFilterForStopWords(query.getModel().getQueryTree().getRoot());
+        return execution.search(query);
+    }
+
+    private void setFilterForStopWords(Item item) {
+        if (item instanceof WordItem) {
+            String word = ((WordItem)item).getWord();
+            word = word.toLowerCase();
+            if (RelatedArticlesByWeakAndSearcher.stopwords.contains(word) ) {
+                item.setFilter(true);
+            }
+        } else if (item instanceof PhraseItem) {
+            return;
+        }
+        else if (item instanceof CompositeItem) {
+            CompositeItem cItem = (CompositeItem)item;
+            for (Item i : cItem.items())
+                setFilterForStopWords(i);
+        }
+    }
+}

vespa-cloud/cord-19-search/src/main/java/ai/vespa/example/cord19/searcher/RelatedArticlesByWeakAndSearcher.java

@@ -97,7 +97,7 @@ private void addWeakAndItem(Article article, boolean includeAbstract, Query quer
         query.getModel().getQueryTree().setRoot(andItem);
     }
 
-    private static Set<String> stopwords = Set.of(
+    protected static Set<String> stopwords = Set.of(
                 "i", "me", "my", "myself", "we", "our", "ours", "ourselves",
                 "you", "your", "yours", "yourself", "yourselves", "he", "him", "his", "himself", "she", "her", "hers",
                 "herself", "it", "its", "itself", "they", "them", "their", "theirs", "themselves", "what", "which", "who", "whom",

vespa-cloud/cord-19-search/src/test/java/ai/vespa/example/cord19/searcher/BoldingSearcherTest.java

@@ -0,0 +1,34 @@
+package ai.vespa.example.cord19.searcher;
+
+import com.yahoo.component.chain.Chain;
+import com.yahoo.search.Query;
+import com.yahoo.search.Result;
+import com.yahoo.search.Searcher;
+import com.yahoo.search.searchchain.Execution;
+import org.junit.jupiter.api.Test;
+import static org.junit.jupiter.api.Assertions.assertEquals;
+
+public class BoldingSearcherTest {
+
+    @Test
+    public void testThatStopWordIsAnnotedAsFilter(){
+        Query input = new Query("?query=basic+reproduction+numbers+For+covid-19+IN+%22south+korea%22&type=any");
+        Result result = execute(input, new BoldingSearcher());
+        assertEquals("OR basic reproduction numbers |For (AND covid 19) |IN \"south korea\"",
+                result.getQuery().getModel().getQueryTree().toString());
+
+    }
+    @Test
+    public void testThatPhraseUntouched(){
+        Query input = new Query("?query=temperature+%22impact+on%22+viral+transmission&type=any");
+        Result result = execute(input, new BoldingSearcher());
+        assertEquals("OR temperature \"impact on\" viral transmission",
+                result.getQuery().getModel().getQueryTree().toString());
+
+    }
+
+    private Result execute(Query query, Searcher... searcher) {
+        Execution execution = new Execution(new Chain<>(searcher), Execution.Context.createContextStub());
+        return execution.search(query);
+    }
+}