Return correct term statistics whem a field is not found in a shard

If you ask for the term vectors of an artificial document with
term_statistics=true, but a shard does not have any terms of the doc's
field(s), it returns the doc's term vectors values as the shard-level
term statistics. This commit fixes that to return 0 for ttf and also
field-level aggregated statistics.

This closes elastic#21906

core/src/main/java/org/elasticsearch/action/termvectors/TermVectorsWriter.java

@@ -71,7 +71,7 @@ void setFields(Fields termVectorsByField, Set<String> selectedFields, EnumSet<Fl
 
             // if no terms found, take the retrieved term vector fields for stats
             if (topLevelTerms == null) {
-                topLevelTerms = fieldTermVector;
+                topLevelTerms = EMPTY_TERMS;
             }
 
             TermsEnum topLevelIterator = topLevelTerms.iterator();
@@ -292,4 +292,30 @@ private void writePotentiallyNegativeVLong(long value) throws IOException {
         // further...
         output.writeVLong(Math.max(0, value + 1));
     }
+
+    /** Implements an empty {@link Terms}. */
+    private static final Terms EMPTY_TERMS = new Terms() {
+        @Override public TermsEnum iterator() throws IOException { return EMPTY_TERMS_ENUM; }
+        @Override public long size() throws IOException { return 0; }
+        @Override public long getSumTotalTermFreq() throws IOException { return 0; }
+        @Override public long getSumDocFreq() throws IOException { return 0; }
+        @Override public int getDocCount() throws IOException { return 0; }
+        @Override public boolean hasFreqs() { return false; }
+        @Override public boolean hasOffsets() { return false; }
+        @Override public boolean hasPositions() { return false; }
+        @Override public boolean hasPayloads() { return false; }
+    };
+
+    /** Implements an empty {@link TermsEnum}. */
+    private static final TermsEnum EMPTY_TERMS_ENUM = new TermsEnum() {
+        @Override public BytesRef next() throws IOException { return null; }
+        @Override public SeekStatus seekCeil(BytesRef text) throws IOException { return SeekStatus.END; }
+        @Override public void seekExact(long ord) throws IOException { }
+        @Override public BytesRef term() throws IOException { return null; }
+        @Override public long ord() throws IOException { return 0; }
+        @Override public int docFreq() throws IOException { return 0; }
+        @Override public long totalTermFreq() throws IOException { return 0; }
+        @Override public PostingsEnum postings(PostingsEnum reuse, int flags) throws IOException { return null; }
+    };
+
 }

core/src/test/java/org/elasticsearch/action/termvectors/GetTermVectorsIT.java

@@ -848,6 +848,16 @@ public void testArtificialNoDoc() throws IOException {
                 .get();
         assertThat(resp.isExists(), equalTo(true));
         checkBrownFoxTermVector(resp.getFields(), "field1", false);
+
+        // Since the index is empty, all of artificial document's "term_statistics" should be 0/absent
+        Terms terms = resp.getFields().terms("field1");
+        assertEquals("sumDocFreq should be 0 for a non-existing field!", 0, terms.getSumDocFreq());
+        assertEquals("sumTotalTermFreq should be 0 for a non-existing field!", 0, terms.getSumTotalTermFreq());
+        TermsEnum termsEnum = terms.iterator(); // we're guaranteed to receive terms for that field
+        while (termsEnum.next() != null) {
+            String term = termsEnum.term().utf8ToString();
+            assertEquals("term [" + term + "] does not exist in the index; ttf should be 0!", 0, termsEnum.totalTermFreq());
+        }
     }
 
     public void testPerFieldAnalyzer() throws IOException {