Check positions on MultiPhraseQueries as well as phrase queries (#129326) (#129352)

Backport #129326 to 8.19

Author: thecoop

docs/changelog/129326.yaml

@@ -0,0 +1,6 @@
+pr: 129326
+summary: Check positions on `MultiPhraseQueries` as well as phrase queries
+area: Search
+type: bug
+issues:
+ - 123871

modules/analysis-common/src/yamlRestTest/resources/rest-api-spec/test/search.query/50_queries_with_synonyms.yml

@@ -200,3 +200,50 @@
                 query: quick brown
                 operator: and
   - match: { hits.total: 2 }
+---
+# Tests that this returns 400, and not 500 (#123871)
+"Test multi_match phrase with no positions":
+  - requires:
+      cluster_features: [ "search.multi.match.checks.positions" ]
+      reason: "This previously resulted in a 5xx error code"
+  - do:
+      indices.create:
+        index: test
+        body:
+          settings:
+            analysis:
+              filter:
+                syns:
+                  type: synonym
+                  synonyms: [ "quick,fast" ]
+              analyzer:
+                syns:
+                  tokenizer: standard
+                  filter: [ "syns" ]
+          mappings:
+            properties:
+              field1:
+                type: text
+                index_options: freqs
+                analyzer: syns
+
+  - do:
+      index:
+        index: test
+        id: "1"
+        body:
+          field1: the quick brown fox
+
+  - do:
+      catch: bad_request
+      search:
+        body:
+          query:
+            multi_match:
+              query: the fast brown
+              type: phrase
+              fields:
+                - field1
+  - match: { status: 400 }
+  - match: { error.root_cause.0.type: query_shard_exception }
+  - match: { error.root_cause.0.reason: "failed to create query: field:[field1] was indexed without position data; cannot run MultiPhraseQuery" }

modules/mapper-extras/src/main/java/org/elasticsearch/index/mapper/extras/SearchAsYouTypeFieldMapper.java

@@ -362,7 +362,7 @@ public Query prefixQuery(
 
         private void checkForPositions() {
             if (getTextSearchInfo().hasPositions() == false) {
-                throw new IllegalStateException("field:[" + name() + "] was indexed without position data; cannot run PhraseQuery");
+                throw new IllegalArgumentException("field:[" + name() + "] was indexed without position data; cannot run PhraseQuery");
             }
         }
 

server/src/internalClusterTest/java/org/elasticsearch/search/query/QueryStringIT.java

@@ -189,9 +189,9 @@ public void testPhraseQueryOnFieldWithNoPositions() throws Exception {
             Exception.class,
             prepareSearch("test").setQuery(queryStringQuery("f4:\"eggplant parmesan\"").lenient(false))
         );
-        IllegalStateException ise = (IllegalStateException) ExceptionsHelper.unwrap(exc, IllegalStateException.class);
-        assertNotNull(ise);
-        assertThat(ise.getMessage(), containsString("field:[f4] was indexed without position data; cannot run PhraseQuery"));
+        IllegalArgumentException iae = (IllegalArgumentException) ExceptionsHelper.unwrap(exc, IllegalArgumentException.class);
+        assertNotNull(iae);
+        assertThat(iae.getMessage(), containsString("field:[f4] was indexed without position data; cannot run PhraseQuery"));
     }
 
     public void testBooleanStrictQuery() throws Exception {

server/src/main/java/org/elasticsearch/index/mapper/TextFieldMapper.java

@@ -897,17 +897,19 @@ public IntervalsSource rangeIntervals(
             return Intervals.range(lowerTerm, upperTerm, includeLower, includeUpper, IndexSearcher.getMaxClauseCount());
         }
 
-        private void checkForPositions() {
+        private void checkForPositions(boolean multi) {
             if (getTextSearchInfo().hasPositions() == false) {
-                throw new IllegalStateException("field:[" + name() + "] was indexed without position data; cannot run PhraseQuery");
+                throw new IllegalArgumentException(
+                    "field:[" + name() + "] was indexed without position data; cannot run " + (multi ? "MultiPhraseQuery" : "PhraseQuery")
+                );
             }
         }
 
         @Override
         public Query phraseQuery(TokenStream stream, int slop, boolean enablePosIncrements, SearchExecutionContext context)
             throws IOException {
             String field = name();
-            checkForPositions();
+            checkForPositions(false);
             // we can't use the index_phrases shortcut with slop, if there are gaps in the stream,
             // or if the incoming token stream is the output of a token graph due to
             // https://issues.apache.org/jira/browse/LUCENE-8916
@@ -942,6 +944,7 @@ public Query phraseQuery(TokenStream stream, int slop, boolean enablePosIncremen
         public Query multiPhraseQuery(TokenStream stream, int slop, boolean enablePositionIncrements, SearchExecutionContext context)
             throws IOException {
             String field = name();
+            checkForPositions(true);
             if (indexPhrases && slop == 0 && hasGaps(stream) == false) {
                 stream = new FixedShingleFilter(stream, 2);
                 field = field + FAST_PHRASE_SUFFIX;
@@ -962,7 +965,7 @@ private static int countTokens(TokenStream ts) throws IOException {
         @Override
         public Query phrasePrefixQuery(TokenStream stream, int slop, int maxExpansions, SearchExecutionContext context) throws IOException {
             if (countTokens(stream) > 1) {
-                checkForPositions();
+                checkForPositions(false);
             }
             return analyzePhrasePrefix(stream, slop, maxExpansions);
         }

server/src/main/java/org/elasticsearch/search/SearchFeatures.java

@@ -26,9 +26,15 @@ public Set<NodeFeature> getFeatures() {
         "search.completion_field.duplicate.support"
     );
     public static final NodeFeature INT_SORT_FOR_INT_SHORT_BYTE_FIELDS = new NodeFeature("search.sort.int_sort_for_int_short_byte_fields");
+    static final NodeFeature MULTI_MATCH_CHECKS_POSITIONS = new NodeFeature("search.multi.match.checks.positions");
 
     @Override
     public Set<NodeFeature> getTestFeatures() {
-        return Set.of(RETRIEVER_RESCORER_ENABLED, COMPLETION_FIELD_SUPPORTS_DUPLICATE_SUGGESTIONS, INT_SORT_FOR_INT_SHORT_BYTE_FIELDS);
+        return Set.of(
+            RETRIEVER_RESCORER_ENABLED,
+            COMPLETION_FIELD_SUPPORTS_DUPLICATE_SUGGESTIONS,
+            INT_SORT_FOR_INT_SHORT_BYTE_FIELDS,
+            MULTI_MATCH_CHECKS_POSITIONS
+        );
     }
 }