Fix the "highlight.max_analyzer_offset" request parameter with "plain…

…" highlighter (opensearch-project#10919) (cherry picked from commit 5d327a2)
drunken-monkey · Feb 17, 2024 · 02b6504 · 02b6504
1 parent 3b47fa3
commit 02b6504
Show file tree

Hide file tree

Showing 3 changed files with 28 additions and 1 deletion.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -160,6 +160,7 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 - Prevent setting remote_snapshot store type on index creation ([#11867](https://github.com/opensearch-project/OpenSearch/pull/11867))
 - [BUG] Fix remote shards balancer when filtering throttled nodes ([#11724](https://github.com/opensearch-project/OpenSearch/pull/11724))
 - [Bug] Check phase name before SearchRequestOperationsListener onPhaseStart ([#12094](https://github.com/opensearch-project/OpenSearch/pull/12094))
+- Fix the "highlight.max_analyzer_offset" request parameter with "plain" highlighter ([#10919](https://github.com/opensearch-project/OpenSearch/pull/10919))
 
 ### Security
 

diff --git a/...pi-spec/src/main/resources/rest-api-spec/test/search.highlight/30_max_analyzed_offset.yml b/...pi-spec/src/main/resources/rest-api-spec/test/search.highlight/30_max_analyzed_offset.yml
@@ -78,3 +78,15 @@ setup:
           index: test1
           body: {"query" : {"match" : {"field2" : "fox"}}, "highlight" : {"type" : "plain", "fields" : {"field2" : {}}}}
   - match: { error.root_cause.0.type: "illegal_argument_exception" }
+
+---
+"Plain highlighter on a field WITHOUT OFFSETS using max_analyzer_offset should SUCCEED":
+  - skip:
+      version: " - 2.1.99"
+      reason: only starting supporting the parameter max_analyzer_offset on version 2.2
+  - do:
+      search:
+        rest_total_hits_as_int: true
+        index: test1
+        body: {"query" : {"match" : {"field1" : "quick"}}, "highlight" : {"type" : "plain", "fields" : {"field1" : {"max_analyzer_offset": 10}}}}
+  - match: {hits.hits.0.highlight.field1.0: "The <em>quick</em> "}
diff --git a/server/src/main/java/org/opensearch/search/fetch/subphase/highlight/PlainHighlighter.java b/server/src/main/java/org/opensearch/search/fetch/subphase/highlight/PlainHighlighter.java
@@ -123,13 +123,27 @@ public HighlightField highlight(FieldHighlightContext fieldContext) throws IOExc
         List<Object> textsToHighlight;
         Analyzer analyzer = context.mapperService().documentMapper().mappers().indexAnalyzer();
         final int maxAnalyzedOffset = context.getIndexSettings().getHighlightMaxAnalyzedOffset();
+        final Integer fieldMaxAnalyzedOffset = field.fieldOptions().maxAnalyzerOffset();
+        if (fieldMaxAnalyzedOffset != null && fieldMaxAnalyzedOffset > maxAnalyzedOffset) {
+            throw new IllegalArgumentException(
+                "max_analyzer_offset has exceeded ["
+                    + maxAnalyzedOffset
+                    + "] - maximum allowed to be analyzed for highlighting. "
+                    + "This maximum can be set by changing the ["
+                    + IndexSettings.MAX_ANALYZED_OFFSET_SETTING.getKey()
+                    + "] index level setting. "
+                    + "For large texts, indexing with offsets or term vectors is recommended!"
+            );
+        }
 
         textsToHighlight = HighlightUtils.loadFieldValues(fieldType, context.getQueryShardContext(), hitContext, fieldContext.forceSource);
 
         for (Object textToHighlight : textsToHighlight) {
             String text = convertFieldValue(fieldType, textToHighlight);
             int textLength = text.length();
-            if (textLength > maxAnalyzedOffset) {
+            if (fieldMaxAnalyzedOffset != null && textLength > fieldMaxAnalyzedOffset) {
+                text = text.substring(0, fieldMaxAnalyzedOffset);
+            } else if (textLength > maxAnalyzedOffset) {
                 throw new IllegalArgumentException(
                     "The length of ["
                         + fieldContext.fieldName