Relax limit on max string size in CBOR, Smile, YAML (elastic#103930)

Remove the rough limit on string length from Jackson 2.15. The limit was already relaxed for JSON in elastic#96031, this extends that change to other XContent types.

Refs: elastic#96031
Fixes: elastic#104009
(cherry picked from commit a359b1f)

Author: stu-elastic

libs/x-content/impl/src/main/java/org/elasticsearch/xcontent/provider/XContentImplUtils.java

@@ -0,0 +1,21 @@
+/*
+ * Copyright Elasticsearch B.V. and/or licensed to Elasticsearch B.V. under one
+ * or more contributor license agreements. Licensed under the Elastic License
+ * 2.0 and the Server Side Public License, v 1; you may not use this file except
+ * in compliance with, at your election, the Elastic License 2.0 or the Server
+ * Side Public License, v 1.
+ */
+
+package org.elasticsearch.xcontent.provider;
+
+import com.fasterxml.jackson.core.JsonFactory;
+import com.fasterxml.jackson.core.StreamReadConstraints;
+import com.fasterxml.jackson.core.TSFBuilder;
+
+public class XContentImplUtils {
+    public static <F extends JsonFactory, B extends TSFBuilder<F, B>> F configure(TSFBuilder<F, B> builder) {
+        // jackson 2.15 introduced a max string length. We have other limits in place to constrain max doc size,
+        // so here we set to max value (2GiB) so as not to constrain further than those existing limits.
+        return builder.streamReadConstraints(StreamReadConstraints.builder().maxStringLength(Integer.MAX_VALUE).build()).build();
+    }
+}

libs/x-content/impl/src/main/java/org/elasticsearch/xcontent/provider/cbor/CborXContentImpl.java

@@ -21,6 +21,7 @@
 import org.elasticsearch.xcontent.XContentParser;
 import org.elasticsearch.xcontent.XContentParserConfiguration;
 import org.elasticsearch.xcontent.XContentType;
+import org.elasticsearch.xcontent.provider.XContentImplUtils;
 
 import java.io.IOException;
 import java.io.InputStream;
@@ -45,7 +46,7 @@ public static XContent cborXContent() {
     }
 
     static {
-        cborFactory = new CBORFactory();
+        cborFactory = XContentImplUtils.configure(CBORFactory.builder());
         cborFactory.configure(CBORFactory.Feature.FAIL_ON_SYMBOL_HASH_OVERFLOW, false); // this trips on many mappings now...
         // Do not automatically close unclosed objects/arrays in com.fasterxml.jackson.dataformat.cbor.CBORGenerator#close() method
         cborFactory.configure(JsonGenerator.Feature.AUTO_CLOSE_JSON_CONTENT, false);

libs/x-content/impl/src/main/java/org/elasticsearch/xcontent/provider/json/JsonXContentImpl.java

@@ -13,14 +13,14 @@
 import com.fasterxml.jackson.core.JsonFactoryBuilder;
 import com.fasterxml.jackson.core.JsonGenerator;
 import com.fasterxml.jackson.core.JsonParser;
-import com.fasterxml.jackson.core.StreamReadConstraints;
 
 import org.elasticsearch.xcontent.XContent;
 import org.elasticsearch.xcontent.XContentBuilder;
 import org.elasticsearch.xcontent.XContentGenerator;
 import org.elasticsearch.xcontent.XContentParser;
 import org.elasticsearch.xcontent.XContentParserConfiguration;
 import org.elasticsearch.xcontent.XContentType;
+import org.elasticsearch.xcontent.provider.XContentImplUtils;
 
 import java.io.IOException;
 import java.io.InputStream;
@@ -46,12 +46,7 @@ public static final XContent jsonXContent() {
     }
 
     static {
-        var builder = new JsonFactoryBuilder();
-        // jackson 2.15 introduced a max string length. We have other limits in place to constrain max doc size,
-        // so here we set to max value (2GiB) so as not to constrain further than those existing limits.
-        builder.streamReadConstraints(StreamReadConstraints.builder().maxStringLength(Integer.MAX_VALUE).build());
-
-        jsonFactory = builder.build();
+        jsonFactory = XContentImplUtils.configure(new JsonFactoryBuilder());
         jsonFactory.configure(JsonGenerator.Feature.QUOTE_FIELD_NAMES, true);
         jsonFactory.configure(JsonParser.Feature.ALLOW_COMMENTS, true);
         jsonFactory.configure(JsonFactory.Feature.FAIL_ON_SYMBOL_HASH_OVERFLOW, false); // this trips on many mappings now...

libs/x-content/impl/src/main/java/org/elasticsearch/xcontent/provider/smile/SmileXContentImpl.java

@@ -21,6 +21,7 @@
 import org.elasticsearch.xcontent.XContentParser;
 import org.elasticsearch.xcontent.XContentParserConfiguration;
 import org.elasticsearch.xcontent.XContentType;
+import org.elasticsearch.xcontent.provider.XContentImplUtils;
 
 import java.io.IOException;
 import java.io.InputStream;
@@ -45,7 +46,7 @@ public static XContent smileXContent() {
     }
 
     static {
-        smileFactory = new SmileFactory();
+        smileFactory = XContentImplUtils.configure(SmileFactory.builder());
         // for now, this is an overhead, might make sense for web sockets
         smileFactory.configure(SmileGenerator.Feature.ENCODE_BINARY_AS_7BIT, false);
         smileFactory.configure(SmileFactory.Feature.FAIL_ON_SYMBOL_HASH_OVERFLOW, false); // this trips on many mappings now...

libs/x-content/impl/src/main/java/org/elasticsearch/xcontent/provider/yaml/YamlXContentImpl.java

@@ -11,13 +11,15 @@
 import com.fasterxml.jackson.core.JsonEncoding;
 import com.fasterxml.jackson.core.JsonParser;
 import com.fasterxml.jackson.dataformat.yaml.YAMLFactory;
+import com.fasterxml.jackson.dataformat.yaml.YAMLParser;
 
 import org.elasticsearch.xcontent.XContent;
 import org.elasticsearch.xcontent.XContentBuilder;
 import org.elasticsearch.xcontent.XContentGenerator;
 import org.elasticsearch.xcontent.XContentParser;
 import org.elasticsearch.xcontent.XContentParserConfiguration;
 import org.elasticsearch.xcontent.XContentType;
+import org.elasticsearch.xcontent.provider.XContentImplUtils;
 
 import java.io.IOException;
 import java.io.InputStream;
@@ -42,7 +44,10 @@ public static XContent yamlXContent() {
     }
 
     static {
-        yamlFactory = new YAMLFactory();
+        yamlFactory = XContentImplUtils.configure(YAMLFactory.builder());
+        // YAMLFactory.builder() differs from new YAMLFactory() in that builder() does not set the default yaml parser feature flags.
+        // So set the only default feature flag, EMPTY_STRING_AS_NULL, here.
+        yamlFactory.configure(YAMLParser.Feature.EMPTY_STRING_AS_NULL, true);
         yamlFactory.configure(JsonParser.Feature.STRICT_DUPLICATE_DETECTION, true);
         yamlFactory.configure(JsonParser.Feature.USE_FAST_DOUBLE_PARSER, true);
         yamlXContent = new YamlXContentImpl();