Better detection of CBOR

CBOR has a special header that is optional, if exists, allows for exact detection. Also, since we know which formats we support in ES, we can support the object major type case.
closes elastic#7640

pom.xml

@@ -234,28 +234,28 @@
         <dependency>
             <groupId>com.fasterxml.jackson.core</groupId>
             <artifactId>jackson-core</artifactId>
-            <version>2.4.2</version>
+            <version>2.4.4</version>
             <scope>compile</scope>
         </dependency>
 
         <dependency>
             <groupId>com.fasterxml.jackson.dataformat</groupId>
             <artifactId>jackson-dataformat-smile</artifactId>
-            <version>2.4.2</version>
+            <version>2.4.4</version>
             <scope>compile</scope>
         </dependency>
 
         <dependency>
             <groupId>com.fasterxml.jackson.dataformat</groupId>
             <artifactId>jackson-dataformat-yaml</artifactId>
-            <version>2.4.2</version>
+            <version>2.4.4</version>
             <scope>compile</scope>
         </dependency>
 
         <dependency>
             <groupId>com.fasterxml.jackson.dataformat</groupId>
             <artifactId>jackson-dataformat-cbor</artifactId>
-            <version>2.4.2</version>
+            <version>2.4.4</version>
             <scope>compile</scope>
         </dependency>
 

src/main/java/org/elasticsearch/common/xcontent/XContentFactory.java

@@ -208,14 +208,16 @@ public static XContentType xContentType(byte[] data) {
      * Guesses the content type based on the provided input stream.
      */
     public static XContentType xContentType(InputStream si) throws IOException {
-        int first = si.read();
-        if (first == -1) {
+        int iFirst = si.read();
+        if (iFirst == -1) {
             return null;
         }
-        int second = si.read();
-        if (second == -1) {
+        byte first = (byte) iFirst;
+        int iSecond = si.read();
+        if (iSecond == -1) {
             return null;
         }
+        byte second = (byte) iSecond;
         if (first == SmileConstants.HEADER_BYTE_1 && second == SmileConstants.HEADER_BYTE_2) {
             int third = si.read();
             if (third == SmileConstants.HEADER_BYTE_3) {
@@ -231,9 +233,27 @@ public static XContentType xContentType(InputStream si) throws IOException {
                 return XContentType.YAML;
             }
         }
-        if (first == (CBORConstants.BYTE_OBJECT_INDEFINITE & 0xff)){
+        // CBOR logic similar to CBORFactory#hasCBORFormat
+        if (first == CBORConstants.BYTE_OBJECT_INDEFINITE){
+            return XContentType.CBOR;
+        }
+        if (CBORConstants.hasMajorType(CBORConstants.MAJOR_TYPE_TAG, first)) {
+            // Actually, specific "self-describe tag" is a very good indicator
+            int iThird = si.read();
+            if (iThird == -1) {
+                return null;
+            }
+            byte third = (byte) iThird;
+            if (first == (byte) 0xD9 && second == (byte) 0xD9 && third == (byte) 0xF7) {
+                return XContentType.CBOR;
+            }
+        }
+        // for small objects, some encoders just encode as major type object, we can safely
+        // say its CBOR since it doesn't contradict SMILE or JSON, and its a last resort
+        if (CBORConstants.hasMajorType(CBORConstants.MAJOR_TYPE_OBJECT, first)) {
             return XContentType.CBOR;
         }
+
         for (int i = 2; i < GUESS_HEADER_LENGTH; i++) {
             int val = si.read();
             if (val == -1) {
@@ -279,9 +299,23 @@ public static XContentType xContentType(BytesReference bytes) {
         if (length > 2 && first == '-' && bytes.get(1) == '-' && bytes.get(2) == '-') {
             return XContentType.YAML;
         }
-        if (first == CBORConstants.BYTE_OBJECT_INDEFINITE){
+        // CBOR logic similar to CBORFactory#hasCBORFormat
+        if (first == CBORConstants.BYTE_OBJECT_INDEFINITE && length > 1){
             return XContentType.CBOR;
         }
+        if (CBORConstants.hasMajorType(CBORConstants.MAJOR_TYPE_TAG, first) && length > 2) {
+            // Actually, specific "self-describe tag" is a very good indicator
+            if (first == (byte) 0xD9 && bytes.get(1) == (byte) 0xD9 && bytes.get(2) == (byte) 0xF7) {
+                return XContentType.CBOR;
+            }
+        }
+        // for small objects, some encoders just encode as major type object, we can safely
+        // say its CBOR since it doesn't contradict SMILE or JSON, and its a last resort
+        if (CBORConstants.hasMajorType(CBORConstants.MAJOR_TYPE_OBJECT, first)) {
+            return XContentType.CBOR;
+        }
+
+        // a last chance for JSON
         for (int i = 0; i < length; i++) {
             if (bytes.get(i) == '{') {
                 return XContentType.JSON;

src/test/java/org/elasticsearch/common/xcontent/XContentFactoryTests.java

@@ -19,6 +19,8 @@
 
 package org.elasticsearch.common.xcontent;
 
+import com.fasterxml.jackson.dataformat.cbor.CBORConstants;
+import com.fasterxml.jackson.dataformat.smile.SmileConstants;
 import org.elasticsearch.common.bytes.BytesArray;
 import org.elasticsearch.common.io.stream.BytesStreamInput;
 import org.elasticsearch.test.ElasticsearchTestCase;
@@ -69,4 +71,21 @@ private void testGuessType(XContentType type) throws IOException {
             assertThat(XContentFactory.xContentType(builder.string()), equalTo(type));
         }
     }
+
+    public void testCBORBasedOnMajorObjectDetection() {
+        // for this {"foo"=> 5} perl encoder for example generates:
+        byte[] bytes = new byte[] {(byte) 0xA1, (byte) 0x43, (byte) 0x66, (byte) 6f, (byte) 6f, (byte) 5};
+        assertThat(XContentFactory.xContentType(bytes), equalTo(XContentType.CBOR));
+
+        // also make sure major type check doesn't collide with SMILE and JSON, just in case
+        assertThat(CBORConstants.hasMajorType(CBORConstants.MAJOR_TYPE_OBJECT, SmileConstants.HEADER_BYTE_1), equalTo(false));
+        assertThat(CBORConstants.hasMajorType(CBORConstants.MAJOR_TYPE_OBJECT, (byte) '{'), equalTo(false));
+        assertThat(CBORConstants.hasMajorType(CBORConstants.MAJOR_TYPE_OBJECT, (byte) ' '), equalTo(false));
+        assertThat(CBORConstants.hasMajorType(CBORConstants.MAJOR_TYPE_OBJECT, (byte) '-'), equalTo(false));
+    }
+
+    public void testCBORBasedOnMagicHeaderDetection() {
+        byte[] bytes = new byte[] {(byte) 0xd9, (byte) 0xd9, (byte) 0xf7};
+        assertThat(XContentFactory.xContentType(bytes), equalTo(XContentType.CBOR));
+    }
 }