add version comparator for StringComparator (#6745)

* add version comparator for StringComparator

* add more test case and docs

docs/content/querying/filters.md

@@ -276,7 +276,7 @@ greater than, less than, greater than or equal to, less than or equal to, and "b
 |upper|String|The upper bound for the filter|no|
 |lowerStrict|Boolean|Perform strict comparison on the lower bound (">" instead of ">=")|no, default: false|
 |upperStrict|Boolean|Perform strict comparison on the upper bound ("<" instead of "<=")|no, default: false|
-|ordering|String|Specifies the sorting order to use when comparing values against the bound. Can be one of the following values: "lexicographic", "alphanumeric", "numeric", "strlen". See [Sorting Orders](./sorting-orders.html) for more details.|no, default: "lexicographic"|
+|ordering|String|Specifies the sorting order to use when comparing values against the bound. Can be one of the following values: "lexicographic", "alphanumeric", "numeric", "strlen", "version". See [Sorting Orders](./sorting-orders.html) for more details.|no, default: "lexicographic"|
 |extractionFn|[Extraction function](#filtering-with-extraction-functions)| Extraction function to apply to the dimension|no|
   
 Bound filters support the use of extraction functions, see [Filtering with Extraction Functions](#filtering-with-extraction-functions) for details.

docs/content/querying/sorting-orders.md

@@ -47,3 +47,8 @@ When comparing two unparseable values (e.g., "hello" and "world"), this ordering
 
 ## Strlen
 Sorts values by the their string lengths. When there is a tie, this comparator falls back to using the String compareTo method.
+
+## Version
+Sorts values as versions, e.g.: "10.0 sorts after 9.0", "1.0.0-SNAPSHOT sorts after 1.0.0".
+
+See https://maven.apache.org/ref/3.6.0/maven-artifact/apidocs/org/apache/maven/artifact/versioning/ComparableVersion.html for more details on how this ordering sorts values.

pom.xml

@@ -749,6 +749,11 @@
                 <artifactId>caffeine</artifactId>
                 <version>${caffeine.version}</version>
             </dependency>
+            <dependency>
+                <groupId>org.apache.maven</groupId>
+                <artifactId>maven-artifact</artifactId>
+                <version>3.6.0</version>
+            </dependency>
 
             <dependency>
                 <groupId>org.apache.calcite</groupId>

processing/pom.xml

@@ -116,6 +116,11 @@
             <artifactId>checker</artifactId>
             <version>${checkerframework.version}</version>
         </dependency>
+        <dependency>
+            <groupId>org.apache.maven</groupId>
+            <artifactId>maven-artifact</artifactId>
+        </dependency>
+
 
         <!-- Tests -->
         <dependency>

processing/src/main/java/org/apache/druid/jackson/StringComparatorModule.java

@@ -36,7 +36,8 @@ public class StringComparatorModule extends SimpleModule
         new NamedType(StringComparators.LexicographicComparator.class, StringComparators.LEXICOGRAPHIC_NAME),
         new NamedType(StringComparators.AlphanumericComparator.class, StringComparators.ALPHANUMERIC_NAME),
         new NamedType(StringComparators.StrlenComparator.class, StringComparators.STRLEN_NAME),
-        new NamedType(StringComparators.NumericComparator.class, StringComparators.NUMERIC_NAME)
+        new NamedType(StringComparators.NumericComparator.class, StringComparators.NUMERIC_NAME),
+        new NamedType(StringComparators.VersionComparator.class, StringComparators.VERSION_NAME)
     );
   }
 

processing/src/main/java/org/apache/druid/query/ordering/StringComparator.java

@@ -39,6 +39,8 @@ public abstract class StringComparator implements Comparator<String>
         return StringComparators.STRLEN;
       case StringComparators.NUMERIC_NAME:
         return StringComparators.NUMERIC;
+      case StringComparators.VERSION_NAME:
+        return StringComparators.VERSION;
       default:
         throw new IAE("Unknown string comparator[%s]", type);
     }

processing/src/main/java/org/apache/druid/query/ordering/StringComparators.java

@@ -24,6 +24,7 @@ import com.google.common.primitives.Ints;
 import com.google.common.primitives.UnsignedBytes;
 import org.apache.druid.common.guava.GuavaUtils;
 import org.apache.druid.java.util.common.StringUtils;
+import org.apache.maven.artifact.versioning.DefaultArtifactVersion;
 
 import java.math.BigDecimal;
 import java.util.Comparator;
@@ -34,16 +35,19 @@ public class StringComparators
   public static final String ALPHANUMERIC_NAME = "alphanumeric";
   public static final String NUMERIC_NAME = "numeric";
   public static final String STRLEN_NAME = "strlen";
+  public static final String VERSION_NAME = "version";
 
   public static final StringComparator LEXICOGRAPHIC = new LexicographicComparator();
   public static final StringComparator ALPHANUMERIC = new AlphanumericComparator();
   public static final StringComparator NUMERIC = new NumericComparator();
   public static final StringComparator STRLEN = new StrlenComparator();
+  public static final StringComparator VERSION = new VersionComparator();
 
   public static final int LEXICOGRAPHIC_CACHE_ID = 0x01;
   public static final int ALPHANUMERIC_CACHE_ID = 0x02;
   public static final int NUMERIC_CACHE_ID = 0x03;
   public static final int STRLEN_CACHE_ID = 0x04;
+  public static final int VERSION_CACHE_ID = 0x05;
 
   public static class LexicographicComparator extends StringComparator
   {
@@ -416,4 +420,51 @@ public class StringComparators
       return new byte[]{(byte) NUMERIC_CACHE_ID};
     }
   }
+
+  public static class VersionComparator extends StringComparator
+  {
+    @Override
+    public int compare(String o1, String o2)
+    {
+      //noinspection StringEquality
+      if (o1 == o2) {
+        return 0;
+      }
+      if (o1 == null) {
+        return -1;
+      }
+      if (o2 == null) {
+        return 1;
+      }
+
+      DefaultArtifactVersion version1 = new DefaultArtifactVersion(o1);
+      DefaultArtifactVersion version2 = new DefaultArtifactVersion(o2);
+      return version1.compareTo(version2);
+    }
+
+    @Override
+    public String toString()
+    {
+      return StringComparators.VERSION_NAME;
+    }
+
+    @Override
+    public boolean equals(Object o)
+    {
+      if (this == o) {
+        return true;
+      }
+      if (o == null || getClass() != o.getClass()) {
+        return false;
+      }
+
+      return true;
+    }
+
+    @Override
+    public byte[] getCacheKey()
+    {
+      return new byte[]{(byte) VERSION_CACHE_ID};
+    }
+  }
 }

processing/src/test/java/org/apache/druid/query/ordering/StringComparatorsTest.java

@@ -142,6 +142,20 @@ public class StringComparatorsTest
     Assert.assertTrue(StringComparators.NUMERIC.compare("CAN'T PARSE THIS", "CAN'T TOUCH THIS") < 0);
   }
 
+  @Test
+  public void testVersionComparator()
+  {
+    commonTest(StringComparators.VERSION);
+
+    Assert.assertTrue(StringComparators.VERSION.compare("02", "002") == 0);
+    Assert.assertTrue(StringComparators.VERSION.compare("1.0", "2.0") < 0);
+    Assert.assertTrue(StringComparators.VERSION.compare("9.1", "10.0") < 0);
+    Assert.assertTrue(StringComparators.VERSION.compare("1.1.1", "2.0") < 0);
+    Assert.assertTrue(StringComparators.VERSION.compare("1.0-SNAPSHOT", "1.0") < 0);
+    Assert.assertTrue(StringComparators.VERSION.compare("2.0.1-xyz-1", "2.0.1-1-xyz") < 0);
+    Assert.assertTrue(StringComparators.VERSION.compare("1.0-SNAPSHOT", "1.0-Final") < 0);
+  }
+
   @Test
   public void testLexicographicComparatorSerdeTest() throws IOException
   {