[SPARK-9157] [SQL] codegen substring

https://issues.apache.org/jira/browse/SPARK-9157

Author: Tarek Auel <tarek.auel@googlemail.com>

Closes #7534 from tarekauel/SPARK-9157 and squashes the following commits:

e65e3e9 [Tarek Auel] [SPARK-9157] indent fix
44e89f8 [Tarek Auel] [SPARK-9157] use EMPTY_UTF8
37d54c4 [Tarek Auel] Merge branch 'master' into SPARK-9157
60732ea [Tarek Auel] [SPARK-9157] created substringSQL in UTF8String
18c3576 [Tarek Auel] [SPARK-9157][SQL] remove slice pos
1a2e611 [Tarek Auel] [SPARK-9157][SQL] codegen substring

2 files changed, 31 insertions, 0 deletions

diff --git a/unsafe/src/main/java/org/apache/spark/unsafe/types/UTF8String.java b/unsafe/src/main/java/org/apache/spark/unsafe/types/UTF8String.java
index ed354f7f87..946d355f1f 100644
--- a/unsafe/src/main/java/org/apache/spark/unsafe/types/UTF8String.java
+++ b/unsafe/src/main/java/org/apache/spark/unsafe/types/UTF8String.java
@@ -165,6 +165,18 @@ public final class UTF8String implements Comparable<UTF8String>, Serializable {
     return fromBytes(bytes);
   }
 
+  public UTF8String substringSQL(int pos, int length) {
+    // Information regarding the pos calculation:
+    // Hive and SQL use one-based indexing for SUBSTR arguments but also accept zero and
+    // negative indices for start positions. If a start index i is greater than 0, it
+    // refers to element i-1 in the sequence. If a start index i is less than 0, it refers
+    // to the -ith element before the end of the sequence. If a start index i is 0, it
+    // refers to the first element.
+    int start = (pos > 0) ? pos -1 : ((pos < 0) ? numChars() + pos : 0);
+    int end = (length == Integer.MAX_VALUE) ? Integer.MAX_VALUE : start + length;
+    return substring(start, end);
+  }
+
   /**
    * Returns whether this contains `substring` or not.
    */
diff --git a/unsafe/src/test/java/org/apache/spark/unsafe/types/UTF8StringSuite.java b/unsafe/src/test/java/org/apache/spark/unsafe/types/UTF8StringSuite.java
index 1f5572c509..e2a5628ff4 100644
--- a/unsafe/src/test/java/org/apache/spark/unsafe/types/UTF8StringSuite.java
+++ b/unsafe/src/test/java/org/apache/spark/unsafe/types/UTF8StringSuite.java
@@ -273,6 +273,25 @@ public class UTF8StringSuite {
   }
 
   @Test
+  public void substringSQL() {
+    UTF8String e = fromString("example");
+    assertEquals(e.substringSQL(0, 2), fromString("ex"));
+    assertEquals(e.substringSQL(1, 2), fromString("ex"));
+    assertEquals(e.substringSQL(0, 7), fromString("example"));
+    assertEquals(e.substringSQL(1, 2), fromString("ex"));
+    assertEquals(e.substringSQL(0, 100), fromString("example"));
+    assertEquals(e.substringSQL(1, 100), fromString("example"));
+    assertEquals(e.substringSQL(2, 2), fromString("xa"));
+    assertEquals(e.substringSQL(1, 6), fromString("exampl"));
+    assertEquals(e.substringSQL(2, 100), fromString("xample"));
+    assertEquals(e.substringSQL(0, 0), fromString(""));
+    assertEquals(e.substringSQL(100, 4), EMPTY_UTF8);
+    assertEquals(e.substringSQL(0, Integer.MAX_VALUE), fromString("example"));
+    assertEquals(e.substringSQL(1, Integer.MAX_VALUE), fromString("example"));
+    assertEquals(e.substringSQL(2, Integer.MAX_VALUE), fromString("xample"));
+  }
+
+  @Test
   public void split() {
     assertTrue(Arrays.equals(fromString("ab,def,ghi").split(fromString(","), -1),
       new UTF8String[]{fromString("ab"), fromString("def"), fromString("ghi")}));