From 4d01aa46489bb2e5f7c143d5e58981b3eb2f4a7d Mon Sep 17 00:00:00 2001
From: Yuming Wang <wgyumg@gmail.com>
Date: Tue, 13 Jun 2017 23:39:06 -0700
Subject: [PATCH] [SPARK-20754][SQL][FOLLOWUP] Add Function Alias For
MOD/POSITION.
## What changes were proposed in this pull request?
https://github.com/apache/spark/pull/18106 Support TRUNC (number), We should also add function alias for `MOD `and `POSITION`.
`POSITION(substr IN str) `is a synonym for `LOCATE(substr,str)`. same as MySQL: https://dev.mysql.com/doc/refman/5.7/en/string-functions.html#function_position
## How was this patch tested?
unit tests
Author: Yuming Wang <wgyumg@gmail.com>
Closes #18206 from wangyum/SPARK-20754-mod&position.
---
.../org/apache/spark/sql/catalyst/parser/SqlBase.g4 | 3 +++
.../spark/sql/catalyst/analysis/FunctionRegistry.scala | 2 ++
.../spark/sql/catalyst/expressions/arithmetic.scala | 2 ++
.../sql/catalyst/expressions/stringExpressions.scala | 4 ++++
.../apache/spark/sql/catalyst/parser/AstBuilder.scala | 7 +++++++
.../catalyst/parser/TableIdentifierParserSuite.scala | 2 +-
.../src/test/resources/sql-tests/inputs/operators.sql | 5 ++++-
.../resources/sql-tests/inputs/string-functions.sql | 3 +++
.../test/resources/sql-tests/results/operators.sql.out | 10 +++++++++-
.../sql-tests/results/string-functions.sql.out | 10 +++++++++-
10 files changed, 44 insertions(+), 4 deletions(-)
diff --git a/sql/catalyst/src/main/antlr4/org/apache/spark/sql/catalyst/parser/SqlBase.g4 b/sql/catalyst/src/main/antlr4/org/apache/spark/sql/catalyst/parser/SqlBase.g4
index 43f7ff5cb4..ef5648c6db 100644
--- a/sql/catalyst/src/main/antlr4/org/apache/spark/sql/catalyst/parser/SqlBase.g4
+++ b/sql/catalyst/src/main/antlr4/org/apache/spark/sql/catalyst/parser/SqlBase.g4
@@ -563,6 +563,7 @@ primaryExpression
| CAST '(' expression AS dataType ')' #cast
| FIRST '(' expression (IGNORE NULLS)? ')' #first
| LAST '(' expression (IGNORE NULLS)? ')' #last
+ | POSITION '(' substr=valueExpression IN str=valueExpression ')' #position
| constant #constantDefault
| ASTERISK #star
| qualifiedName '.' ASTERISK #star
@@ -720,6 +721,7 @@ nonReserved
| SET | RESET
| VIEW | REPLACE
| IF
+ | POSITION
| NO | DATA
| START | TRANSACTION | COMMIT | ROLLBACK | IGNORE
| SORT | CLUSTER | DISTRIBUTE | UNSET | TBLPROPERTIES | SKEWED | STORED | DIRECTORIES | LOCATION
@@ -850,6 +852,7 @@ MACRO: 'MACRO';
IGNORE: 'IGNORE';
IF: 'IF';
+POSITION: 'POSITION';
EQ : '=' | '==';
NSEQ: '<=>';
diff --git a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/FunctionRegistry.scala b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/FunctionRegistry.scala
index 4245b70892..877328164a 100644
--- a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/FunctionRegistry.scala
+++ b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/FunctionRegistry.scala
@@ -240,6 +240,7 @@ object FunctionRegistry {
expression[Log1p]("log1p"),
expression[Log2]("log2"),
expression[Log]("ln"),
+ expression[Remainder]("mod"),
expression[UnaryMinus]("negative"),
expression[Pi]("pi"),
expression[Pmod]("pmod"),
@@ -325,6 +326,7 @@ object FunctionRegistry {
expression[StringTrimLeft]("ltrim"),
expression[JsonTuple]("json_tuple"),
expression[ParseUrl]("parse_url"),
+ expression[StringLocate]("position"),
expression[FormatString]("printf"),
expression[RegExpExtract]("regexp_extract"),
expression[RegExpReplace]("regexp_replace"),
diff --git a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/arithmetic.scala b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/arithmetic.scala
index f2b252259b..ec6e6ba0f0 100644
--- a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/arithmetic.scala
+++ b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/arithmetic.scala
@@ -320,6 +320,8 @@ case class Divide(left: Expression, right: Expression) extends BinaryArithmetic
Examples:
> SELECT 2 _FUNC_ 1.8;
0.2
+ > SELECT MOD(2, 1.8);
+ 0.2
""")
case class Remainder(left: Expression, right: Expression) extends BinaryArithmetic {
diff --git a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/stringExpressions.scala b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/stringExpressions.scala
index 035a1afe8b..717ada225a 100644
--- a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/stringExpressions.scala
+++ b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/stringExpressions.scala
@@ -654,8 +654,12 @@ case class SubstringIndex(strExpr: Expression, delimExpr: Expression, countExpr:
""",
extended = """
Examples:
+ > SELECT _FUNC_('bar', 'foobarbar');
+ 4
> SELECT _FUNC_('bar', 'foobarbar', 5);
7
+ > SELECT POSITION('bar' IN 'foobarbar');
+ 4
""")
// scalastyle:on line.size.limit
case class StringLocate(substr: Expression, str: Expression, start: Expression)
diff --git a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/parser/AstBuilder.scala b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/parser/AstBuilder.scala
index a16611af28..500d999c30 100644
--- a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/parser/AstBuilder.scala
+++ b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/parser/AstBuilder.scala
@@ -1076,6 +1076,13 @@ class AstBuilder(conf: SQLConf) extends SqlBaseBaseVisitor[AnyRef] with Logging
Last(expression(ctx.expression), Literal(ignoreNullsExpr)).toAggregateExpression()
}
+ /**
+ * Create a Position expression.
+ */
+ override def visitPosition(ctx: PositionContext): Expression = withOrigin(ctx) {
+ new StringLocate(expression(ctx.substr), expression(ctx.str))
+ }
+
/**
* Create a (windowed) Function expression.
*/
diff --git a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/parser/TableIdentifierParserSuite.scala b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/parser/TableIdentifierParserSuite.scala
index f33abc5b2e..76be6ee3f5 100644
--- a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/parser/TableIdentifierParserSuite.scala
+++ b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/parser/TableIdentifierParserSuite.scala
@@ -51,7 +51,7 @@ class TableIdentifierParserSuite extends SparkFunSuite {
"rollup", "row", "rows", "set", "smallint", "table", "timestamp", "to", "trigger",
"true", "truncate", "update", "user", "values", "with", "regexp", "rlike",
"bigint", "binary", "boolean", "current_date", "current_timestamp", "date", "double", "float",
- "int", "smallint", "timestamp", "at")
+ "int", "smallint", "timestamp", "at", "position")
val hiveStrictNonReservedKeyword = Seq("anti", "full", "inner", "left", "semi", "right",
"natural", "union", "intersect", "except", "database", "on", "join", "cross", "select", "from",
diff --git a/sql/core/src/test/resources/sql-tests/inputs/operators.sql b/sql/core/src/test/resources/sql-tests/inputs/operators.sql
index 75a0256ad7..3934620577 100644
--- a/sql/core/src/test/resources/sql-tests/inputs/operators.sql
+++ b/sql/core/src/test/resources/sql-tests/inputs/operators.sql
@@ -76,4 +76,7 @@ select floor(0.01);
select floor(-0.10);
-- comparison operator
-select 1 > 0.00001
+select 1 > 0.00001;
+
+-- mod
+select mod(7, 2), mod(7, 0), mod(0, 2), mod(7, null), mod(null, 2), mod(null, null);
diff --git a/sql/core/src/test/resources/sql-tests/inputs/string-functions.sql b/sql/core/src/test/resources/sql-tests/inputs/string-functions.sql
index d82df11251..20c0390664 100644
--- a/sql/core/src/test/resources/sql-tests/inputs/string-functions.sql
+++ b/sql/core/src/test/resources/sql-tests/inputs/string-functions.sql
@@ -15,3 +15,6 @@ select replace('abc', 'b');
-- uuid
select length(uuid()), (uuid() <> uuid());
+
+-- position
+select position('bar' in 'foobarbar'), position(null, 'foobarbar'), position('aaads', null);
diff --git a/sql/core/src/test/resources/sql-tests/results/operators.sql.out b/sql/core/src/test/resources/sql-tests/results/operators.sql.out
index 57e8a612fa..51ccf764d9 100644
--- a/sql/core/src/test/resources/sql-tests/results/operators.sql.out
+++ b/sql/core/src/test/resources/sql-tests/results/operators.sql.out
@@ -1,5 +1,5 @@
-- Automatically generated by SQLQueryTestSuite
--- Number of queries: 50
+-- Number of queries: 51
-- !query 0
@@ -412,3 +412,11 @@ select 1 > 0.00001
struct<(CAST(1 AS BIGINT) > 0):boolean>
-- !query 49 output
true
+
+
+-- !query 50
+select mod(7, 2), mod(7, 0), mod(0, 2), mod(7, null), mod(null, 2), mod(null, null)
+-- !query 50 schema
+struct<(7 % 2):int,(7 % 0):int,(0 % 2):int,(7 % CAST(NULL AS INT)):int,(CAST(NULL AS INT) % 2):int,(CAST(NULL AS DOUBLE) % CAST(NULL AS DOUBLE)):double>
+-- !query 50 output
+1 NULL 0 NULL NULL NULL
diff --git a/sql/core/src/test/resources/sql-tests/results/string-functions.sql.out b/sql/core/src/test/resources/sql-tests/results/string-functions.sql.out
index 4093a7b9fc..52eb554edf 100644
--- a/sql/core/src/test/resources/sql-tests/results/string-functions.sql.out
+++ b/sql/core/src/test/resources/sql-tests/results/string-functions.sql.out
@@ -1,5 +1,5 @@
-- Automatically generated by SQLQueryTestSuite
--- Number of queries: 7
+-- Number of queries: 8
-- !query 0
@@ -78,3 +78,11 @@ select length(uuid()), (uuid() <> uuid())
struct<length(uuid()):int,(NOT (uuid() = uuid())):boolean>
-- !query 6 output
36 true
+
+
+-- !query 7
+select position('bar' in 'foobarbar'), position(null, 'foobarbar'), position('aaads', null)
+-- !query 7 schema
+struct<locate(bar, foobarbar, 1):int,locate(CAST(NULL AS STRING), foobarbar, 1):int,locate(aaads, CAST(NULL AS STRING), 1):int>
+-- !query 7 output
+4 NULL NULL
--
GitLab