Skip to content

[SPARK-43779][SQL] ParseToDate should load the EvalMode in main thread #41298

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
amaliujia wants to merge 4 commits into from
Closed

[SPARK-43779][SQL] ParseToDate should load the EvalMode in main thread #41298

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
dc60e49
[SPARK-43779][SQL] ParseToDate should load the EvalMode in main thread.
amaliujia May 24, 2023
494b547
update
amaliujia May 25, 2023
6441539
updagee
amaliujia May 25, 2023
3b258bf
update
amaliujia May 26, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
9 changes: 6 additions & 3 deletions ...talyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/datetimeExpressions.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2044,12 +2044,15 @@ case class MonthsBetween(
case class ParseToDate(
left: Expression,
format: Option[Expression],
timeZoneId: Option[String] = None)
timeZoneId: Option[String] = None,
ansiEnabled: Boolean = SQLConf.get.ansiEnabled)
extends RuntimeReplaceable with ImplicitCastInputTypes with TimeZoneAwareExpression {

override lazy val replacement: Expression = format.map { f =>
Cast(GetTimestamp(left, f, TimestampType, timeZoneId), DateType, timeZoneId)
}.getOrElse(Cast(left, DateType, timeZoneId)) // backwards compatibility
Cast(GetTimestamp(left, f, TimestampType, timeZoneId, ansiEnabled), DateType, timeZoneId,
EvalMode.fromBoolean(ansiEnabled))
}.getOrElse(Cast(left, DateType, timeZoneId,
EvalMode.fromBoolean(ansiEnabled))) // backwards compatibility

def this(left: Expression, format: Expression) = {
this(left, Option(format))
Expand Down
8 changes: 4 additions & 4 deletions sql/core/src/test/resources/sql-tests/analyzer-results/ansi/date.sql.out
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -148,21 +148,21 @@ select UNIX_DATE(DATE('1970-01-01')), UNIX_DATE(DATE('2020-12-04')), UNIX_DATE(n
-- !query
select to_date(null), to_date('2016-12-31'), to_date('2016-12-31', 'yyyy-MM-dd')
-- !query analysis
Project [to_date(cast(null as string), None, Some(America/Los_Angeles)) AS to_date(NULL)#x, to_date(2016-12-31, None, Some(America/Los_Angeles)) AS to_date(2016-12-31)#x, to_date(2016-12-31, Some(yyyy-MM-dd), Some(America/Los_Angeles)) AS to_date(2016-12-31, yyyy-MM-dd)#x]
Project [to_date(cast(null as string), None, Some(America/Los_Angeles), true) AS to_date(NULL)#x, to_date(2016-12-31, None, Some(America/Los_Angeles), true) AS to_date(2016-12-31)#x, to_date(2016-12-31, Some(yyyy-MM-dd), Some(America/Los_Angeles), true) AS to_date(2016-12-31, yyyy-MM-dd)#x]
+- OneRowRelation


-- !query
select to_date("16", "dd")
-- !query analysis
Project [to_date(16, Some(dd), Some(America/Los_Angeles)) AS to_date(16, dd)#x]
Project [to_date(16, Some(dd), Some(America/Los_Angeles), true) AS to_date(16, dd)#x]
+- OneRowRelation


-- !query
select to_date("02-29", "MM-dd")
-- !query analysis
Project [to_date(02-29, Some(MM-dd), Some(America/Los_Angeles)) AS to_date(02-29, MM-dd)#x]
Project [to_date(02-29, Some(MM-dd), Some(America/Los_Angeles), true) AS to_date(02-29, MM-dd)#x]
+- OneRowRelation


Expand Down Expand Up @@ -729,7 +729,7 @@ select date '2012-01-01' - interval '2-2' year to month,
-- !query
select to_date('26/October/2015', 'dd/MMMMM/yyyy')
-- !query analysis
Project [to_date(26/October/2015, Some(dd/MMMMM/yyyy), Some(America/Los_Angeles)) AS to_date(26/October/2015, dd/MMMMM/yyyy)#x]
Project [to_date(26/October/2015, Some(dd/MMMMM/yyyy), Some(America/Los_Angeles), true) AS to_date(26/October/2015, dd/MMMMM/yyyy)#x]
+- OneRowRelation


Expand Down
4 changes: 2 additions & 2 deletions sql/core/src/test/resources/sql-tests/analyzer-results/ansi/datetime-parsing-invalid.sql.out
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -128,14 +128,14 @@ Project [from_csv(StructField(date,DateType,true), (dateFormat,yyyy-DDD), 2018-3
-- !query
select to_date("2020-01-27T20:06:11.847", "yyyy-MM-dd HH:mm:ss.SSS")
-- !query analysis
Project [to_date(2020-01-27T20:06:11.847, Some(yyyy-MM-dd HH:mm:ss.SSS), Some(America/Los_Angeles)) AS to_date(2020-01-27T20:06:11.847, yyyy-MM-dd HH:mm:ss.SSS)#x]
Project [to_date(2020-01-27T20:06:11.847, Some(yyyy-MM-dd HH:mm:ss.SSS), Some(America/Los_Angeles), true) AS to_date(2020-01-27T20:06:11.847, yyyy-MM-dd HH:mm:ss.SSS)#x]
+- OneRowRelation


-- !query
select to_date("Unparseable", "yyyy-MM-dd HH:mm:ss.SSS")
-- !query analysis
Project [to_date(Unparseable, Some(yyyy-MM-dd HH:mm:ss.SSS), Some(America/Los_Angeles)) AS to_date(Unparseable, yyyy-MM-dd HH:mm:ss.SSS)#x]
Project [to_date(Unparseable, Some(yyyy-MM-dd HH:mm:ss.SSS), Some(America/Los_Angeles), true) AS to_date(Unparseable, yyyy-MM-dd HH:mm:ss.SSS)#x]
+- OneRowRelation


Expand Down
8 changes: 4 additions & 4 deletions sql/core/src/test/resources/sql-tests/analyzer-results/date.sql.out
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -148,21 +148,21 @@ select UNIX_DATE(DATE('1970-01-01')), UNIX_DATE(DATE('2020-12-04')), UNIX_DATE(n
-- !query
select to_date(null), to_date('2016-12-31'), to_date('2016-12-31', 'yyyy-MM-dd')
-- !query analysis
Project [to_date(cast(null as string), None, Some(America/Los_Angeles)) AS to_date(NULL)#x, to_date(2016-12-31, None, Some(America/Los_Angeles)) AS to_date(2016-12-31)#x, to_date(2016-12-31, Some(yyyy-MM-dd), Some(America/Los_Angeles)) AS to_date(2016-12-31, yyyy-MM-dd)#x]
Project [to_date(cast(null as string), None, Some(America/Los_Angeles), false) AS to_date(NULL)#x, to_date(2016-12-31, None, Some(America/Los_Angeles), false) AS to_date(2016-12-31)#x, to_date(2016-12-31, Some(yyyy-MM-dd), Some(America/Los_Angeles), false) AS to_date(2016-12-31, yyyy-MM-dd)#x]
+- OneRowRelation


-- !query
select to_date("16", "dd")
-- !query analysis
Project [to_date(16, Some(dd), Some(America/Los_Angeles)) AS to_date(16, dd)#x]
Project [to_date(16, Some(dd), Some(America/Los_Angeles), false) AS to_date(16, dd)#x]
+- OneRowRelation


-- !query
select to_date("02-29", "MM-dd")
-- !query analysis
Project [to_date(02-29, Some(MM-dd), Some(America/Los_Angeles)) AS to_date(02-29, MM-dd)#x]
Project [to_date(02-29, Some(MM-dd), Some(America/Los_Angeles), false) AS to_date(02-29, MM-dd)#x]
+- OneRowRelation


Expand Down Expand Up @@ -804,7 +804,7 @@ select date '2012-01-01' - interval '2-2' year to month,
-- !query
select to_date('26/October/2015', 'dd/MMMMM/yyyy')
-- !query analysis
Project [to_date(26/October/2015, Some(dd/MMMMM/yyyy), Some(America/Los_Angeles)) AS to_date(26/October/2015, dd/MMMMM/yyyy)#x]
Project [to_date(26/October/2015, Some(dd/MMMMM/yyyy), Some(America/Los_Angeles), false) AS to_date(26/October/2015, dd/MMMMM/yyyy)#x]
+- OneRowRelation


Expand Down
8 changes: 4 additions & 4 deletions sql/core/src/test/resources/sql-tests/analyzer-results/datetime-legacy.sql.out
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -148,21 +148,21 @@ select UNIX_DATE(DATE('1970-01-01')), UNIX_DATE(DATE('2020-12-04')), UNIX_DATE(n
-- !query
select to_date(null), to_date('2016-12-31'), to_date('2016-12-31', 'yyyy-MM-dd')
-- !query analysis
Project [to_date(cast(null as string), None, Some(America/Los_Angeles)) AS to_date(NULL)#x, to_date(2016-12-31, None, Some(America/Los_Angeles)) AS to_date(2016-12-31)#x, to_date(2016-12-31, Some(yyyy-MM-dd), Some(America/Los_Angeles)) AS to_date(2016-12-31, yyyy-MM-dd)#x]
Project [to_date(cast(null as string), None, Some(America/Los_Angeles), false) AS to_date(NULL)#x, to_date(2016-12-31, None, Some(America/Los_Angeles), false) AS to_date(2016-12-31)#x, to_date(2016-12-31, Some(yyyy-MM-dd), Some(America/Los_Angeles), false) AS to_date(2016-12-31, yyyy-MM-dd)#x]
+- OneRowRelation


-- !query
select to_date("16", "dd")
-- !query analysis
Project [to_date(16, Some(dd), Some(America/Los_Angeles)) AS to_date(16, dd)#x]
Project [to_date(16, Some(dd), Some(America/Los_Angeles), false) AS to_date(16, dd)#x]
+- OneRowRelation


-- !query
select to_date("02-29", "MM-dd")
-- !query analysis
Project [to_date(02-29, Some(MM-dd), Some(America/Los_Angeles)) AS to_date(02-29, MM-dd)#x]
Project [to_date(02-29, Some(MM-dd), Some(America/Los_Angeles), false) AS to_date(02-29, MM-dd)#x]
+- OneRowRelation


Expand Down Expand Up @@ -804,7 +804,7 @@ select date '2012-01-01' - interval '2-2' year to month,
-- !query
select to_date('26/October/2015', 'dd/MMMMM/yyyy')
-- !query analysis
Project [to_date(26/October/2015, Some(dd/MMMMM/yyyy), Some(America/Los_Angeles)) AS to_date(26/October/2015, dd/MMMMM/yyyy)#x]
Project [to_date(26/October/2015, Some(dd/MMMMM/yyyy), Some(America/Los_Angeles), false) AS to_date(26/October/2015, dd/MMMMM/yyyy)#x]
+- OneRowRelation


Expand Down
4 changes: 2 additions & 2 deletions sql/core/src/test/resources/sql-tests/analyzer-results/datetime-parsing-invalid.sql.out
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -128,14 +128,14 @@ Project [from_csv(StructField(date,DateType,true), (dateFormat,yyyy-DDD), 2018-3
-- !query
select to_date("2020-01-27T20:06:11.847", "yyyy-MM-dd HH:mm:ss.SSS")
-- !query analysis
Project [to_date(2020-01-27T20:06:11.847, Some(yyyy-MM-dd HH:mm:ss.SSS), Some(America/Los_Angeles)) AS to_date(2020-01-27T20:06:11.847, yyyy-MM-dd HH:mm:ss.SSS)#x]
Project [to_date(2020-01-27T20:06:11.847, Some(yyyy-MM-dd HH:mm:ss.SSS), Some(America/Los_Angeles), false) AS to_date(2020-01-27T20:06:11.847, yyyy-MM-dd HH:mm:ss.SSS)#x]
+- OneRowRelation


-- !query
select to_date("Unparseable", "yyyy-MM-dd HH:mm:ss.SSS")
-- !query analysis
Project [to_date(Unparseable, Some(yyyy-MM-dd HH:mm:ss.SSS), Some(America/Los_Angeles)) AS to_date(Unparseable, yyyy-MM-dd HH:mm:ss.SSS)#x]
Project [to_date(Unparseable, Some(yyyy-MM-dd HH:mm:ss.SSS), Some(America/Los_Angeles), false) AS to_date(Unparseable, yyyy-MM-dd HH:mm:ss.SSS)#x]
+- OneRowRelation


Expand Down
8 changes: 4 additions & 4 deletions sql/core/src/test/resources/sql-tests/analyzer-results/group-by-filter.sql.out
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -102,7 +102,7 @@ SELECT COUNT(id) FILTER (WHERE hiredate = date "2001-01-01") FROM emp
-- !query
SELECT COUNT(id) FILTER (WHERE hiredate = to_date('2001-01-01 00:00:00')) FROM emp
-- !query analysis
Aggregate [count(id#x) FILTER (WHERE (hiredate#x = to_date(2001-01-01 00:00:00, None, Some(America/Los_Angeles)))) AS count(id) FILTER (WHERE (hiredate = to_date(2001-01-01 00:00:00)))#xL]
Aggregate [count(id#x) FILTER (WHERE (hiredate#x = to_date(2001-01-01 00:00:00, None, Some(America/Los_Angeles), false))) AS count(id) FILTER (WHERE (hiredate = to_date(2001-01-01 00:00:00)))#xL]
+- SubqueryAlias emp
+- View (`EMP`, [id#x,emp_name#x,hiredate#x,salary#x,dept_id#x])
+- Project [cast(id#x as int) AS id#x, cast(emp_name#x as string) AS emp_name#x, cast(hiredate#x as date) AS hiredate#x, cast(salary#x as double) AS salary#x, cast(dept_id#x as int) AS dept_id#x]
Expand Down Expand Up @@ -162,7 +162,7 @@ Aggregate [count(distinct id#x) AS count(DISTINCT id)#xL, count(distinct id#x) F
-- !query
SELECT COUNT(DISTINCT id) FILTER (WHERE hiredate = to_timestamp("2001-01-01 00:00:00")), COUNT(DISTINCT id) FILTER (WHERE hiredate = to_date('2001-01-01 00:00:00')) FROM emp
-- !query analysis
Aggregate [count(distinct id#x) FILTER (WHERE (cast(hiredate#x as timestamp) = to_timestamp(2001-01-01 00:00:00, None, TimestampType, Some(America/Los_Angeles), false))) AS count(DISTINCT id) FILTER (WHERE (hiredate = to_timestamp(2001-01-01 00:00:00)))#xL, count(distinct id#x) FILTER (WHERE (hiredate#x = to_date(2001-01-01 00:00:00, None, Some(America/Los_Angeles)))) AS count(DISTINCT id) FILTER (WHERE (hiredate = to_date(2001-01-01 00:00:00)))#xL]
Aggregate [count(distinct id#x) FILTER (WHERE (cast(hiredate#x as timestamp) = to_timestamp(2001-01-01 00:00:00, None, TimestampType, Some(America/Los_Angeles), false))) AS count(DISTINCT id) FILTER (WHERE (hiredate = to_timestamp(2001-01-01 00:00:00)))#xL, count(distinct id#x) FILTER (WHERE (hiredate#x = to_date(2001-01-01 00:00:00, None, Some(America/Los_Angeles), false))) AS count(DISTINCT id) FILTER (WHERE (hiredate = to_date(2001-01-01 00:00:00)))#xL]
+- SubqueryAlias emp
+- View (`EMP`, [id#x,emp_name#x,hiredate#x,salary#x,dept_id#x])
+- Project [cast(id#x as int) AS id#x, cast(emp_name#x as string) AS emp_name#x, cast(hiredate#x as date) AS hiredate#x, cast(salary#x as double) AS salary#x, cast(dept_id#x as int) AS dept_id#x]
Expand Down Expand Up @@ -356,7 +356,7 @@ SELECT dept_id, SUM(salary) FILTER (WHERE hiredate > date "2003-01-01") FROM emp
-- !query
SELECT dept_id, SUM(salary) FILTER (WHERE hiredate > to_date("2003-01-01")) FROM emp GROUP BY dept_id
-- !query analysis
Aggregate [dept_id#x], [dept_id#x, sum(salary#x) FILTER (WHERE (hiredate#x > to_date(2003-01-01, None, Some(America/Los_Angeles)))) AS sum(salary) FILTER (WHERE (hiredate > to_date(2003-01-01)))#x]
Aggregate [dept_id#x], [dept_id#x, sum(salary#x) FILTER (WHERE (hiredate#x > to_date(2003-01-01, None, Some(America/Los_Angeles), false))) AS sum(salary) FILTER (WHERE (hiredate > to_date(2003-01-01)))#x]
+- SubqueryAlias emp
+- View (`EMP`, [id#x,emp_name#x,hiredate#x,salary#x,dept_id#x])
+- Project [cast(id#x as int) AS id#x, cast(emp_name#x as string) AS emp_name#x, cast(hiredate#x as date) AS hiredate#x, cast(salary#x as double) AS salary#x, cast(dept_id#x as int) AS dept_id#x]
Expand Down Expand Up @@ -464,7 +464,7 @@ SELECT 'foo', SUM(salary) FILTER (WHERE hiredate >= date "2003-01-01") FROM emp
-- !query
SELECT 'foo', SUM(salary) FILTER (WHERE hiredate >= to_date("2003-01-01")) FROM emp GROUP BY 1
-- !query analysis
Aggregate [foo], [foo AS foo#x, sum(salary#x) FILTER (WHERE (hiredate#x >= to_date(2003-01-01, None, Some(America/Los_Angeles)))) AS sum(salary) FILTER (WHERE (hiredate >= to_date(2003-01-01)))#x]
Aggregate [foo], [foo AS foo#x, sum(salary#x) FILTER (WHERE (hiredate#x >= to_date(2003-01-01, None, Some(America/Los_Angeles), false))) AS sum(salary) FILTER (WHERE (hiredate >= to_date(2003-01-01)))#x]
+- SubqueryAlias emp
+- View (`EMP`, [id#x,emp_name#x,hiredate#x,salary#x,dept_id#x])
+- Project [cast(id#x as int) AS id#x, cast(emp_name#x as string) AS emp_name#x, cast(hiredate#x as date) AS hiredate#x, cast(salary#x as double) AS salary#x, cast(dept_id#x as int) AS dept_id#x]
Expand Down
4 changes: 2 additions & 2 deletions sql/core/src/test/resources/sql-tests/analyzer-results/postgreSQL/text.sql.out
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -84,7 +84,7 @@ Project [concat(one) AS concat(one)#x]
-- !query
select concat(1,2,3,'hello',true, false, to_date('20100309','yyyyMMdd'))
-- !query analysis
Project [concat(cast(1 as string), cast(2 as string), cast(3 as string), hello, cast(true as string), cast(false as string), cast(to_date(20100309, Some(yyyyMMdd), Some(America/Los_Angeles)) as string)) AS concat(1, 2, 3, hello, true, false, to_date(20100309, yyyyMMdd))#x]
Project [concat(cast(1 as string), cast(2 as string), cast(3 as string), hello, cast(true as string), cast(false as string), cast(to_date(20100309, Some(yyyyMMdd), Some(America/Los_Angeles), true) as string)) AS concat(1, 2, 3, hello, true, false, to_date(20100309, yyyyMMdd))#x]
+- OneRowRelation


Expand All @@ -98,7 +98,7 @@ Project [concat_ws(#, one) AS concat_ws(#, one)#x]
-- !query
select concat_ws('#',1,2,3,'hello',true, false, to_date('20100309','yyyyMMdd'))
-- !query analysis
Project [concat_ws(#, cast(1 as string), cast(2 as string), cast(3 as string), hello, cast(true as string), cast(false as string), cast(to_date(20100309, Some(yyyyMMdd), Some(America/Los_Angeles)) as string)) AS concat_ws(#, 1, 2, 3, hello, true, false, to_date(20100309, yyyyMMdd))#x]
Project [concat_ws(#, cast(1 as string), cast(2 as string), cast(3 as string), hello, cast(true as string), cast(false as string), cast(to_date(20100309, Some(yyyyMMdd), Some(America/Los_Angeles), true) as string)) AS concat_ws(#, 1, 2, 3, hello, true, false, to_date(20100309, yyyyMMdd))#x]
+- OneRowRelation


Expand Down
Loading