[SPARK-14922][SPARK-17732][SPARK-23866][SQL] Support partition filters in ALTER TABLE DROP PARTITION #20999

mgaido91 · 2018-04-06T16:26:34Z

What changes were proposed in this pull request?

Hive has been supporting for a while the ability of dropping partitions using any kind of comparison operator on them. Spark so far is supporting only dropping partitions by exact values. For instance, Spark doesn't support:

ALTER TABLE mytable DROP PARTITION(mydate < '2018-04-06')

The PR adds the support to this syntax too.

The PR takes input from the effort in #19691 by @DazhuangSu. As such, this closes #19691.

How was this patch tested?

UTs to be added

… PARTITION

gatorsmile · 2018-04-06T16:33:49Z

Just FYI, there are three related PRs:

[SPARK-17732][SQL] ALTER TABLE DROP PARTITION should support comparators #15987
[SPARK-17732][SQL] Revert ALTER TABLE DROP PARTITION should support comparators #16036
[SPARK-14922][SPARK-17732][SQL]ALTER TABLE DROP PARTITION should support comparators #19691

SparkQA · 2018-04-06T19:02:08Z

Test build #88996 has finished for PR 20999 at commit b57a5d1.

This patch fails Spark unit tests.
This patch merges cleanly.
This patch adds no public classes.

mgaido91 · 2018-04-06T20:52:48Z

thanks @gatorsmile , I missed them. I see that #19691 is still open and waiting for review. Probably I should close this one and we can go on on that PR. But I have seen no activity on it for a while, is there any reason?

Thanks.

gatorsmile · 2018-04-08T03:55:17Z

sql/catalyst/src/main/antlr4/org/apache/spark/sql/catalyst/parser/SqlBase.g4

+    ;
+
+dropPartitionVal
+    : identifier (comparisonOperator constant)?


It has to be in this format? partCol1 > 2 How about 2 > partCol1?

yes, in Hive it has to be like this. 2 > partCol1 is not supported by Hive.

Hive also throws antler errors for the case 2 > partCol1?

Hive does throw an error in that case, you mean asking that error is a parsing or another kind of exception?

yea, yes. I like user-understandable error messages.

Hive throws this parser exception:

hive> alter table test1 drop partition(1 > c); NoViableAltException(368@[]) at org.apache.hadoop.hive.ql.parse.HiveParser_IdentifiersParser.identifier(HiveParser_IdentifiersParser.java:12014) at org.apache.hadoop.hive.ql.parse.HiveParser_IdentifiersParser.dropPartitionVal(HiveParser_IdentifiersParser.java:11684) at org.apache.hadoop.hive.ql.parse.HiveParser_IdentifiersParser.dropPartitionSpec(HiveParser_IdentifiersParser.java:11563) at org.apache.hadoop.hive.ql.parse.HiveParser.dropPartitionSpec(HiveParser.java:44851) at org.apache.hadoop.hive.ql.parse.HiveParser.alterStatementSuffixDropPartitions(HiveParser.java:11564) at org.apache.hadoop.hive.ql.parse.HiveParser.alterTableStatementSuffix(HiveParser.java:8000) at org.apache.hadoop.hive.ql.parse.HiveParser.alterStatement(HiveParser.java:7450) at org.apache.hadoop.hive.ql.parse.HiveParser.ddlStatement(HiveParser.java:4340) at org.apache.hadoop.hive.ql.parse.HiveParser.execStatement(HiveParser.java:2497) at org.apache.hadoop.hive.ql.parse.HiveParser.statement(HiveParser.java:1423) at org.apache.hadoop.hive.ql.parse.ParseDriver.parse(ParseDriver.java:209) at org.apache.hadoop.hive.ql.parse.ParseUtils.parse(ParseUtils.java:74) at org.apache.hadoop.hive.ql.parse.ParseUtils.parse(ParseUtils.java:67) at org.apache.hadoop.hive.ql.Driver.compile(Driver.java:615) at org.apache.hadoop.hive.ql.Driver.compileInternal(Driver.java:1829) at org.apache.hadoop.hive.ql.Driver.compileAndRespond(Driver.java:1776) at org.apache.hadoop.hive.ql.Driver.compileAndRespond(Driver.java:1771) at org.apache.hadoop.hive.ql.reexec.ReExecDriver.compileAndRespond(ReExecDriver.java:126) at org.apache.hadoop.hive.ql.reexec.ReExecDriver.run(ReExecDriver.java:214) at org.apache.hadoop.hive.cli.CliDriver.processLocalCmd(CliDriver.java:239) at org.apache.hadoop.hive.cli.CliDriver.processCmd(CliDriver.java:188) at org.apache.hadoop.hive.cli.CliDriver.processLine(CliDriver.java:402) at org.apache.hadoop.hive.cli.CliDriver.executeDriver(CliDriver.java:832) at org.apache.hadoop.hive.cli.CliDriver.run(CliDriver.java:770) at org.apache.hadoop.hive.cli.CliDriver.main(CliDriver.java:694) at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62) at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) at java.lang.reflect.Method.invoke(Method.java:498) at org.apache.hadoop.util.RunJar.run(RunJar.java:221) at org.apache.hadoop.util.RunJar.main(RunJar.java:136) FAILED: ParseException line 1:33 cannot recognize input near '1' '>' 'c' in drop partition statement

so yes, it is analogous to this.

thanks for the check. I still like meaningful messages though, we shold wait for other reviewer's comments.

gatorsmile · 2018-04-08T03:56:58Z

Let us start reviewing that PR.

mgaido91 · 2018-04-08T08:47:31Z

ok, thanks @gatorsmile. Then I am closing this.

mgaido91 · 2018-09-05T16:13:58Z

I am reopening this according to the discussion in #19691.

cc @maropu @DazhuangSu

SparkQA · 2018-09-05T20:21:39Z

Test build #95725 has finished for PR 20999 at commit 148f477.

This patch passes all tests.
This patch merges cleanly.
This patch adds no public classes.

maropu · 2018-09-06T00:09:24Z

still WIP? btw, it seems currently credits can go to multiple developers;
#22324 (comment)

mgaido91 · 2018-09-06T11:27:58Z

it seems currently credits can go to multiple developers;

Yes, but I don't know how to do that. Probably committers can do it in the merging process, so I think the only thing I can do is write in the description who should be credited and then committers can do accordingly. Please correct me if I am wrong. Thanks.

SparkQA · 2018-09-06T17:04:58Z

Test build #95758 has finished for PR 20999 at commit 7d3cf0c.

This patch passes all tests.
This patch merges cleanly.
This patch adds no public classes.

maropu · 2018-09-07T03:14:24Z

you're not wrong, but I just meant you don't need to describe "Credit for that should be given to him." in the PR description.

mgaido91 · 2018-09-07T07:55:41Z

Thanks @maropu , updated.

maropu · 2018-09-08T06:31:56Z

sql/catalyst/src/main/antlr4/org/apache/spark/sql/catalyst/parser/SqlBase.g4

+    ;
+
+dropPartitionVal
+    : identifier (comparisonOperator constant)?


Hive also throws antler errors for the case 2 > partCol1?

maropu · 2018-09-08T06:36:50Z

sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/HiveDDLSuite.scala

+  test("SPARK-14922: Partition filter is not allowed in ADD PARTITION") {
+    withTable("sales") {
+      sql("CREATE TABLE sales(id INT) PARTITIONED BY (country STRING, quarter STRING)")
+      intercept[AnalysisException] {


plz check all the error messages above?

ok, will do

maropu · 2018-09-08T07:41:37Z

sql/core/src/test/scala/org/apache/spark/sql/execution/command/DDLParserSuite.scala

    val tableIdent = TableIdentifier("table_name", None)
-    val expected1_table = AlterTableDropPartitionCommand(
+
+    val expected1_table = AlterTableDropPartitionCommand.fromSpecs(


Can you add tests case to check if the parser can accept the comparators added by this pr?

sure, will do, thanks.

maropu · 2018-09-08T07:54:04Z

sql/core/src/main/scala/org/apache/spark/sql/execution/command/ddl.scala

 import org.apache.spark.sql.catalyst.catalog._
 import org.apache.spark.sql.catalyst.catalog.CatalogTypes.TablePartitionSpec
-import org.apache.spark.sql.catalyst.expressions.{Attribute, AttributeReference}
+import org.apache.spark.sql.catalyst.expressions._


too many imports?

not sure what you mean here. The list of imports would be very long, as I use, EqualTo, And, Literal, Cast, BinaryComparison, etc. I can list all them, but I am not sure it is worth. What do you think?

I just wanted to check if your IDE wrongly folded this import, or not. It's ok.

maropu · 2018-09-08T07:58:58Z

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/parser/AstBuilder.scala

+      ctx.dropPartitionVal().asScala.map { pFilter =>
+        if (pFilter.identifier() == null || pFilter.constant() == null ||
+            pFilter.comparisonOperator() == null) {
+          throw new ParseException(s"Invalid partition spec: ${pFilter.getText}", ctx)


Can you add tests for this exception in DDLParserSuite.scala?

sure, will do ASAP, thanks.

maropu · 2018-09-08T08:04:56Z

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/parser/AstBuilder.scala

+      ctx: DropPartitionSpecContext): Seq[Expression] = {
+    withOrigin(ctx) {
+      ctx.dropPartitionVal().asScala.map { pFilter =>
+        if (pFilter.identifier() == null || pFilter.constant() == null ||


no chance pFilter.identifier() == null?

I am not sure. The other 2 conditions can definitely be true, but I am not sure about this. I think it is safer to check it.

I saw no null check in the other partition spec, then I thought so; https://github.com/apache/spark/blob/master/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/parser/AstBuilder.scala#L274

maropu · 2018-09-08T08:09:24Z

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/parser/AstBuilder.scala

+        // running the command. The type is not relevant, it is replaced during the real resolution
+        val partition =
+          AttributeReference(pFilter.identifier().getText, StringType)()
+        val value = Literal(visitStringConstant(pFilter.constant()))


val value = Literal(visitStringConstant(pFilter.constant()), StringType) for better readablilty?

~~thanks, will update~~

sorry, I'd need to do Literal.create(visitStringConstant(pFilter.constant()), StringType). Everywhere else here we are using Literal and not Literal.create. So I am not sure it makes sense to change it, but I have not a strong opinion on this. If you prefer, I can switch to Literal.create. Thanks.

ok, thanks for the check. It's ok to keep the current one.

maropu · 2018-09-08T08:14:44Z

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/parser/AstBuilder.scala

+        // We cannot use UnresolvedAttribute because resolution is performed after Analysis, when
+        // running the command. The type is not relevant, it is replaced during the real resolution
+        val partition =
+          AttributeReference(pFilter.identifier().getText, StringType)()


Looks weird and why can we use UnresolvedAttribute in #19691?
https://github.com/apache/spark/pull/19691/files#diff-9847f5cef7cf7fbc5830fbc6b779ee10R293
(Sorry, but probably I miss something?)

Well, the answer is in def expressions in QueryPlan. In #19691, we end up with a Seq[(TablePartitionSpec, Seq[Expression])], so the expressions there are not recognized/considered by the Analyzer. In this PR we have Seq[Expression] (which is way cleaner IMHO and address comment https://github.com/apache/spark/pull/19691/files#r193002268), so these expressions are considered by the Analyzer.

Let me have more time to check this behaviour.

For example, how about this approach? (you tried already?) 52506f1

It added unresolved a logical plan (an input relation and filters) for AlterTableDropPartitionCommand, then resolved the plan in AlterTableDropPartitionCommand.run and computed partition specs based on the resolved expressions?

Your approach has the same issue, ie. would fail because the LogicalPlan you added is not resolved after analysis.

I may be missing something here, so sorry if I am not understanding something, but I think the issue is that the analyzer is called anyway before the AlterTableDropPartitionCommand.run command and it fails because of the unresolved attributes. Moreover, in your code I don't see it being part neither of children nor of innerChildren.

I think the alternative here is to add a rule to the analyzer for this, but it seems an overkill to me.

yea, if you put Expression in the class fields of AlterTableDropPartitionCommand (https://github.com/apache/spark/pull/20999/files#diff-54979ed5797b4a6193cf663dc23baca5R524), it fails in Analyzer.executeAndCheck. But, if we put LogicalPlan in the class fields, IIUC it doesn't fail there.

oh, now I see, sorry. What about then having a Seq[Filter] instead? In order to avoid the splitDisjunctivePredicates?

Ya, looks good to me. But, I'm not sure which one is the right approach, so we'd be better to wait for other reviewer's comments here, too. cc: @gatorsmile @viirya

sure, thanks @maropu.

maropu · 2018-09-08T08:15:49Z

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/parser/AstBuilder.scala

+   * - Less then or Equal: '<='
+   * - Greater than: '>'
+   * - Greater then or Equal: '>='
+   */


Hive also supports all the comparators above?

yes, it does

mgaido91

Thanks for your comments @maropu. I'll update the PR accordingly asap.

mgaido91 · 2018-09-08T09:35:44Z

sql/catalyst/src/main/antlr4/org/apache/spark/sql/catalyst/parser/SqlBase.g4

+    ;
+
+dropPartitionVal
+    : identifier (comparisonOperator constant)?


Hive does throw an error in that case, you mean asking that error is a parsing or another kind of exception?

mgaido91 · 2018-09-08T09:36:58Z

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/parser/AstBuilder.scala

+      ctx: DropPartitionSpecContext): Seq[Expression] = {
+    withOrigin(ctx) {
+      ctx.dropPartitionVal().asScala.map { pFilter =>
+        if (pFilter.identifier() == null || pFilter.constant() == null ||


I am not sure. The other 2 conditions can definitely be true, but I am not sure about this. I think it is safer to check it.

mgaido91 · 2018-09-08T09:37:50Z

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/parser/AstBuilder.scala

+      ctx.dropPartitionVal().asScala.map { pFilter =>
+        if (pFilter.identifier() == null || pFilter.constant() == null ||
+            pFilter.comparisonOperator() == null) {
+          throw new ParseException(s"Invalid partition spec: ${pFilter.getText}", ctx)


sure, will do ASAP, thanks.

mgaido91 · 2018-09-08T09:43:01Z

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/parser/AstBuilder.scala

+        // We cannot use UnresolvedAttribute because resolution is performed after Analysis, when
+        // running the command. The type is not relevant, it is replaced during the real resolution
+        val partition =
+          AttributeReference(pFilter.identifier().getText, StringType)()


Well, the answer is in def expressions in QueryPlan. In #19691, we end up with a Seq[(TablePartitionSpec, Seq[Expression])], so the expressions there are not recognized/considered by the Analyzer. In this PR we have Seq[Expression] (which is way cleaner IMHO and address comment https://github.com/apache/spark/pull/19691/files#r193002268), so these expressions are considered by the Analyzer.

mgaido91 · 2018-09-08T09:43:21Z

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/parser/AstBuilder.scala

+        // running the command. The type is not relevant, it is replaced during the real resolution
+        val partition =
+          AttributeReference(pFilter.identifier().getText, StringType)()
+        val value = Literal(visitStringConstant(pFilter.constant()))


~~thanks, will update~~

sorry, I'd need to do Literal.create(visitStringConstant(pFilter.constant()), StringType). Everywhere else here we are using Literal and not Literal.create. So I am not sure it makes sense to change it, but I have not a strong opinion on this. If you prefer, I can switch to Literal.create. Thanks.

mgaido91 · 2018-09-08T09:50:37Z

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/parser/AstBuilder.scala

+   * - Less then or Equal: '<='
+   * - Greater than: '>'
+   * - Greater then or Equal: '>='
+   */


yes, it does

mgaido91 · 2018-09-08T09:51:59Z

sql/core/src/main/scala/org/apache/spark/sql/execution/command/ddl.scala

 import org.apache.spark.sql.catalyst.catalog._
 import org.apache.spark.sql.catalyst.catalog.CatalogTypes.TablePartitionSpec
-import org.apache.spark.sql.catalyst.expressions.{Attribute, AttributeReference}
+import org.apache.spark.sql.catalyst.expressions._


not sure what you mean here. The list of imports would be very long, as I use, EqualTo, And, Literal, Cast, BinaryComparison, etc. I can list all them, but I am not sure it is worth. What do you think?

mgaido91 · 2018-09-08T09:52:27Z

sql/core/src/test/scala/org/apache/spark/sql/execution/command/DDLParserSuite.scala

    val tableIdent = TableIdentifier("table_name", None)
-    val expected1_table = AlterTableDropPartitionCommand(
+
+    val expected1_table = AlterTableDropPartitionCommand.fromSpecs(


sure, will do, thanks.

mgaido91 · 2018-09-08T09:53:08Z

sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/HiveDDLSuite.scala

+  test("SPARK-14922: Partition filter is not allowed in ADD PARTITION") {
+    withTable("sales") {
+      sql("CREATE TABLE sales(id INT) PARTITIONED BY (country STRING, quarter STRING)")
+      intercept[AnalysisException] {


ok, will do

cloud-fan · 2018-10-15T08:16:17Z

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/namedExpressions.scala

+ */
+case class PartitioningAttribute(name: String)
+  extends Attribute with Unevaluable {
+  override val exprId: ExprId = NamedExpression.newExprId


even it's a fake attribute, we should not change the exprId when this expression gets copied. Can we move exprId to the constructor?

cloud-fan · 2018-10-15T08:16:56Z

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/namedExpressions.scala

+  // Not really needed and used. We just need a dataType to be used during analysis for resolving
+  // the expressions. The String type is used because all the literals in PARTITION operations are
+  // parsed as strings and eventually casted later.
+  override def dataType: DataType = StringType


If it's not needed, can we throw exception here? We may need to override toString though.

Probably I should improve the comment then. it's misleading: this is actually needed because otherwise we may hit exceptions since the dataType is checked when running checkInputDataTypes of the comparison operator containing it. I'll improve the comment.

cloud-fan · 2018-10-15T08:18:04Z

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/parser/AstBuilder.scala

+   * Create a partition specification map with filters.
+   */
+  override def visitDropPartitionSpec(
+      ctx: DropPartitionSpecContext): Seq[Expression] = {


nit: can we move withOrigin(ctx) here? i.e.

def xxx(): T = withOrigin(ctx) { ... }

cloud-fan · 2018-10-15T08:23:44Z

sql/core/src/main/scala/org/apache/spark/sql/execution/command/ddl.scala

+      }
    }

    catalog.dropPartitions(


does hive have an API to drop partitions with a predicate? I think the current approach is very inefficient with non-equal partition predicates.

unfortunately, no. I checked https://github.com/apache/hive/blob/master/ql/src/java/org/apache/hadoop/hive/ql/metadata/Hive.java but I could find none.

So the implementation here is similar to how hive implements it?

Yes, this is my understanding. You can check DDLTaks.dropPartitions.

cloud-fan · 2018-10-15T11:23:40Z

retest this please

SparkQA · 2018-10-15T14:43:04Z

Test build #97389 has finished for PR 20999 at commit 2085088.

This patch fails Spark unit tests.
This patch merges cleanly.
This patch adds the following public classes (experimental):
case class PartitioningAttribute(

SparkQA · 2018-10-15T14:49:37Z

Test build #97382 has finished for PR 20999 at commit 2085088.

This patch fails Spark unit tests.
This patch merges cleanly.
This patch adds the following public classes (experimental):
case class PartitioningAttribute(

SparkQA · 2018-10-15T19:08:44Z

Test build #97398 has finished for PR 20999 at commit 146aa32.

This patch passes all tests.
This patch merges cleanly.
This patch adds no public classes.

mgaido91 · 2018-10-17T09:17:57Z

any more comments @cloud-fan @maropu @viirya ?

cloud-fan · 2018-10-17T15:09:33Z

It seems like Hive can drop partitions directly with a partition expression: https://github.com/apache/hive/blob/master/standalone-metastore/metastore-server/src/test/java/org/apache/hadoop/hive/metastore/HiveMetaStoreClientPreCatalog.java#L1016

can you double check this part?

mgaido91 · 2018-10-18T13:13:55Z

Yes, you're right @cloud-fan. I missed that, thanks. I am going to update this with the required change, but it is going to be a huge change. Thanks.

mgaido91 · 2018-10-29T15:21:34Z

@cloud-fan I checked the feasibility of doing what you suggested, but unfortunately that method is very hard to be used (and I am not sure how safe it is). The main issues are:

it doesn't accept a string for the filters, but its encoding in a byte array. The point is that the encoding is performed by Hive utils and relying on them might be very hard (especially it is hard to track any change Hive makes there across the versions);
a minor issue is that also an integer is required there and I have not been able to understand what it represents and therefore how to use it.

I am not sure if then we can go ahead with this solution, even though it is suboptimal and eventually create a new ticket for leveraging the Hive metastore API you mentioned in a separate PR, as it is not going to be trivial to do that. What do you think?

jinxing64 · 2018-12-26T12:01:08Z

@mgaido91
This is really helpful feature; Thanks working on this.
Is there any progress now ?

mgaido91 · 2018-12-26T13:41:26Z

thanks fir your comment @jinxing64.

I am waiting for an answer for the question above. Currently it is a sub-optimal solution, but I think we can go ahead with it as the Hive API is quite hard to be used, so I'd consider the comment by @cloud-fan as an optimization to be done in a followup PR.

If committers can help providing their opinion on this and keeping in review on this, I am active on this PR and it is my goal as well to make this in for 3.0.

sleep1661 · 2019-07-09T09:52:40Z

Is there any progress? @mgaido91 @cloud-fan

mgaido91 · 2019-07-11T10:58:15Z

I am waiting for committers' opinion on the topic above...

SparkQA · 2019-07-17T18:15:18Z

Test build #107795 has finished for PR 20999 at commit 146aa32.

This patch fails Java style tests.
This patch does not merge cleanly.
This patch adds no public classes.

SparkQA · 2019-07-20T09:49:20Z

Test build #107939 has finished for PR 20999 at commit 25533a0.

This patch fails to build.
This patch merges cleanly.
This patch adds no public classes.

SparkQA · 2019-07-20T14:46:53Z

Test build #107940 has finished for PR 20999 at commit 9676061.

This patch passes all tests.
This patch merges cleanly.
This patch adds no public classes.

weixiuli · 2019-08-26T08:00:49Z

It seems like Hive can drop partitions directly with a partition expression: https://github.com/apache/hive/blob/master/standalone-metastore/metastore-server/src/test/java/org/apache/hadoop/hive/metastore/HiveMetaStoreClientPreCatalog.java#L1016

can you double check this part?

Is there any progress? @mgaido91 @cloud-fan

mgaido91 · 2019-08-26T08:05:45Z

@weixiuli see #20999 (comment)

weixiuli · 2019-10-28T09:42:34Z

@mgaido91 @cloud-fan Thanks working on this, this PR has NOT been updated for a long time, i have some ideas to solve the above question in #26280, PTAL and kindly review.

github-actions · 2020-02-06T00:08:11Z

We're closing this PR because it hasn't been updated in a while. This isn't a judgement on the merit of the PR in any way. It's just a way of keeping the PR queue manageable.
If you'd like to revive this PR, please reopen it and ask a committer to remove the Stale tag!

AngersZhuuuu · 2022-01-05T06:50:13Z

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/namedExpressions.scala

+
+  override lazy val canonicalized: Expression = this.copy(exprId = ExprId(0))
+
+  override def withExprId(newExprId: ExprId): Attribute = throw new UnsupportedOperationException


override def sql: String = name

Can make the error message more clear.

AngersZhuuuu · 2022-01-05T06:51:14Z

sql/core/src/main/scala/org/apache/spark/sql/execution/command/ddl.scala

+          sparkSession.sessionState.conf.resolver) :: Nil
+      }
    }



Should check resolvedSpecs here to throw error message if total resolved spec is empty.

AngersZhuuuu · 2022-01-05T06:52:16Z

@viirya Any more suggestion on this function? It's really useful for warehouse with so many partition.

[WIP][SPARK-23866][SQL] Support partition filters in ALTER TABLE DROP…

b57a5d1

… PARTITION

gatorsmile reviewed Apr 8, 2018

View reviewed changes

mgaido91 closed this Apr 8, 2018

adding UT from SPARK-14922 by DazhuangSu

148f477

mgaido91 reopened this Sep 5, 2018

mgaido91 changed the title ~~[WIP][SPARK-23866][SQL] Support partition filters in ALTER TABLE DROP PARTITION~~ [WIP][SPARK-14922][SPARK-23866][SQL] Support partition filters in ALTER TABLE DROP PARTITION Sep 5, 2018

mgaido91 changed the title ~~[WIP][SPARK-14922][SPARK-23866][SQL] Support partition filters in ALTER TABLE DROP PARTITION~~ [SPARK-14922][SPARK-23866][SQL] Support partition filters in ALTER TABLE DROP PARTITION Sep 6, 2018

move expression build logic to AstBuilder

7d3cf0c

mgaido91 changed the title ~~[SPARK-14922][SPARK-23866][SQL] Support partition filters in ALTER TABLE DROP PARTITION~~ [SPARK-14922][SPARK-17732][SPARK-23866][SQL] Support partition filters in ALTER TABLE DROP PARTITION Sep 6, 2018

maropu reviewed Sep 8, 2018

View reviewed changes

mgaido91 commented Sep 8, 2018

View reviewed changes

cloud-fan reviewed Oct 15, 2018

View reviewed changes

address comments

2085088

fix ut failure

146aa32

dongjoon-hyun added the SQL label Jun 14, 2019

Merge branch 'master' of github.com:apache/spark into SPARK-23866

25533a0

fix

9676061

weixiuli mentioned this pull request Oct 28, 2019

[SPARK-14922][SPARK-17732][SPARK-23866][SQL] Support partition filter in ALTER TABLE DROP PARTITION and batch dropping PARTITIONS #26280

Closed

github-actions bot added the Stale label Feb 6, 2020

github-actions bot closed this Feb 7, 2020

AngersZhuuuu reviewed Jan 5, 2022

View reviewed changes


		override lazy val canonicalized: Expression = this.copy(exprId = ExprId(0))

		override def withExprId(newExprId: ExprId): Attribute = throw new UnsupportedOperationException

[SPARK-14922][SPARK-17732][SPARK-23866][SQL] Support partition filters in ALTER TABLE DROP PARTITION #20999

[SPARK-14922][SPARK-17732][SPARK-23866][SQL] Support partition filters in ALTER TABLE DROP PARTITION #20999

Uh oh!

Conversation

mgaido91 commented Apr 6, 2018 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

What changes were proposed in this pull request?

How was this patch tested?

Uh oh!

gatorsmile commented Apr 6, 2018

Uh oh!

SparkQA commented Apr 6, 2018

Uh oh!

mgaido91 commented Apr 6, 2018

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

gatorsmile commented Apr 8, 2018

Uh oh!

mgaido91 commented Apr 8, 2018

Uh oh!

mgaido91 commented Sep 5, 2018

Uh oh!

SparkQA commented Sep 5, 2018

Uh oh!

maropu commented Sep 6, 2018

Uh oh!

mgaido91 commented Sep 6, 2018

Uh oh!

SparkQA commented Sep 6, 2018

Uh oh!

maropu commented Sep 7, 2018

Uh oh!

mgaido91 commented Sep 7, 2018

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

maropu Sep 8, 2018 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

mgaido91 Sep 8, 2018 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

mgaido91 commented Apr 6, 2018 •

edited

Loading

maropu Sep 8, 2018 •

edited

Loading

mgaido91 Sep 8, 2018 •

edited

Loading