## What changes were proposed in this pull request?
This PR adds check whether Java code generated by Catalyst can be compiled by `janino` correctly or not into `TPCDSQuerySuite`. Before this PR, this suite only checks whether analysis can be performed correctly or not.
This check will be able to avoid unexpected performance degrade by interpreter execution due to a Java compilation error.
## How was this patch tested?
Existing a test case, but updated it.
Author: Kazuaki Ishizaki <ishizaki@jp.ibm.com>
Closes#19971 from kiszk/SPARK-22774.
## What changes were proposed in this pull request?
`ColumnVector.anyNullsSet` is not called anywhere except tests, and we can easily replace it with `ColumnVector.numNulls > 0`
## How was this patch tested?
existing tests
Author: Wenchen Fan <wenchen@databricks.com>
Closes#19980 from cloud-fan/minor.
## What changes were proposed in this pull request?
These dictionary related APIs are special to `WritableColumnVector` and should not be in `ColumnVector`, which will be public soon.
## How was this patch tested?
existing tests
Author: Wenchen Fan <wenchen@databricks.com>
Closes#19970 from cloud-fan/final.
SQLConf allows some callers to define a custom default value for
configs, and that complicates a little bit the handling of fallback
config entries, since most of the default value resolution is
hidden by the config code.
This change peaks into the internals of these fallback configs
to figure out the correct default value, and also returns the
current human-readable default when showing the default value
(e.g. through "set -v").
Author: Marcelo Vanzin <vanzin@cloudera.com>
Closes#19974 from vanzin/SPARK-22779.
## What changes were proposed in this pull request?
This PR provides DataSourceV2 API support for structured streaming, including new pieces needed to support continuous processing [SPARK-20928]. High level summary:
- DataSourceV2 includes new mixins to support micro-batch and continuous reads and writes. For reads, we accept an optional user specified schema rather than using the ReadSupportWithSchema model, because doing so would severely complicate the interface.
- DataSourceV2Reader includes new interfaces to read a specific microbatch or read continuously from a given offset. These follow the same setter pattern as the existing Supports* mixins so that they can work with SupportsScanUnsafeRow.
- DataReader (the per-partition reader) has a new subinterface ContinuousDataReader only for continuous processing. This reader has a special method to check progress, and next() blocks for new input rather than returning false.
- Offset, an abstract representation of position in a streaming query, is ported to the public API. (Each type of reader will define its own Offset implementation.)
- DataSourceV2Writer has a new subinterface ContinuousWriter only for continuous processing. Commits to this interface come tagged with an epoch number, as the execution engine will continue to produce new epoch commits as the task continues indefinitely.
Note that this PR does not propose to change the existing DataSourceV2 batch API, or deprecate the existing streaming source/sink internal APIs in spark.sql.execution.streaming.
## How was this patch tested?
Toy implementations of the new interfaces with unit tests.
Author: Jose Torres <jose@databricks.com>
Closes#19925 from joseph-torres/continuous-api.
## What changes were proposed in this pull request?
This pr fixed a compilation error of TPCDS `q75`/`q77` caused by #19813;
```
java.util.concurrent.ExecutionException: org.codehaus.commons.compiler.CompileException: File 'generated.java', Line 371, Column 16: failed to compile: org.codehaus.commons.compiler.CompileException: File 'generated.java', Line 371, Column 16: Expression "bhj_matched" is not an rvalue
at com.google.common.util.concurrent.AbstractFuture$Sync.getValue(AbstractFuture.java:306)
at com.google.common.util.concurrent.AbstractFuture$Sync.get(AbstractFuture.java:293)
at com.google.common.util.concurrent.AbstractFuture.get(AbstractFuture.java:116)
at com.google.common.util.concurrent.Uninterruptibles.getUninterruptibly(Uninterruptibles.java:135)
```
## How was this patch tested?
Manually checked `q75`/`q77` can be properly compiled
Author: Takeshi Yamamuro <yamamuro@apache.org>
Closes#19969 from maropu/SPARK-22600-FOLLOWUP.
## What changes were proposed in this pull request?
See jira description for the bug : https://issues.apache.org/jira/browse/SPARK-22042
Fix done in this PR is: In `EnsureRequirements`, apply `ReorderJoinPredicates` over the input tree before doing its core logic. Since the tree is transformed bottom-up, we can assure that the children are resolved before doing `ReorderJoinPredicates`.
Theoretically this will guarantee to cover all such cases while keeping the code simple. My small grudge is for cosmetic reasons. This PR will look weird given that we don't call rules from other rules (not to my knowledge). I could have moved all the logic for `ReorderJoinPredicates` into `EnsureRequirements` but that will make it a but crowded. I am happy to discuss if there are better options.
## How was this patch tested?
Added a new test case
Author: Tejas Patil <tejasp@fb.com>
Closes#19257 from tejasapatil/SPARK-22042_ReorderJoinPredicates.
## What changes were proposed in this pull request?
We need to add some helper code to make testing ML transformers & models easier with streaming data. These tests might help us catch any remaining issues and we could encourage future PRs to use these tests to prevent new Models & Transformers from having issues.
I add a `MLTest` trait which extends `StreamTest` trait, and override `createSparkSession`. So ML testsuite can only extend `MLTest`, to use both ML & Stream test util functions.
I only modify one testcase in `LinearRegressionSuite`, for first pass review.
Link to #19746
## How was this patch tested?
`MLTestSuite` added.
Author: WeichenXu <weichen.xu@databricks.com>
Closes#19843 from WeichenXu123/ml_stream_test_helper.
## What changes were proposed in this pull request?
SPARK-22543 fixes the 64kb compile error for deeply nested expression for non-wholestage codegen. This PR extends it to support wholestage codegen.
This patch brings some util methods in to extract necessary parameters for an expression if it is split to a function.
The util methods are put in object `ExpressionCodegen` under `codegen`. The main entry is `getExpressionInputParams` which returns all necessary parameters to evaluate the given expression in a split function.
This util methods can be used to split expressions too. This is a TODO item later.
## How was this patch tested?
Added test.
Author: Liang-Chi Hsieh <viirya@gmail.com>
Closes#19813 from viirya/reduce-expr-code-for-wholestage.
## What changes were proposed in this pull request?
We have two methods to reference an object `addReferenceMinorObj` and `addReferenceObj `. The latter creates a new global variable, which means new entries in the constant pool.
The PR unifies the two method in a single `addReferenceObj` which returns the code to access the object in the `references` array and doesn't add new mutable states.
## How was this patch tested?
added UTs.
Author: Marco Gaido <mgaido@hortonworks.com>
Closes#19916 from mgaido91/SPARK-22716.
In order to enable truncate for PostgreSQL databases in Spark JDBC, a change is needed to the query used for truncating a PostgreSQL table. By default, PostgreSQL will automatically truncate any descendant tables if a TRUNCATE query is executed. As this may result in (unwanted) side-effects, the query used for the truncate should be specified separately for PostgreSQL, specifying only to TRUNCATE a single table.
## What changes were proposed in this pull request?
Add `getTruncateQuery` function to `JdbcDialect.scala`, with default query. Overridden this function for PostgreSQL to only truncate a single table. Also sets `isCascadingTruncateTable` to false, as this will allow truncates for PostgreSQL.
## How was this patch tested?
Existing tests all pass. Added test for `getTruncateQuery`
Author: Daniel van der Ende <daniel.vanderende@gmail.com>
Closes#19911 from danielvdende/SPARK-22717.
## What changes were proposed in this pull request?
In the previous PRs, https://github.com/apache/spark/pull/17832 and https://github.com/apache/spark/pull/17835 , we convert `TIMESTAMP WITH TIME ZONE` and `TIME WITH TIME ZONE` to `TIMESTAMP` for all the JDBC sources. However, this conversion could be risky since it does not respect our SQL configuration `spark.sql.session.timeZone`.
In addition, each vendor might have different semantics for these two types. For example, Postgres simply returns `TIMESTAMP` types for `TIMESTAMP WITH TIME ZONE`. For such supports, we should do it case by case. This PR reverts the general support of `TIMESTAMP WITH TIME ZONE` and `TIME WITH TIME ZONE` for JDBC sources, except ORACLE Dialect.
When supporting the ORACLE's `TIMESTAMP WITH TIME ZONE`, we only support it when the JVM default timezone is the same as the user-specified configuration `spark.sql.session.timeZone` (whose default is the JVM default timezone). Now, we still treat `TIMESTAMP WITH TIME ZONE` as `TIMESTAMP` when fetching the values via the Oracle JDBC connector, whose client converts the timestamp values with time zone to the timestamp values using the local JVM default timezone (a test case is added to `OracleIntegrationSuite.scala` in this PR for showing the behavior). Thus, to avoid any future behavior change, we will not support it if JVM default timezone is different from `spark.sql.session.timeZone`
No regression because the previous two PRs were just merged to be unreleased master branch.
## How was this patch tested?
Added the test cases
Author: gatorsmile <gatorsmile@gmail.com>
Closes#19939 from gatorsmile/timezoneUpdate.
## What changes were proposed in this pull request?
Before we deliver the Hive compatibility mode, we plan to write a set of test cases that can be easily run in both Spark and Hive sides. We can easily compare whether they are the same or not. When new typeCoercion rules are added, we also can easily track the changes. These test cases can also be backported to the previous Spark versions for determining the changes we made.
This PR is the first attempt for improving the test coverage for type coercion compatibility. We generate these test cases for our binary comparison and ImplicitTypeCasts based on the Apache Derby test cases in https://github.com/apache/derby/blob/10.14/java/testing/org/apache/derbyTesting/functionTests/tests/lang/implicitConversions.sql
## How was this patch tested?
N/A
Author: gatorsmile <gatorsmile@gmail.com>
Closes#19918 from gatorsmile/typeCoercionTests.
## What changes were proposed in this pull request?
During https://github.com/apache/spark/pull/19882, `conf` is mistakenly used to switch ORC implementation between `native` and `hive`. To affect `OrcTest` correctly, `spark.conf` should be used.
## How was this patch tested?
Pass the tests.
Author: Dongjoon Hyun <dongjoon@apache.org>
Closes#19931 from dongjoon-hyun/SPARK-22672-2.
## What changes were proposed in this pull request?
Int96 data written by impala vs data written by hive & spark is stored slightly differently -- they use a different offset for the timezone. This adds an option "spark.sql.parquet.int96TimestampConversion" (false by default) to adjust timestamps if and only if the writer is impala (or more precisely, if the parquet file's "createdBy" metadata does not start with "parquet-mr"). This matches the existing behavior in hive from HIVE-9482.
## How was this patch tested?
Unit test added, existing tests run via jenkins.
Author: Imran Rashid <irashid@cloudera.com>
Author: Henry Robinson <henry@apache.org>
Closes#19769 from squito/SPARK-12297_skip_conversion.
## What changes were proposed in this pull request?
#19416 changed the format in which rows were encoded in the state store. However, this can break existing streaming queries with the old format in unpredictable ways (potentially crashing the JVM). Hence I am reverting this for now. This will be re-applied in the future after we start saving more metadata in checkpoints to signify which version of state row format the existing streaming query is running. Then we can decode old and new formats accordingly.
## How was this patch tested?
Existing tests.
Author: Tathagata Das <tathagata.das1565@gmail.com>
Closes#19924 from tdas/SPARK-22187-1.
## What changes were proposed in this pull request?
This PR support for pushing down filters for DateType in ORC
## How was this patch tested?
Pass the Jenkins with newly add and updated test cases.
Author: Dongjoon Hyun <dongjoon@apache.org>
Closes#18995 from dongjoon-hyun/SPARK-21787.
…a-2.12 and JDK9
## What changes were proposed in this pull request?
Some compile error after upgrading to scala-2.12
```javascript
spark_source/core/src/main/scala/org/apache/spark/executor/Executor.scala:455: ambiguous reference to overloaded definition, method limit in class ByteBuffer of type (x$1: Int)java.nio.ByteBuffer
method limit in class Buffer of type ()Int
match expected type ?
val resultSize = serializedDirectResult.limit
error
```
The limit method was moved from ByteBuffer to the superclass Buffer and it can no longer be called without (). The same reason for position method.
```javascript
/home/zly/prj/oss/jdk9_HOS_SOURCE/spark_source/sql/hive/src/main/scala/org/apache/spark/sql/hive/execution/ScriptTransformationExec.scala:427: ambiguous reference to overloaded definition, [error] both method putAll in class Properties of type (x$1: java.util.Map[_, _])Unit [error] and method putAll in class Hashtable of type (x$1: java.util.Map[_ <: Object, _ <: Object])Unit [error] match argument types (java.util.Map[String,String])
[error] props.putAll(outputSerdeProps.toMap.asJava)
[error] ^
```
This is because the key type is Object instead of String which is unsafe.
## How was this patch tested?
running tests
Please review http://spark.apache.org/contributing.html before opening a pull request.
Author: kellyzly <kellyzly@126.com>
Closes#19854 from kellyzly/SPARK-22660.
## What changes were proposed in this pull request?
To support vectorization in native OrcFileFormat later, we need to use `buildReaderWithPartitionValues` instead of `buildReader` like ParquetFileFormat. This PR replaces `buildReader` with `buildReaderWithPartitionValues`.
## How was this patch tested?
Pass the Jenkins with the existing test cases.
Author: Dongjoon Hyun <dongjoon@apache.org>
Closes#19907 from dongjoon-hyun/SPARK-ORC-BUILD-READER.
- Implemented methods getInt, getLong, getBoolean for DataSourceV2Options
- Added new unit tests to exercise these methods
Author: Sunitha Kambhampati <skambha@us.ibm.com>
Closes#19902 from skambha/spark22452.
## What changes were proposed in this pull request?
Similar to https://github.com/apache/spark/pull/19842 , we should also make `ColumnarRow` an immutable view, and move forward to make `ColumnVector` public.
## How was this patch tested?
Existing tests.
The performance concern should be same as https://github.com/apache/spark/pull/19842 .
Author: Wenchen Fan <wenchen@databricks.com>
Closes#19898 from cloud-fan/row-id.
## What changes were proposed in this pull request?
Since SPARK-20682, we have two `OrcFileFormat`s. This PR refactors ORC tests with three principles (with a few exceptions)
1. Move test suite into `sql/core`.
2. Create `HiveXXX` test suite in `sql/hive` by reusing `sql/core` test suite.
3. `OrcTest` will provide common helper functions and `val orcImp: String`.
**Test Suites**
*Native OrcFileFormat*
- org.apache.spark.sql.hive.orc
- OrcFilterSuite
- OrcPartitionDiscoverySuite
- OrcQuerySuite
- OrcSourceSuite
- o.a.s.sql.hive.orc
- OrcHadoopFsRelationSuite
*Hive built-in OrcFileFormat*
- o.a.s.sql.hive.orc
- HiveOrcFilterSuite
- HiveOrcPartitionDiscoverySuite
- HiveOrcQuerySuite
- HiveOrcSourceSuite
- HiveOrcHadoopFsRelationSuite
**Hierarchy**
```
OrcTest
-> OrcSuite
-> OrcSourceSuite
-> OrcQueryTest
-> OrcQuerySuite
-> OrcPartitionDiscoveryTest
-> OrcPartitionDiscoverySuite
-> OrcFilterSuite
HadoopFsRelationTest
-> OrcHadoopFsRelationSuite
-> HiveOrcHadoopFsRelationSuite
```
Please note the followings.
- Unlike the other test suites, `OrcHadoopFsRelationSuite` doesn't inherit `OrcTest`. It is inside `sql/hive` like `ParquetHadoopFsRelationSuite` due to the dependencies and follows the existing convention to use `val dataSourceName: String`
- `OrcFilterSuite`s cannot reuse test cases due to the different function signatures using Hive 1.2.1 ORC classes and Apache ORC 1.4.1 classes.
## How was this patch tested?
Pass the Jenkins tests with reorganized test suites.
Author: Dongjoon Hyun <dongjoon@apache.org>
Closes#19882 from dongjoon-hyun/SPARK-22672.
## What changes were proposed in this pull request?
There was a bug in Univocity Parser that causes the issue in SPARK-22516. This was fixed by upgrading from 2.5.4 to 2.5.9 version of the library :
**Executing**
```
spark.read.option("header","true").option("inferSchema", "true").option("multiLine", "true").option("comment", "g").csv("test_file_without_eof_char.csv").show()
```
**Before**
```
ERROR Executor: Exception in task 0.0 in stage 6.0 (TID 6)
com.univocity.parsers.common.TextParsingException: java.lang.IllegalArgumentException - Unable to skip 1 lines from line 2. End of input reached
...
Internal state when error was thrown: line=3, column=0, record=2, charIndex=31
at com.univocity.parsers.common.AbstractParser.handleException(AbstractParser.java:339)
at com.univocity.parsers.common.AbstractParser.parseNext(AbstractParser.java:475)
at org.apache.spark.sql.execution.datasources.csv.UnivocityParser$$anon$1.next(UnivocityParser.scala:281)
at scala.collection.Iterator$$anon$11.next(Iterator.scala:409)
```
**After**
```
+-------+-------+
|column1|column2|
+-------+-------+
| abc| def|
+-------+-------+
```
## How was this patch tested?
The already existing `CSVSuite.commented lines in CSV data` test was extended to parse the file also in multiline mode. The test input file was modified to also include a comment in the last line.
Author: smurakozi <smurakozi@gmail.com>
Closes#19906 from smurakozi/SPARK-22516.
## What changes were proposed in this pull request?
This is a follow-up of https://github.com/apache/spark/pull/19871 to improve an exception message.
## How was this patch tested?
Pass the Jenkins.
Author: Dongjoon Hyun <dongjoon@apache.org>
Closes#19903 from dongjoon-hyun/orc_exception.
## What changes were proposed in this pull request?
The SQL `Analyzer` goes through a whole query plan even most part of it is analyzed. This increases the time spent on query analysis for long pipelines in ML, especially.
This patch adds a logical node called `AnalysisBarrier` that wraps an analyzed logical plan to prevent it from analysis again. The barrier is applied to the analyzed logical plan in `Dataset`. It won't change the output of wrapped logical plan and just acts as a wrapper to hide it from analyzer. New operations on the dataset will be put on the barrier, so only the new nodes created will be analyzed.
This analysis barrier will be removed at the end of analysis stage.
## How was this patch tested?
Added tests.
Author: Liang-Chi Hsieh <viirya@gmail.com>
Closes#19873 from viirya/SPARK-20392-reopen.
## What changes were proposed in this pull request?
During [SPARK-22488](https://github.com/apache/spark/pull/19713) to fix view resolution issue, there occurs a regression at `2.2.1` and `master` branch like the following. This PR fixes that.
```scala
scala> spark.version
res2: String = 2.2.1
scala> sql("DROP TABLE IF EXISTS t").show
17/12/04 21:01:06 WARN DropTableCommand: org.apache.spark.sql.AnalysisException:
Table or view not found: t;
org.apache.spark.sql.AnalysisException: Table or view not found: t;
```
## How was this patch tested?
Manual.
Author: Dongjoon Hyun <dongjoon@apache.org>
Closes#19888 from dongjoon-hyun/SPARK-22686.
## What changes were proposed in this pull request?
This PR aims to provide a configuration to choose the default `OrcFileFormat` from legacy `sql/hive` module or new `sql/core` module.
For example, this configuration will affects the following operations.
```scala
spark.read.orc(...)
```
```sql
CREATE TABLE t
USING ORC
...
```
## How was this patch tested?
Pass the Jenkins with new test suites.
Author: Dongjoon Hyun <dongjoon@apache.org>
Closes#19871 from dongjoon-hyun/spark-sql-orc-enabled.
## What changes were proposed in this pull request?
PropagateTypes are called twice in TypeCoercion. We do not need to call it twice. Instead, we should call it after each change on the types.
## How was this patch tested?
The existing tests
Author: gatorsmile <gatorsmile@gmail.com>
Closes#19874 from gatorsmile/deduplicatePropagateTypes.
## What changes were proposed in this pull request?
The `HashAggregateExec` whole stage codegen path is a little messy and hard to understand, this code cleans it up a little bit, especially for the fast hash map part.
## How was this patch tested?
existing tests
Author: Wenchen Fan <wenchen@databricks.com>
Closes#19869 from cloud-fan/hash-agg.
## What changes were proposed in this pull request?
Since [SPARK-2883](https://issues.apache.org/jira/browse/SPARK-2883), Apache Spark supports Apache ORC inside `sql/hive` module with Hive dependency. This PR aims to add a new ORC data source inside `sql/core` and to replace the old ORC data source eventually. This PR resolves the following three issues.
- [SPARK-20682](https://issues.apache.org/jira/browse/SPARK-20682): Add new ORCFileFormat based on Apache ORC 1.4.1
- [SPARK-15474](https://issues.apache.org/jira/browse/SPARK-15474): ORC data source fails to write and read back empty dataframe
- [SPARK-21791](https://issues.apache.org/jira/browse/SPARK-21791): ORC should support column names with dot
## How was this patch tested?
Pass the Jenkins with the existing all tests and new tests for SPARK-15474 and SPARK-21791.
Author: Dongjoon Hyun <dongjoon@apache.org>
Author: Wenchen Fan <wenchen@databricks.com>
Closes#19651 from dongjoon-hyun/SPARK-20682.
## What changes were proposed in this pull request?
Use a separate Spark event queue for StreamingQueryListenerBus so that if there are many non-streaming events, streaming query listeners don't need to wait for other Spark listeners and can catch up.
## How was this patch tested?
Jenkins
Author: Shixiong Zhu <zsxwing@gmail.com>
Closes#19838 from zsxwing/SPARK-22638.
## What changes were proposed in this pull request?
When user tries to load data with a non existing hdfs file path system is not validating it and the load command operation is getting successful.
This is misleading to the user. already there is a validation in the scenario of none existing local file path. This PR has added validation in the scenario of nonexisting hdfs file path
## How was this patch tested?
UT has been added for verifying the issue, also snapshots has been added after the verification in a spark yarn cluster
Author: sujith71955 <sujithchacko.2010@gmail.com>
Closes#19823 from sujith71955/master_LoadComand_Issue.
## What changes were proposed in this pull request?
This PR introduces a way to explicitly range-partition a Dataset. So far, only round-robin and hash partitioning were possible via `df.repartition(...)`, but sometimes range partitioning might be desirable: e.g. when writing to disk, for better compression without the cost of global sort.
The current implementation piggybacks on the existing `RepartitionByExpression` `LogicalPlan` and simply adds the following logic: If its expressions are of type `SortOrder`, then it will do `RangePartitioning`; otherwise `HashPartitioning`. This was by far the least intrusive solution I could come up with.
## How was this patch tested?
Unit test for `RepartitionByExpression` changes, a test to ensure we're not changing the behavior of existing `.repartition()` and a few end-to-end tests in `DataFrameSuite`.
Author: Adrian Ionescu <adrian@databricks.com>
Closes#19828 from adrian-ionescu/repartitionByRange.
## What changes were proposed in this pull request?
How to reproduce:
```scala
import org.apache.spark.sql.execution.joins.BroadcastHashJoinExec
spark.createDataFrame(Seq((1, "4"), (2, "2"))).toDF("key", "value").createTempView("table1")
spark.createDataFrame(Seq((1, "1"), (2, "2"))).toDF("key", "value").createTempView("table2")
val bl = sql("SELECT /*+ MAPJOIN(t1) */ * FROM table1 t1 JOIN table2 t2 ON t1.key = t2.key").queryExecution.executedPlan
println(bl.children.head.asInstanceOf[BroadcastHashJoinExec].buildSide)
```
The result is `BuildRight`, but should be `BuildLeft`. This PR fix this issue.
## How was this patch tested?
unit tests
Author: Yuming Wang <wgyumg@gmail.com>
Closes#19714 from wangyum/SPARK-22489.
## What changes were proposed in this pull request?
To make `ColumnVector` public, `ColumnarArray` need to be public too, and we should not have mutable public fields in a public class. This PR proposes to make `ColumnarArray` an immutable view of the data, and always create a new instance of `ColumnarArray` in `ColumnVector#getArray`
## How was this patch tested?
new benchmark in `ColumnarBatchBenchmark`
Author: Wenchen Fan <wenchen@databricks.com>
Closes#19842 from cloud-fan/column-vector.
## What changes were proposed in this pull request?
As a step to make `ColumnVector` public, the `ColumnarRow` returned by `ColumnVector#getStruct` should be immutable.
However we do need the mutability of `ColumnaRow` for the fast vectorized hashmap in hash aggregate. To solve this, this PR introduces a `MutableColumnarRow` for this use case.
## How was this patch tested?
existing test.
Author: Wenchen Fan <wenchen@databricks.com>
Closes#19847 from cloud-fan/mutable-row.
## What changes were proposed in this pull request?
Currently, in the optimize rule `PropagateEmptyRelation`, the following cases is not handled:
1. empty relation as right child in left outer join
2. empty relation as left child in right outer join
3. empty relation as right child in left semi join
4. empty relation as right child in left anti join
5. only one empty relation in full outer join
case 1 / 2 / 5 can be treated as **Cartesian product** and cause exception. See the new test cases.
## How was this patch tested?
Unit test
Author: Wang Gengliang <ltnwgl@gmail.com>
Closes#19825 from gengliangwang/SPARK-22615.
## What changes were proposed in this pull request?
For SQL write jobs, we only set metrics for the SQL listener and display them in the SQL plan UI. We should also set metrics for Spark task output metrics, which will be shown in spark job UI.
## How was this patch tested?
test it manually. For a simple write job
```
spark.range(1000).write.parquet("/tmp/p1")
```
now the spark job UI looks like
Author: Wenchen Fan <wenchen@databricks.com>
Closes#19833 from cloud-fan/ui.
## What changes were proposed in this pull request?
`CatalogImpl.refreshTable` uses `foreach(..)` to refresh all tables in a view. This traverses all nodes in the subtree and calls `LogicalPlan.refresh()` on these nodes. However `LogicalPlan.refresh()` is also refreshing its children, as a result refreshing a large view can be quite expensive.
This PR just calls `LogicalPlan.refresh()` on the top node.
## How was this patch tested?
Existing tests.
Author: Herman van Hovell <hvanhovell@databricks.com>
Closes#19837 from hvanhovell/SPARK-22637.
## What changes were proposed in this pull request?
* JIRA: [SPARK-22431](https://issues.apache.org/jira/browse/SPARK-22431) : Creating Permanent view with illegal type
**Description:**
- It is possible in Spark SQL to create a permanent view that uses an nested field with an illegal name.
- For example if we create the following view:
```create view x as select struct('a' as `$q`, 1 as b) q```
- A simple select fails with the following exception:
```
select * from x;
org.apache.spark.SparkException: Cannot recognize hive type string: struct<$q:string,b:int>
at org.apache.spark.sql.hive.client.HiveClientImpl$.fromHiveColumn(HiveClientImpl.scala:812)
at org.apache.spark.sql.hive.client.HiveClientImpl$$anonfun$getTableOption$1$$anonfun$apply$11$$anonfun$7.apply(HiveClientImpl.scala:378)
at org.apache.spark.sql.hive.client.HiveClientImpl$$anonfun$getTableOption$1$$anonfun$apply$11$$anonfun$7.apply(HiveClientImpl.scala:378)
...
```
**Issue/Analysis**: Right now, we can create a view with a schema that cannot be read back by Spark from the Hive metastore. For more details, please see the discussion about the analysis and proposed fix options in comment 1 and comment 2 in the [SPARK-22431](https://issues.apache.org/jira/browse/SPARK-22431)
**Proposed changes**:
- Fix the hive table/view codepath to check whether the schema datatype is parseable by Spark before persisting it in the metastore. This change is localized to HiveClientImpl to do the check similar to the check in FromHiveColumn. This is fail-fast and we will avoid the scenario where we write something to the metastore that we are unable to read it back.
- Added new unit tests
- Ran the sql related unit test suites ( hive/test, sql/test, catalyst/test) OK
With the fix:
```
create view x as select struct('a' as `$q`, 1 as b) q;
17/11/28 10:44:55 ERROR SparkSQLDriver: Failed in [create view x as select struct('a' as `$q`, 1 as b) q]
org.apache.spark.SparkException: Cannot recognize hive type string: struct<$q:string,b:int>
at org.apache.spark.sql.hive.client.HiveClientImpl$.org$apache$spark$sql$hive$client$HiveClientImpl$$getSparkSQLDataType(HiveClientImpl.scala:884)
at org.apache.spark.sql.hive.client.HiveClientImpl$$anonfun$org$apache$spark$sql$hive$client$HiveClientImpl$$verifyColumnDataType$1.apply(HiveClientImpl.scala:906)
at org.apache.spark.sql.hive.client.HiveClientImpl$$anonfun$org$apache$spark$sql$hive$client$HiveClientImpl$$verifyColumnDataType$1.apply(HiveClientImpl.scala:906)
at scala.collection.Iterator$class.foreach(Iterator.scala:893)
...
```
## How was this patch tested?
- New unit tests have been added.
hvanhovell, Please review and share your thoughts/comments. Thank you so much.
Author: Sunitha Kambhampati <skambha@us.ibm.com>
Closes#19747 from skambha/spark22431.
## What changes were proposed in this pull request?
Currently, relation size is computed as the sum of file size, which is error-prone because storage format like parquet may have a much smaller file size compared to in-memory size. When we choose broadcast join based on file size, there's a risk of OOM. But if the number of rows is available in statistics, we can get a better estimation by `numRows * rowSize`, which helps to alleviate this problem.
## How was this patch tested?
Added a new test case for data source table and hive table.
Author: Zhenhua Wang <wzh_zju@163.com>
Author: Zhenhua Wang <wangzhenhua@huawei.com>
Closes#19743 from wzhfy/better_leaf_size.
## What changes were proposed in this pull request?
When converting Pandas DataFrame/Series from/to Spark DataFrame using `toPandas()` or pandas udfs, timestamp values behave to respect Python system timezone instead of session timezone.
For example, let's say we use `"America/Los_Angeles"` as session timezone and have a timestamp value `"1970-01-01 00:00:01"` in the timezone. Btw, I'm in Japan so Python timezone would be `"Asia/Tokyo"`.
The timestamp value from current `toPandas()` will be the following:
```
>>> spark.conf.set("spark.sql.session.timeZone", "America/Los_Angeles")
>>> df = spark.createDataFrame([28801], "long").selectExpr("timestamp(value) as ts")
>>> df.show()
+-------------------+
| ts|
+-------------------+
|1970-01-01 00:00:01|
+-------------------+
>>> df.toPandas()
ts
0 1970-01-01 17:00:01
```
As you can see, the value becomes `"1970-01-01 17:00:01"` because it respects Python timezone.
As we discussed in #18664, we consider this behavior is a bug and the value should be `"1970-01-01 00:00:01"`.
## How was this patch tested?
Added tests and existing tests.
Author: Takuya UESHIN <ueshin@databricks.com>
Closes#19607 from ueshin/issues/SPARK-22395.
## What changes were proposed in this pull request?
In PySpark API Document, DataFrame.write.csv() says that setting the quote parameter to an empty string should turn off quoting. Instead, it uses the [null character](https://en.wikipedia.org/wiki/Null_character) as the quote.
This PR fixes the doc.
## How was this patch tested?
Manual.
```
cd python/docs
make html
open _build/html/pyspark.sql.html
```
Author: gaborgsomogyi <gabor.g.somogyi@gmail.com>
Closes#19814 from gaborgsomogyi/SPARK-22484.
## What changes were proposed in this pull request?
Code generation is disabled for CaseWhen when the number of branches is higher than `spark.sql.codegen.maxCaseBranches` (which defaults to 20). This was done to prevent the well known 64KB method limit exception.
This PR proposes to support code generation also in those cases (without causing exceptions of course). As a side effect, we could get rid of the `spark.sql.codegen.maxCaseBranches` configuration.
## How was this patch tested?
existing UTs
Author: Marco Gaido <mgaido@hortonworks.com>
Author: Marco Gaido <marcogaido91@gmail.com>
Closes#19752 from mgaido91/SPARK-22520.
## What changes were proposed in this pull request?
Currently, relation stats is the same whether cbo is enabled or not. While relation (`LogicalRelation` or `HiveTableRelation`) is a `LogicalPlan`, its behavior is inconsistent with other plans. This can cause confusion when user runs EXPLAIN COST commands. Besides, when CBO is disabled, we apply the size-only estimation strategy, so there's no need to propagate other catalog statistics to relation.
## How was this patch tested?
Enhanced existing tests case and added a test case.
Author: Zhenhua Wang <wangzhenhua@huawei.com>
Closes#19757 from wzhfy/catalog_stats_conversion.
## What changes were proposed in this pull request?
`ColumnVector#loadBytes` is only used as an optimization for reading UTF8String in `WritableColumnVector`, this PR moves this optimization to `WritableColumnVector` and simplified it.
## How was this patch tested?
existing test
Author: Wenchen Fan <wenchen@databricks.com>
Closes#19815 from cloud-fan/load-bytes.
## What changes were proposed in this pull request?
`nullsNativeAddress` and `valuesNativeAddress` are only used in tests and benchmark, no need to be top class API.
## How was this patch tested?
existing tests
Author: Wenchen Fan <wenchen@databricks.com>
Closes#19818 from cloud-fan/minor.
## What changes were proposed in this pull request?
`ctx.currentVars` means the input variables for the current operator, which is already decided in `CodegenSupport`, we can set it there instead of `doConsume`.
also add more comments to help people understand the codegen framework.
After this PR, we now have a principle about setting `ctx.currentVars` and `ctx.INPUT_ROW`:
1. for non-whole-stage-codegen path, never set them. (permit some special cases like generating ordering)
2. for whole-stage-codegen `produce` path, mostly we don't need to set them, but blocking operators may need to set them for expressions that produce data from data source, sort buffer, aggregate buffer, etc.
3. for whole-stage-codegen `consume` path, mostly we don't need to set them because `currentVars` is automatically set to child input variables and `INPUT_ROW` is mostly not used. A few plans need to tweak them as they may have different inputs, or they use the input row.
## How was this patch tested?
existing tests.
Author: Wenchen Fan <wenchen@databricks.com>
Closes#19803 from cloud-fan/codegen.
## What changes were proposed in this pull request?
A frequently reported issue of Spark is the Java 64kb compile error. This is because Spark generates a very big method and it's usually caused by 3 reasons:
1. a deep expression tree, e.g. a very complex filter condition
2. many individual expressions, e.g. expressions can have many children, operators can have many expressions.
3. a deep query plan tree (with whole stage codegen)
This PR focuses on 1. There are already several patches(#15620#18972#18641) trying to fix this issue and some of them are already merged. However this is an endless job as every non-leaf expression has this issue.
This PR proposes to fix this issue in `Expression.genCode`, to make sure the code for a single expression won't grow too big.
According to maropu 's benchmark, no regression is found with TPCDS (thanks maropu !): https://docs.google.com/spreadsheets/d/1K3_7lX05-ZgxDXi9X_GleNnDjcnJIfoSlSCDZcL4gdg/edit?usp=sharing
## How was this patch tested?
existing test
Author: Wenchen Fan <wenchen@databricks.com>
Author: Wenchen Fan <cloud0fan@gmail.com>
Closes#19767 from cloud-fan/codegen.
## What changes were proposed in this pull request?
Let’s say I have a nested AND expression shown below and p2 can not be pushed down,
(p1 AND p2) OR p3
In current Spark code, during data source filter translation, (p1 AND p2) is returned as p1 only and p2 is simply lost. This issue occurs with JDBC data source and is similar to [SPARK-12218](https://github.com/apache/spark/pull/10362) for Parquet. When we have AND nested below another expression, we should either push both legs or nothing.
Note that:
- The current Spark code will always split conjunctive predicate before it determines if a predicate can be pushed down or not
- If I have (p1 AND p2) AND p3, it will be split into p1, p2, p3. There won't be nested AND expression.
- The current Spark code logic for OR is OK. It either pushes both legs or nothing.
The same translation method is also called by Data Source V2.
## How was this patch tested?
Added new unit test cases to JDBCSuite
gatorsmile
Author: Jia Li <jiali@us.ibm.com>
Closes#19776 from jliwork/spark-22548.
## What changes were proposed in this pull request?
Added the histogram representation to the output of the `DESCRIBE EXTENDED table_name column_name` command.
## How was this patch tested?
Modified SQL UT and checked output
Please review http://spark.apache.org/contributing.html before opening a pull request.
Author: Marco Gaido <mgaido@hortonworks.com>
Closes#19774 from mgaido91/SPARK-22475.
## What changes were proposed in this pull request?
This PR is to clean the usage of addMutableState and splitExpressions
1. replace hardcoded type string to ctx.JAVA_BOOLEAN etc.
2. create a default value of the initCode for ctx.addMutableStats
3. Use named arguments when calling `splitExpressions `
## How was this patch tested?
The existing test cases
Author: gatorsmile <gatorsmile@gmail.com>
Closes#19790 from gatorsmile/codeClean.
This PR enables to use ``OffHeapColumnVector`` when ``spark.sql.columnVector.offheap.enable`` is set to ``true``. While ``ColumnVector`` has two implementations ``OnHeapColumnVector`` and ``OffHeapColumnVector``, only ``OnHeapColumnVector`` is always used.
This PR implements the followings
- Pass ``OffHeapColumnVector`` to ``ColumnarBatch.allocate()`` when ``spark.sql.columnVector.offheap.enable`` is set to ``true``
- Free all of off-heap memory regions by ``OffHeapColumnVector.close()``
- Ensure to call ``OffHeapColumnVector.close()``
Use existing tests
Author: Kazuaki Ishizaki <ishizaki@jp.ibm.com>
Closes#17436 from kiszk/SPARK-20101.
## What changes were proposed in this pull request?
ScalaTest 3.0 uses an implicit `Signaler`. This PR makes it sure all Spark tests uses `ThreadSignaler` explicitly which has the same default behavior of interrupting a thread on the JVM like ScalaTest 2.2.x. This will reduce potential flakiness.
## How was this patch tested?
This is testsuite-only update. This should passes the Jenkins tests.
Author: Dongjoon Hyun <dongjoon@apache.org>
Closes#19784 from dongjoon-hyun/use_thread_signaler.
## What changes were proposed in this pull request?
Pass the FileSystem created using the correct Hadoop conf into `globPathIfNecessary` so that it can pick up user's hadoop configurations, such as credentials.
## How was this patch tested?
Jenkins
Author: Shixiong Zhu <zsxwing@gmail.com>
Closes#19771 from zsxwing/fix-file-stream-conf.
## What changes were proposed in this pull request?
* Add a "function type" argument to pandas_udf.
* Add a new public enum class `PandasUdfType` in pyspark.sql.functions
* Refactor udf related code from pyspark.sql.functions to pyspark.sql.udf
* Merge "PythonUdfType" and "PythonEvalType" into a single enum class "PythonEvalType"
Example:
```
from pyspark.sql.functions import pandas_udf, PandasUDFType
pandas_udf('double', PandasUDFType.SCALAR):
def plus_one(v):
return v + 1
```
## Design doc
https://docs.google.com/document/d/1KlLaa-xJ3oz28xlEJqXyCAHU3dwFYkFs_ixcUXrJNTc/edit
## How was this patch tested?
Added PandasUDFTests
## TODO:
* [x] Implement proper enum type for `PandasUDFType`
* [x] Update documentation
* [x] Add more tests in PandasUDFTests
Author: Li Jin <ice.xelloss@gmail.com>
Closes#19630 from icexelloss/spark-22409-pandas-udf-type.
## What changes were proposed in this pull request?
`ColumnarBatch` provides features to do fast filter and project in a columnar fashion, however this feature is never used by Spark, as Spark uses whole stage codegen and processes the data in a row fashion. This PR proposes to remove these unused features as we won't switch to columnar execution in the near future. Even we do, I think this part needs a proper redesign.
This is also a step to make `ColumnVector` public, as we don't wanna expose these features to users.
## How was this patch tested?
existing tests
Author: Wenchen Fan <wenchen@databricks.com>
Closes#19766 from cloud-fan/vector.
## What changes were proposed in this pull request?
Do not include jdbc properties which may contain credentials in logging a logical plan with `SaveIntoDataSourceCommand` in it.
## How was this patch tested?
building locally and trying to reproduce (per the steps in https://issues.apache.org/jira/browse/SPARK-22479):
```
== Parsed Logical Plan ==
SaveIntoDataSourceCommand org.apache.spark.sql.execution.datasources.jdbc.JdbcRelationProvider570127fa, Map(dbtable -> test20, driver -> org.postgresql.Driver, url -> *********(redacted), password -> *********(redacted)), ErrorIfExists
+- Range (0, 100, step=1, splits=Some(8))
== Analyzed Logical Plan ==
SaveIntoDataSourceCommand org.apache.spark.sql.execution.datasources.jdbc.JdbcRelationProvider570127fa, Map(dbtable -> test20, driver -> org.postgresql.Driver, url -> *********(redacted), password -> *********(redacted)), ErrorIfExists
+- Range (0, 100, step=1, splits=Some(8))
== Optimized Logical Plan ==
SaveIntoDataSourceCommand org.apache.spark.sql.execution.datasources.jdbc.JdbcRelationProvider570127fa, Map(dbtable -> test20, driver -> org.postgresql.Driver, url -> *********(redacted), password -> *********(redacted)), ErrorIfExists
+- Range (0, 100, step=1, splits=Some(8))
== Physical Plan ==
Execute SaveIntoDataSourceCommand
+- SaveIntoDataSourceCommand org.apache.spark.sql.execution.datasources.jdbc.JdbcRelationProvider570127fa, Map(dbtable -> test20, driver -> org.postgresql.Driver, url -> *********(redacted), password -> *********(redacted)), ErrorIfExists
+- Range (0, 100, step=1, splits=Some(8))
```
Author: osatici <osatici@palantir.com>
Closes#19708 from onursatici/os/redact-jdbc-creds.
## What changes were proposed in this pull request?
This fixes a problem caused by #15880
`select '1.5' > 0.5; // Result is NULL in Spark but is true in Hive.
`
When compare string and numeric, cast them as double like Hive.
Author: liutang123 <liutang123@yeah.net>
Closes#19692 from liutang123/SPARK-22469.
## What changes were proposed in this pull request?
Logically the `Array` doesn't belong to `ColumnVector`, and `Row` doesn't belong to `ColumnarBatch`. e.g. `ColumnVector` needs to return `Array` for `getArray`, and `Row` for `getStruct`. `Array` and `Row` can return each other with the `getArray`/`getStruct` methods.
This is also a step to make `ColumnVector` public, it's cleaner to have `Array` and `Row` as top-level classes.
This PR is just code moving around, with 2 renaming: `Array` -> `VectorBasedArray`, `Row` -> `VectorBasedRow`.
## How was this patch tested?
existing tests.
Author: Wenchen Fan <wenchen@databricks.com>
Closes#19740 from cloud-fan/vector.
This change replaces the SQLListener with a new implementation that
saves the data to the same store used by the SparkContext's status
store. For that, the types used by the old SQLListener had to be
updated a bit so that they're more serialization-friendly.
The interface for getting data from the store was abstracted into
a new class, SQLAppStatusStore (following the convention used in
core).
Another change is the way that the SQL UI hooks up into the core
UI or the SHS. The old "SparkHistoryListenerFactory" was replaced
with a new "AppStatePlugin" that more explicitly differentiates
between the two use cases: processing events, and showing the UI.
Both live apps and the SHS use this new API (previously, it was
restricted to the SHS).
Note on the above: this causes a slight change of behavior for
live apps; the SQL tab will only show up after the first execution
is started.
The metrics gathering code was re-worked a bit so that the types
used are less memory hungry and more serialization-friendly. This
reduces memory usage when using in-memory stores, and reduces load
times when using disk stores.
Tested with existing and added unit tests. Note one unit test was
disabled because it depends on SPARK-20653, which isn't in yet.
Author: Marcelo Vanzin <vanzin@cloudera.com>
Closes#19681 from vanzin/SPARK-20652.
## What changes were proposed in this pull request?
Equi-height histogram is effective in cardinality estimation, and more accurate than basic column stats (min, max, ndv, etc) especially in skew distribution. So we need to support it.
For equi-height histogram, all buckets (intervals) have the same height (frequency).
In this PR, we use a two-step method to generate an equi-height histogram:
1. use `ApproximatePercentile` to get percentiles `p(0), p(1/n), p(2/n) ... p((n-1)/n), p(1)`;
2. construct range values of buckets, e.g. `[p(0), p(1/n)], [p(1/n), p(2/n)] ... [p((n-1)/n), p(1)]`, and use `ApproxCountDistinctForIntervals` to count ndv in each bucket. Each bucket is of the form: `(lowerBound, higherBound, ndv)`.
## How was this patch tested?
Added new test cases and modified some existing test cases.
Author: Zhenhua Wang <wangzhenhua@huawei.com>
Author: Zhenhua Wang <wzh_zju@163.com>
Closes#19479 from wzhfy/generate_histogram.
## What changes were proposed in this pull request?
There is a concern that Spark-side codegen row-by-row filtering might be faster than Parquet's one in general due to type-boxing and additional fuction calls which Spark's one tries to avoid.
So, this PR adds an option to disable/enable record-by-record filtering in Parquet side.
It sets the default to `false` to take the advantage of the improvement.
This was also discussed in https://github.com/apache/spark/pull/14671.
## How was this patch tested?
Manually benchmarks were performed. I generated a billion (1,000,000,000) records and tested equality comparison concatenated with `OR`. This filter combinations were made from 5 to 30.
It seem indeed Spark-filtering is faster in the test case and the gap increased as the filter tree becomes larger.
The details are as below:
**Code**
``` scala
test("Parquet-side filter vs Spark-side filter - record by record") {
withTempPath { path =>
val N = 1000 * 1000 * 1000
val df = spark.range(N).toDF("a")
df.write.parquet(path.getAbsolutePath)
val benchmark = new Benchmark("Parquet-side vs Spark-side", N)
Seq(5, 10, 20, 30).foreach { num =>
val filterExpr = (0 to num).map(i => s"a = $i").mkString(" OR ")
benchmark.addCase(s"Parquet-side filter - number of filters [$num]", 3) { _ =>
withSQLConf(SQLConf.PARQUET_VECTORIZED_READER_ENABLED.key -> false.toString,
SQLConf.PARQUET_RECORD_FILTER_ENABLED.key -> true.toString) {
// We should strip Spark-side filter to compare correctly.
stripSparkFilter(
spark.read.parquet(path.getAbsolutePath).filter(filterExpr)).count()
}
}
benchmark.addCase(s"Spark-side filter - number of filters [$num]", 3) { _ =>
withSQLConf(SQLConf.PARQUET_VECTORIZED_READER_ENABLED.key -> false.toString,
SQLConf.PARQUET_RECORD_FILTER_ENABLED.key -> false.toString) {
spark.read.parquet(path.getAbsolutePath).filter(filterExpr).count()
}
}
}
benchmark.run()
}
}
```
**Result**
```
Parquet-side vs Spark-side: Best/Avg Time(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------
Parquet-side filter - number of filters [5] 4268 / 4367 234.3 4.3 0.8X
Spark-side filter - number of filters [5] 3709 / 3741 269.6 3.7 0.9X
Parquet-side filter - number of filters [10] 5673 / 5727 176.3 5.7 0.6X
Spark-side filter - number of filters [10] 3588 / 3632 278.7 3.6 0.9X
Parquet-side filter - number of filters [20] 8024 / 8440 124.6 8.0 0.4X
Spark-side filter - number of filters [20] 3912 / 3946 255.6 3.9 0.8X
Parquet-side filter - number of filters [30] 11936 / 12041 83.8 11.9 0.3X
Spark-side filter - number of filters [30] 3929 / 3978 254.5 3.9 0.8X
```
Author: hyukjinkwon <gurwls223@gmail.com>
Closes#15049 from HyukjinKwon/SPARK-17310.
## What changes were proposed in this pull request?
This change uses Arrow to optimize the creation of a Spark DataFrame from a Pandas DataFrame. The input df is sliced according to the default parallelism. The optimization is enabled with the existing conf "spark.sql.execution.arrow.enabled" and is disabled by default.
## How was this patch tested?
Added new unit test to create DataFrame with and without the optimization enabled, then compare results.
Author: Bryan Cutler <cutlerb@gmail.com>
Author: Takuya UESHIN <ueshin@databricks.com>
Closes#19459 from BryanCutler/arrow-createDataFrame-from_pandas-SPARK-20791.
## What changes were proposed in this pull request?
This PR changes `AND` or `OR` code generation to place condition and then expressions' generated code into separated methods if these size could be large. When the method is newly generated, variables for `isNull` and `value` are declared as an instance variable to pass these values (e.g. `isNull1409` and `value1409`) to the callers of the generated method.
This PR resolved two cases:
* large code size of left expression
* large code size of right expression
## How was this patch tested?
Added a new test case into `CodeGenerationSuite`
Author: Kazuaki Ishizaki <ishizaki@jp.ibm.com>
Closes#18972 from kiszk/SPARK-21720.
## What changes were proposed in this pull request?
This PR makes Spark to be able to read Parquet TIMESTAMP_MICROS values, and add a new config to allow Spark to write timestamp values to parquet as TIMESTAMP_MICROS type.
## How was this patch tested?
new test
Author: Wenchen Fan <wenchen@databricks.com>
Closes#19702 from cloud-fan/parquet.
## What changes were proposed in this pull request?
The current internal `table()` API of `SparkSession` bypasses the Analyzer and directly calls `sessionState.catalog.lookupRelation` API. This skips the view resolution logics in our Analyzer rule `ResolveRelations`. This internal API is widely used by various DDL commands, public and internal APIs.
Users might get the strange error caused by view resolution when the default database is different.
```
Table or view not found: t1; line 1 pos 14
org.apache.spark.sql.AnalysisException: Table or view not found: t1; line 1 pos 14
at org.apache.spark.sql.catalyst.analysis.package$AnalysisErrorAt.failAnalysis(package.scala:42)
```
This PR is to fix it by enforcing it to use `ResolveRelations` to resolve the table.
## How was this patch tested?
Added a test case and modified the existing test cases
Author: gatorsmile <gatorsmile@gmail.com>
Closes#19713 from gatorsmile/viewResolution.
## What changes were proposed in this pull request?
For the few Dataset actions such as `foreach`, currently no SQL metrics are visible in the SQL tab of SparkUI. It is because it binds wrongly to Dataset's `QueryExecution`. As the actions directly evaluate on the RDD which has individual `QueryExecution`, to show correct SQL metrics on UI, we should bind to RDD's `QueryExecution`.
## How was this patch tested?
Manually test. Screenshot is attached in the PR.
Author: Liang-Chi Hsieh <viirya@gmail.com>
Closes#19689 from viirya/SPARK-22462.
## What changes were proposed in this pull request?
Fix to allow recovery on console , avoid checkpoint exception
## How was this patch tested?
existing tests
manual tests [ Replicating error and seeing no checkpoint error after fix]
Author: Rekha Joshi <rekhajoshm@gmail.com>
Author: rjoshi2 <rekhajoshm@gmail.com>
Closes#19407 from rekhajoshm/SPARK-21667.
## What changes were proposed in this pull request?
In `spark-sql` module tests there are deprecations warnings caused by the usage of deprecated methods of `java.sql.Date` and the usage of the deprecated `AsyncAssertions.Waiter` class.
This PR replace the deprecated methods of `java.sql.Date` with non-deprecated ones (using `Calendar` where needed). It replaces also the deprecated `org.scalatest.concurrent.AsyncAssertions.Waiter` with `org.scalatest.concurrent.Waiters._`.
## How was this patch tested?
existing UTs
Author: Marco Gaido <mgaido@hortonworks.com>
Closes#19696 from mgaido91/SPARK-22473.
## What changes were proposed in this pull request?
One powerful feature of `Dataset` is, we can easily map SQL rows to Scala/Java objects and do runtime null check automatically.
For example, let's say we have a parquet file with schema `<a: int, b: string>`, and we have a `case class Data(a: Int, b: String)`. Users can easily read this parquet file into `Data` objects, and Spark will throw NPE if column `a` has null values.
However the null checking is left behind for top-level primitive values. For example, let's say we have a parquet file with schema `<a: Int>`, and we read it into Scala `Int`. If column `a` has null values, we will get some weird results.
```
scala> val ds = spark.read.parquet(...).as[Int]
scala> ds.show()
+----+
|v |
+----+
|null|
|1 |
+----+
scala> ds.collect
res0: Array[Long] = Array(0, 1)
scala> ds.map(_ * 2).show
+-----+
|value|
+-----+
|-2 |
|2 |
+-----+
```
This is because internally Spark use some special default values for primitive types, but never expect users to see/operate these default value directly.
This PR adds null check for top-level primitive values
## How was this patch tested?
new test
Author: Wenchen Fan <wenchen@databricks.com>
Closes#19707 from cloud-fan/bug.
Continuation of PR#19528 (https://github.com/apache/spark/pull/19529#issuecomment-340252119)
The problem with the maven build in the previous PR was the new tests.... the creation of a spark session outside the tests meant there was more than one spark session around at a time.
I was using the spark session outside the tests so that the tests could share data; I've changed it so that each test creates the data anew.
Author: Nathan Kronenfeld <nicole.oresme@gmail.com>
Author: Nathan Kronenfeld <nkronenfeld@uncharted.software>
Closes#19705 from nkronenfeld/alternative-style-tests-2.
## What changes were proposed in this pull request?
Current ML's Bucketizer can only bin a column of continuous features. If a dataset has thousands of of continuous columns needed to bin, we will result in thousands of ML stages. It is inefficient regarding query planning and execution.
We should have a type of bucketizer that can bin a lot of columns all at once. It would need to accept an list of arrays of split points to correspond to the columns to bin, but it might make things more efficient by replacing thousands of stages with just one.
This current approach in this patch is to add a new `MultipleBucketizerInterface` for this purpose. `Bucketizer` now extends this new interface.
### Performance
Benchmarking using the test dataset provided in JIRA SPARK-20392 (blockbuster.csv).
The ML pipeline includes 2 `StringIndexer`s and 1 `MultipleBucketizer` or 137 `Bucketizer`s to bin 137 input columns with the same splits. Then count the time to transform the dataset.
MultipleBucketizer: 3352 ms
Bucketizer: 51512 ms
## How was this patch tested?
Jenkins tests.
Please review http://spark.apache.org/contributing.html before opening a pull request.
Author: Liang-Chi Hsieh <viirya@gmail.com>
Closes#17819 from viirya/SPARK-20542.
## What changes were proposed in this pull request?
For a class with field name of special characters, e.g.:
```scala
case class MyType(`field.1`: String, `field 2`: String)
```
Although we can manipulate DataFrame/Dataset, the field names are encoded:
```scala
scala> val df = Seq(MyType("a", "b"), MyType("c", "d")).toDF
df: org.apache.spark.sql.DataFrame = [field$u002E1: string, field$u00202: string]
scala> df.as[MyType].collect
res7: Array[MyType] = Array(MyType(a,b), MyType(c,d))
```
It causes resolving problem when we try to convert the data with non-encoded field names:
```scala
spark.read.json(path).as[MyType]
...
[info] org.apache.spark.sql.AnalysisException: cannot resolve '`field$u002E1`' given input columns: [field 2, fie
ld.1];
[info] at org.apache.spark.sql.catalyst.analysis.package$AnalysisErrorAt.failAnalysis(package.scala:42)
...
```
We should use decoded field name in Dataset schema.
## How was this patch tested?
Added tests.
Author: Liang-Chi Hsieh <viirya@gmail.com>
Closes#19664 from viirya/SPARK-22442.
## What changes were proposed in this pull request?
This PR proposes to add `errorifexists` to SparkR API and fix the rest of them describing the mode, mainly, in API documentations as well.
This PR also replaces `convertToJSaveMode` to `setWriteMode` so that string as is is passed to JVM and executes:
b034f2565f/sql/core/src/main/scala/org/apache/spark/sql/DataFrameWriter.scala (L72-L82)
and remove the duplication here:
3f958a9992/sql/core/src/main/scala/org/apache/spark/sql/api/r/SQLUtils.scala (L187-L194)
## How was this patch tested?
Manually checked the built documentation. These were mainly found by `` grep -r `error` `` and `grep -r 'error'`.
Also, unit tests added in `test_sparkSQL.R`.
Author: hyukjinkwon <gurwls223@gmail.com>
Closes#19673 from HyukjinKwon/SPARK-21640-followup.
## What changes were proposed in this pull request?
This PR adds support for a new function called `dayofweek` that returns the day of the week of the given argument as an integer value in the range 1-7, where 1 represents Sunday.
## How was this patch tested?
Unit tests and manual tests.
Author: ptkool <michael.styles@shopify.com>
Closes#19672 from ptkool/day_of_week_function.
## What changes were proposed in this pull request?
`spark.sql.statistics.autoUpdate.size` should be `spark.sql.statistics.size.autoUpdate.enabled`. The previous name is confusing as users may treat it as a size config.
This config is in master branch only, no backward compatibility issue.
## How was this patch tested?
N/A
Author: Wenchen Fan <wenchen@databricks.com>
Closes#19667 from cloud-fan/minor.
## What changes were proposed in this pull request?
clarify exception behaviors for all data source v2 interfaces.
## How was this patch tested?
document change only
Author: Wenchen Fan <wenchen@databricks.com>
Closes#19623 from cloud-fan/data-source-exception.
## What changes were proposed in this pull request?
`CodegenContext.copyResult` is kind of a global status for whole stage codegen. But the tricky part is, it is only used to transfer an information from child to parent when calling the `consume` chain. We have to be super careful in `produce`/`consume`, to set it to true when producing multiple result rows, and set it to false in operators that start new pipeline(like sort).
This PR moves the `copyResult` to `CodegenSupport`, and call it at `WholeStageCodegenExec`. This is much easier to reason about.
## How was this patch tested?
existing tests
Author: Wenchen Fan <wenchen@databricks.com>
Closes#19656 from cloud-fan/whole-sage.
…
## What changes were proposed in this pull request?
override JDBCDialects methods quoteIdentifier, getTableExistsQuery and getSchemaQuery in AggregatedDialect
## How was this patch tested?
Test the new implementation in JDBCSuite test("Aggregated dialects")
Author: Huaxin Gao <huaxing@us.ibm.com>
Closes#19658 from huaxingao/spark-22443.
## What changes were proposed in this pull request?
Next fit decreasing bin packing algorithm is used to combine splits in DataSourceScanExec but the comment incorrectly states that first fit decreasing algorithm is used. The current implementation doesn't go back to a previously used bin other than the bin that the last element was put into.
Author: Vinitha Gankidi <vgankidi@netflix.com>
Closes#19634 from vgankidi/SPARK-22412.
## What changes were proposed in this pull request?
When we insert `BatchEvalPython` for Python UDFs into a query plan, if its child has some outputs that are not used by the original parent node, `BatchEvalPython` will still take those outputs and save into the queue. When the data for those outputs are big, it is easily to generate big spill on disk.
For example, the following reproducible code is from the JIRA ticket.
```python
from pyspark.sql.functions import *
from pyspark.sql.types import *
lines_of_file = [ "this is a line" for x in xrange(10000) ]
file_obj = [ "this_is_a_foldername/this_is_a_filename", lines_of_file ]
data = [ file_obj for x in xrange(5) ]
small_df = spark.sparkContext.parallelize(data).map(lambda x : (x[0], x[1])).toDF(["file", "lines"])
exploded = small_df.select("file", explode("lines"))
def split_key(s):
return s.split("/")[1]
split_key_udf = udf(split_key, StringType())
with_filename = exploded.withColumn("filename", split_key_udf("file"))
with_filename.explain(True)
```
The physical plan before/after this change:
Before:
```
*Project [file#0, col#5, pythonUDF0#14 AS filename#9]
+- BatchEvalPython [split_key(file#0)], [file#0, lines#1, col#5, pythonUDF0#14]
+- Generate explode(lines#1), true, false, [col#5]
+- Scan ExistingRDD[file#0,lines#1]
```
After:
```
*Project [file#0, col#5, pythonUDF0#14 AS filename#9]
+- BatchEvalPython [split_key(file#0)], [col#5, file#0, pythonUDF0#14]
+- *Project [col#5, file#0]
+- Generate explode(lines#1), true, false, [col#5]
+- Scan ExistingRDD[file#0,lines#1]
```
Before this change, `lines#1` is a redundant input to `BatchEvalPython`. This patch removes it by adding a Project.
## How was this patch tested?
Manually test.
Author: Liang-Chi Hsieh <viirya@gmail.com>
Closes#19642 from viirya/SPARK-22410.
## What changes were proposed in this pull request?
Added a test class to check NULL handling behavior.
The expected behavior is defined as the one of the most well-known databases as specified here: https://sqlite.org/nulls.html.
SparkSQL behaves like other DBs:
- Adding anything to null gives null -> YES
- Multiplying null by zero gives null -> YES
- nulls are distinct in SELECT DISTINCT -> NO
- nulls are distinct in a UNION -> NO
- "CASE WHEN null THEN 1 ELSE 0 END" is 0? -> YES
- "null OR true" is true -> YES
- "not (null AND false)" is true -> YES
- null in aggregation are skipped -> YES
## How was this patch tested?
Added test class
Author: Marco Gaido <mgaido@hortonworks.com>
Closes#19653 from mgaido91/SPARK-22418.
forward-port https://github.com/apache/spark/pull/19622 to master branch.
This bug doesn't exist in master because we've added hive bucketing support and the hive bucketing metadata can be recognized by Spark, but we should still port it to master: 1) there may be other unsupported hive metadata removed by Spark. 2) reduce code difference between master and 2.2 to ease the backport in the feature.
***
When we alter table schema, we set the new schema to spark `CatalogTable`, convert it to hive table, and finally call `hive.alterTable`. This causes a problem in Spark 2.2, because hive bucketing metedata is not recognized by Spark, which means a Spark `CatalogTable` representing a hive table is always non-bucketed, and when we convert it to hive table and call `hive.alterTable`, the original hive bucketing metadata will be removed.
To fix this bug, we should read out the raw hive table metadata, update its schema, and call `hive.alterTable`. By doing this we can guarantee only the schema is changed, and nothing else.
Author: Wenchen Fan <wenchen@databricks.com>
Closes#19644 from cloud-fan/infer.
## What changes were proposed in this pull request?
According to the [discussion](https://github.com/apache/spark/pull/19571#issuecomment-339472976) on SPARK-15474, we will add new OrcFileFormat in `sql/core` module and allow users to use both old and new OrcFileFormat.
To do that, `OrcOptions` should be visible in `sql/core` module, too. Previously, it was `private[orc]` in `sql/hive`. This PR removes `private[orc]` because we don't use `private[sql]` in `sql/execution` package after [SPARK-16964](https://github.com/apache/spark/pull/14554).
## How was this patch tested?
Pass the Jenkins with the existing tests.
Author: Dongjoon Hyun <dongjoon@apache.org>
Closes#19636 from dongjoon-hyun/SPARK-22416.
## What changes were proposed in this pull request?
Adding a global limit on top of the distinct values before sorting and collecting will reduce the overall work in the case where we have more distinct values. We will also eagerly perform a collect rather than a take because we know we only have at most (maxValues + 1) rows.
## How was this patch tested?
Existing tests cover sorted order
Author: Patrick Woody <pwoody@palantir.com>
Closes#19629 from pwoody/SPARK-22408.
## What changes were proposed in this pull request?
This patch includes some doc updates for data source API v2. I was reading the code and noticed some minor issues.
## How was this patch tested?
This is a doc only change.
Author: Reynold Xin <rxin@databricks.com>
Closes#19626 from rxin/dsv2-update.
## What changes were proposed in this pull request?
Write HDFSBackedStateStoreProvider.loadMap non-recursively. This prevents stack overflow if too many deltas stack up in a low memory environment.
## How was this patch tested?
existing unit tests for functional equivalence, new unit test to check for stack overflow
Author: Jose Torres <jose@databricks.com>
Closes#19611 from joseph-torres/SPARK-22305.
## What changes were proposed in this pull request?
Both `ReadSupport` and `ReadTask` have a method called `createReader`, but they create different things. This could cause some confusion for data source developers. The same issue exists between `WriteSupport` and `DataWriterFactory`, both of which have a method called `createWriter`. This PR renames the method of `ReadTask`/`DataWriterFactory` to `createDataReader`/`createDataWriter`.
Besides, the name of `RowToInternalRowDataWriterFactory` is not correct, because it actually converts `InternalRow`s to `Row`s. It should be renamed `InternalRowDataWriterFactory`.
## How was this patch tested?
Only renaming, should be covered by existing tests.
Author: Zhenhua Wang <wzh_zju@163.com>
Closes#19610 from wzhfy/rename.
## What changes were proposed in this pull request?
When Hive support is not on, users can hit unresolved plan node when trying to call `INSERT OVERWRITE DIRECTORY` using Hive format.
```
"unresolved operator 'InsertIntoDir true, Storage(Location: /private/var/folders/vx/j0ydl5rn0gd9mgrh1pljnw900000gn/T/spark-b4227606-9311-46a8-8c02-56355bf0e2bc, Serde Library: org.apache.hadoop.hive.ql.io.orc.OrcSerde, InputFormat: org.apache.hadoop.hive.ql.io.orc.OrcInputFormat, OutputFormat: org.apache.hadoop.hive.ql.io.orc.OrcOutputFormat), hive, true;;
```
This PR is to issue a better error message.
## How was this patch tested?
Added a test case.
Author: gatorsmile <gatorsmile@gmail.com>
Closes#19608 from gatorsmile/hivesupportInsertOverwrite.
## What changes were proposed in this pull request?
In `UnsafeInMemorySorter`, one record may take 32 bytes: 1 `long` for pointer, 1 `long` for key-prefix, and another 2 `long`s as the temporary buffer for radix sort.
In `UnsafeExternalSorter`, we set the `DEFAULT_NUM_ELEMENTS_FOR_SPILL_THRESHOLD` to be `1024 * 1024 * 1024 / 2`, and hoping the max size of point array to be 8 GB. However this is wrong, `1024 * 1024 * 1024 / 2 * 32` is actually 16 GB, and if we grow the point array before reach this limitation, we may hit the max-page-size error.
Users may see exception like this on large dataset:
```
Caused by: java.lang.IllegalArgumentException: Cannot allocate a page with more than 17179869176 bytes
at org.apache.spark.memory.TaskMemoryManager.allocatePage(TaskMemoryManager.java:241)
at org.apache.spark.memory.MemoryConsumer.allocatePage(MemoryConsumer.java:121)
at org.apache.spark.util.collection.unsafe.sort.UnsafeExternalSorter.acquireNewPageIfNecessary(UnsafeExternalSorter.java:374)
at org.apache.spark.util.collection.unsafe.sort.UnsafeExternalSorter.insertRecord(UnsafeExternalSorter.java:396)
at org.apache.spark.sql.execution.UnsafeExternalRowSorter.insertRow(UnsafeExternalRowSorter.java:94)
...
```
Setting `DEFAULT_NUM_ELEMENTS_FOR_SPILL_THRESHOLD` to a smaller number is not enough, users can still set the config to a big number and trigger the too large page size issue. This PR fixes it by explicitly handling the too large page size exception in the sorter and spill.
This PR also change the type of `spark.shuffle.spill.numElementsForceSpillThreshold` to int, because it's only compared with `numRecords`, which is an int. This is an internal conf so we don't have a serious compatibility issue.
## How was this patch tested?
TODO
Author: Wenchen Fan <wenchen@databricks.com>
Closes#18251 from cloud-fan/sort.
## What changes were proposed in this pull request?
This PR fixes the conversion error when reads data from a PostgreSQL table that contains columns of `uuid[]`, `inet[]` and `cidr[]` data types.
For example, create a table with the uuid[] data type, and insert the test data.
```SQL
CREATE TABLE users
(
id smallint NOT NULL,
name character varying(50),
user_ids uuid[],
PRIMARY KEY (id)
)
INSERT INTO users ("id", "name","user_ids")
VALUES (1, 'foo', ARRAY
['7be8aaf8-650e-4dbb-8186-0a749840ecf2'
,'205f9bfc-018c-4452-a605-609c0cfad228']::UUID[]
)
```
Then it will throw the following exceptions when trying to load the data.
```
java.lang.ClassCastException: [Ljava.util.UUID; cannot be cast to [Ljava.lang.String;
at org.apache.spark.sql.execution.datasources.jdbc.JdbcUtils$$anonfun$14.apply(JdbcUtils.scala:459)
at org.apache.spark.sql.execution.datasources.jdbc.JdbcUtils$$anonfun$14.apply(JdbcUtils.scala:458)
...
```
## How was this patch tested?
Added test in `PostgresIntegrationSuite`.
Author: Jen-Ming Chung <jenmingisme@gmail.com>
Closes#19567 from jmchung/SPARK-22291.
## What changes were proposed in this pull request?
This is a followup of https://github.com/apache/spark/pull/17075 , to fix the bug in codegen path.
## How was this patch tested?
new regression test
Author: Wenchen Fan <wenchen@databricks.com>
Closes#19576 from cloud-fan/bug.
## What changes were proposed in this pull request?
AggUtils.planStreamingAggregation has some comments about DISTINCT aggregates,
while streaming aggregation does not support DISTINCT.
This seems to have been wrongly copy-pasted over.
## How was this patch tested?
Only a comment change.
Author: Juliusz Sompolski <julek@databricks.com>
Closes#18937 from juliuszsompolski/streaming-agg-doc.
## What changes were proposed in this pull request?
`ArrowEvalPythonExec` and `FlatMapGroupsInPandasExec` are refering config values of `SQLConf` in function for `mapPartitions`/`mapPartitionsInternal`, but we should capture them in Driver.
## How was this patch tested?
Added a test and existing tests.
Author: Takuya UESHIN <ueshin@databricks.com>
Closes#19587 from ueshin/issues/SPARK-22370.
## What changes were proposed in this pull request?
Seems that end users can be confused by the union's behavior on Dataset of typed objects. We can clarity it more in the document of `union` function.
## How was this patch tested?
Only document change.
Author: Liang-Chi Hsieh <viirya@gmail.com>
Closes#19570 from viirya/SPARK-22335.
## What changes were proposed in this pull request?
Canonicalized plans are not supposed to be executed. I ran into a case in which there's some code that accidentally calls execute on a canonicalized plan. This patch throws a more explicit exception when that happens.
## How was this patch tested?
Added a test case in SparkPlanSuite.
Author: Reynold Xin <rxin@databricks.com>
Closes#18828 from rxin/SPARK-21619.
## What changes were proposed in this pull request?
https://issues.apache.org/jira/browse/SPARK-22333
In current version, users can use CURRENT_DATE() and CURRENT_TIMESTAMP() without specifying braces.
However, when a table has columns named as "current_date" or "current_timestamp", it will still be parsed as function call.
There are many such cases in our production cluster. We get the wrong answer due to this inappropriate behevior. In general, ColumnReference should get higher priority than timeFunctionCall.
## How was this patch tested?
unit test
manul test
Author: donnyzone <wellfengzhu@gmail.com>
Closes#19559 from DonnyZone/master.
## What changes were proposed in this pull request?
Adds a new optimisation rule 'ReplaceExceptWithNotFilter' that replaces Except logical with Filter operator and schedule it before applying 'ReplaceExceptWithAntiJoin' rule. This way we can avoid expensive join operation if one or both of the datasets of the Except operation are fully derived out of Filters from a same parent.
## How was this patch tested?
The patch is tested locally using spark-shell + unit test.
Author: Sathiya <sathiya.kumar@polytechnique.edu>
Closes#19451 from sathiyapk/SPARK-22181-optimize-exceptWithFilter.
## What changes were proposed in this pull request?
SPARK-18016 introduced `NestedClass` to avoid that the many methods generated by `splitExpressions` contribute to the outer class' constant pool, making it growing too much. Unfortunately, despite their definition is stored in the `NestedClass`, they all are invoked in the outer class and for each method invocation, there are two entries added to the constant pool: a `Methodref` and a `Utf8` entry (you can easily check this compiling a simple sample class with `janinoc` and looking at its Constant Pool). This limits the scalability of the solution with very large methods which are split in a lot of small ones. This means that currently we are generating classes like this one:
```
class SpecificUnsafeProjection extends org.apache.spark.sql.catalyst.expressions.UnsafeProjection {
...
public UnsafeRow apply(InternalRow i) {
rowWriter.zeroOutNullBytes();
apply_0(i);
apply_1(i);
...
nestedClassInstance.apply_862(i);
nestedClassInstance.apply_863(i);
...
nestedClassInstance1.apply_1612(i);
nestedClassInstance1.apply_1613(i);
...
}
...
private class NestedClass {
private void apply_862(InternalRow i) { ... }
private void apply_863(InternalRow i) { ... }
...
}
private class NestedClass1 {
private void apply_1612(InternalRow i) { ... }
private void apply_1613(InternalRow i) { ... }
...
}
}
```
This PR reduce the Constant Pool size of the outer class by adding a new method to each nested class: in this method we invoke all the small methods generated by `splitExpression` in that nested class. In this way, in the outer class there is only one method invocation per nested class, reducing by orders of magnitude the entries in its constant pool because of method invocations. This means that after the patch the generated code becomes:
```
class SpecificUnsafeProjection extends org.apache.spark.sql.catalyst.expressions.UnsafeProjection {
...
public UnsafeRow apply(InternalRow i) {
rowWriter.zeroOutNullBytes();
apply_0(i);
apply_1(i);
...
nestedClassInstance.apply(i);
nestedClassInstance1.apply(i);
...
}
...
private class NestedClass {
private void apply_862(InternalRow i) { ... }
private void apply_863(InternalRow i) { ... }
...
private void apply(InternalRow i) {
apply_862(i);
apply_863(i);
...
}
}
private class NestedClass1 {
private void apply_1612(InternalRow i) { ... }
private void apply_1613(InternalRow i) { ... }
...
private void apply(InternalRow i) {
apply_1612(i);
apply_1613(i);
...
}
}
}
```
## How was this patch tested?
Added UT and existing UTs
Author: Marco Gaido <mgaido@hortonworks.com>
Author: Marco Gaido <marcogaido91@gmail.com>
Closes#19480 from mgaido91/SPARK-22226.
## What changes were proposed in this pull request?
This PR is to clean the related codes majorly based on the today's code review on https://github.com/apache/spark/pull/19559
## How was this patch tested?
N/A
Author: gatorsmile <gatorsmile@gmail.com>
Closes#19585 from gatorsmile/trivialFixes.
## What changes were proposed in this pull request?
Adding date and timestamp support with Arrow for `toPandas()` and `pandas_udf`s. Timestamps are stored in Arrow as UTC and manifested to the user as timezone-naive localized to the Python system timezone.
## How was this patch tested?
Added Scala tests for date and timestamp types under ArrowConverters, ArrowUtils, and ArrowWriter suites. Added Python tests for `toPandas()` and `pandas_udf`s with date and timestamp types.
Author: Bryan Cutler <cutlerb@gmail.com>
Author: Takuya UESHIN <ueshin@databricks.com>
Closes#18664 from BryanCutler/arrow-date-timestamp-SPARK-21375.
## What changes were proposed in this pull request?
It's possible that users create a `Dataset`, and call `collect` of this `Dataset` in many threads at the same time. Currently `Dataset#collect` just call `encoder.fromRow` to convert spark rows to objects of type T, and this encoder is per-dataset. This means `Dataset#collect` is not thread-safe, because the encoder uses a projection to output the object to a re-usable row.
This PR fixes this problem, by creating a new projection when calling `Dataset#collect`, so that we have the re-usable row for each method call, instead of each Dataset.
## How was this patch tested?
N/A
Author: Wenchen Fan <wenchen@databricks.com>
Closes#19577 from cloud-fan/encoder.
## What changes were proposed in this pull request?
This is a regression introduced by #14207. After Spark 2.1, we store the inferred schema when creating the table, to avoid inferring schema again at read path. However, there is one special case: overlapped columns between data and partition. For this case, it breaks the assumption of table schema that there is on ovelap between data and partition schema, and partition columns should be at the end. The result is, for Spark 2.1, the table scan has incorrect schema that puts partition columns at the end. For Spark 2.2, we add a check in CatalogTable to validate table schema, which fails at this case.
To fix this issue, a simple and safe approach is to fallback to old behavior when overlapeed columns detected, i.e. store empty schema in metastore.
## How was this patch tested?
new regression test
Author: Wenchen Fan <wenchen@databricks.com>
Closes#19579 from cloud-fan/bug2.
## What changes were proposed in this pull request?
Add a flag "spark.sql.files.ignoreMissingFiles" to parallel the existing flag "spark.sql.files.ignoreCorruptFiles".
## How was this patch tested?
new unit test
Author: Jose Torres <jose@databricks.com>
Closes#19581 from joseph-torres/SPARK-22366.
## What changes were proposed in this pull request?
Support unit tests of external code (i.e., applications that use spark) using scalatest that don't want to use FunSuite. SharedSparkContext already supports this, but SharedSQLContext does not.
I've introduced SharedSparkSession as a parent to SharedSQLContext, written in a way that it does support all scalatest styles.
## How was this patch tested?
There are three new unit test suites added that just test using FunSpec, FlatSpec, and WordSpec.
Author: Nathan Kronenfeld <nicole.oresme@gmail.com>
Closes#19529 from nkronenfeld/alternative-style-tests-2.
## What changes were proposed in this pull request?
Removed one unused method.
## How was this patch tested?
Existing tests.
Author: Liang-Chi Hsieh <viirya@gmail.com>
Closes#19508 from viirya/SPARK-20783-followup.
## What changes were proposed in this pull request?
Scala 2.12's `Future` defines two new methods to implement, `transform` and `transformWith`. These can be implemented naturally in Spark's `FutureAction` extension and subclasses, but, only in terms of the new methods that don't exist in Scala 2.11. To support both at the same time, reflection is used to implement these.
## How was this patch tested?
Existing tests.
Author: Sean Owen <sowen@cloudera.com>
Closes#19561 from srowen/SPARK-22322.
## What changes were proposed in this pull request?
Rewritten error message for clarity. Added extra information in case of attribute name collision, hinting the user to double-check referencing two different tables
## How was this patch tested?
No functional changes, only final message has changed. It has been tested manually against the situation proposed in the JIRA ticket. Automated tests in repository pass.
This PR is original work from me and I license this work to the Spark project
Author: Ruben Berenguel Montoro <ruben@mostlymaths.net>
Author: Ruben Berenguel Montoro <ruben@dreamattic.com>
Author: Ruben Berenguel <ruben@mostlymaths.net>
Closes#17100 from rberenguel/SPARK-13947-error-message.
## What changes were proposed in this pull request?
We enable table cache `InMemoryTableScanExec` to provide `ColumnarBatch` now. But the cached batches are retrieved without pruning. In this case, we still need to do partition batch pruning.
## How was this patch tested?
Existing tests.
Author: Liang-Chi Hsieh <viirya@gmail.com>
Closes#19569 from viirya/SPARK-22348.
## What changes were proposed in this pull request?
The current implementation of `ApproxCountDistinctForIntervals` is `ImperativeAggregate`. The number of `aggBufferAttributes` is the number of total words in the hllppHelper array. Each hllppHelper has 52 words by default relativeSD.
Since this aggregate function is used in equi-height histogram generation, and the number of buckets in histogram is usually hundreds, the number of `aggBufferAttributes` can easily reach tens of thousands or even more.
This leads to a huge method in codegen and causes error:
```
org.codehaus.janino.JaninoRuntimeException: Code of method "apply(Lorg/apache/spark/sql/catalyst/InternalRow;)Lorg/apache/spark/sql/catalyst/expressions/UnsafeRow;" of class "org.apache.spark.sql.catalyst.expressions.GeneratedClass$SpecificUnsafeProjection" grows beyond 64 KB.
```
Besides, huge generated methods also result in performance regression.
In this PR, we change its implementation to `TypedImperativeAggregate`. After the fix, `ApproxCountDistinctForIntervals` can deal with more than thousands endpoints without throwing codegen error, and improve performance from `20 sec` to `2 sec` in a test case of 500 endpoints.
## How was this patch tested?
Test by an added test case and existing tests.
Author: Zhenhua Wang <wangzhenhua@huawei.com>
Closes#19506 from wzhfy/change_forIntervals_typedAgg.
TIMESTAMP (-101), BINARY_DOUBLE (101) and BINARY_FLOAT (100) are handled in OracleDialect
## What changes were proposed in this pull request?
When a oracle table contains columns whose type is BINARY_FLOAT or BINARY_DOUBLE, spark sql fails to load a table with SQLException
```
java.sql.SQLException: Unsupported type 101
at org.apache.spark.sql.execution.datasources.jdbc.JdbcUtils$.org$apache$spark$sql$execution$datasources$jdbc$JdbcUtils$$getCatalystType(JdbcUtils.scala:235)
at org.apache.spark.sql.execution.datasources.jdbc.JdbcUtils$$anonfun$8.apply(JdbcUtils.scala:292)
at org.apache.spark.sql.execution.datasources.jdbc.JdbcUtils$$anonfun$8.apply(JdbcUtils.scala:292)
at scala.Option.getOrElse(Option.scala:121)
at org.apache.spark.sql.execution.datasources.jdbc.JdbcUtils$.getSchema(JdbcUtils.scala:291)
at org.apache.spark.sql.execution.datasources.jdbc.JDBCRDD$.resolveTable(JDBCRDD.scala:64)
at org.apache.spark.sql.execution.datasources.jdbc.JDBCRelation.<init>(JDBCRelation.scala:113)
at org.apache.spark.sql.execution.datasources.jdbc.JdbcRelationProvider.createRelation(JdbcRelationProvider.scala:47)
at org.apache.spark.sql.execution.datasources.DataSource.resolveRelation(DataSource.scala:306)
at org.apache.spark.sql.DataFrameReader.load(DataFrameReader.scala:178)
at org.apache.spark.sql.DataFrameReader.load(DataFrameReader.scala:146)
```
## How was this patch tested?
I updated a UT which covers type conversion test for types (-101, 100, 101), on top of that I tested this change against actual table with those columns and it was able to read and write to the table.
Author: Kohki Nishio <taroplus@me.com>
Closes#19548 from taroplus/oracle_sql_types_101.
## What changes were proposed in this pull request?
When [SPARK-19261](https://issues.apache.org/jira/browse/SPARK-19261) implements `ALTER TABLE ADD COLUMNS`, ORC data source is omitted due to SPARK-14387, SPARK-16628, and SPARK-18355. Now, those issues are fixed and Spark 2.3 is [using Spark schema to read ORC table instead of ORC file schema](e6e36004af). This PR enables `ALTER TABLE ADD COLUMNS` for ORC data source.
## How was this patch tested?
Pass the updated and added test cases.
Author: Dongjoon Hyun <dongjoon@apache.org>
Closes#19545 from dongjoon-hyun/SPARK-21929.
## What changes were proposed in this pull request?
Plan equality should be computed by `canonicalized`, so we can remove unnecessary `hashCode` and `equals` methods.
## How was this patch tested?
Existing tests.
Author: Zhenhua Wang <wangzhenhua@huawei.com>
Closes#19539 from wzhfy/remove_equals.
## What changes were proposed in this pull request?
This is a follow-up of #18732.
This pr modifies `GroupedData.apply()` method to convert pandas udf to grouped udf implicitly.
## How was this patch tested?
Exisiting tests.
Author: Takuya UESHIN <ueshin@databricks.com>
Closes#19517 from ueshin/issues/SPARK-20396/fup2.
## What changes were proposed in this pull request?
spark does not support grouping__id, it has grouping_id() instead.
But it is not convenient for hive user to change to spark-sql
so this pr is to replace grouping__id with grouping_id()
hive user need not to alter their scripts
## How was this patch tested?
test with SQLQuerySuite.scala
Author: CenYuhai <yuhai.cen@ele.me>
Closes#18270 from cenyuhai/SPARK-21055.
## What changes were proposed in this pull request?
This is a very trivial PR, simply marking `strategies` in `SparkPlanner` with the `override` keyword for clarity since it is overriding `strategies` in `QueryPlanner` two levels up in the class hierarchy. I was reading through the code to learn a bit and got stuck on this fact for a little while, so I figured this may be helpful so that another developer new to the project doesn't get stuck where I was.
I did not make a JIRA ticket for this because it is so trivial, but I'm happy to do so to adhere to the contribution guidelines if required.
## How was this patch tested?
(Please explain how this patch was tested. E.g. unit tests, integration tests, manual tests)
(If this patch involves UI changes, please attach a screenshot; otherwise, remove this)
Please review http://spark.apache.org/contributing.html before opening a pull request.
Author: Eric Perry <eric@ericjperry.com>
Closes#19537 from ericjperry/override-strategies.
## What changes were proposed in this pull request?
A working prototype for data source v2 write path.
The writing framework is similar to the reading framework. i.e. `WriteSupport` -> `DataSourceV2Writer` -> `DataWriterFactory` -> `DataWriter`.
Similar to the `FileCommitPotocol`, the writing API has job and task level commit/abort to support the transaction.
## How was this patch tested?
new tests
Author: Wenchen Fan <wenchen@databricks.com>
Closes#19269 from cloud-fan/data-source-v2-write.
## What changes were proposed in this pull request?
Fix java style issues
## How was this patch tested?
Run `./dev/lint-java` locally since it's not run on Jenkins
Author: Andrew Ash <andrew@andrewash.com>
Closes#19486 from ash211/aash/fix-lint-java.
## What changes were proposed in this pull request?
This PR addresses the comments by gatorsmile on [the previous PR](https://github.com/apache/spark/pull/19494).
## How was this patch tested?
Previous UT and added UT.
Author: Marco Gaido <marcogaido91@gmail.com>
Closes#19522 from mgaido91/SPARK-22249_FOLLOWUP.
## What changes were proposed in this pull request?
To let the same aggregate function that appear multiple times in an Aggregate be evaluated only once, we need to deduplicate the aggregate expressions. The original code was trying to use a "distinct" call to get a set of aggregate expressions, but did not work, since the "distinct" did not compare semantic equality. And even if it did, further work should be done in result expression rewriting.
In this PR, I changed the "set" to a map mapping the semantic identity of a aggregate expression to itself. Thus, later on, when rewriting result expressions (i.e., output expressions), the aggregate expression reference can be fixed.
## How was this patch tested?
Added a new test in SQLQuerySuite
Author: maryannxue <maryann.xue@gmail.com>
Closes#19488 from maryannxue/spark-22266.
## What changes were proposed in this pull request?
Complex state-updating and/or timeout-handling logic in mapGroupsWithState functions may require taking decisions based on the current event-time watermark and/or processing time. Currently, you can use the SQL function `current_timestamp` to get the current processing time, but it needs to be passed inserted in every row with a select, and then passed through the encoder, which isn't efficient. Furthermore, there is no way to get the current watermark.
This PR exposes both of them through the GroupState API.
Additionally, it also cleans up some of the GroupState docs.
## How was this patch tested?
New unit tests
Author: Tathagata Das <tathagata.das1565@gmail.com>
Closes#19495 from tdas/SPARK-22278.
## What changes were proposed in this pull request?
In Average.scala, it has
```
override lazy val evaluateExpression = child.dataType match {
case DecimalType.Fixed(p, s) =>
// increase the precision and scale to prevent precision loss
val dt = DecimalType.bounded(p + 14, s + 4)
Cast(Cast(sum, dt) / Cast(count, dt), resultType)
case _ =>
Cast(sum, resultType) / Cast(count, resultType)
}
def setChild (newchild: Expression) = {
child = newchild
}
```
It is possible that Cast(count, dt), resultType) will make the precision of the decimal number bigger than 38, and this causes over flow. Since count is an integer and doesn't need a scale, I will cast it using DecimalType.bounded(38,0)
## How was this patch tested?
In DataFrameSuite, I will add a test case.
Please review http://spark.apache.org/contributing.html before opening a pull request.
Author: Huaxin Gao <huaxing@us.ibm.com>
Closes#19496 from huaxingao/spark-22271.
## What changes were proposed in this pull request?
Evaluate one-sided conditions early in stream-stream joins.
This is in addition to normal filter pushdown, because integrating it with the join logic allows it to take place in outer join scenarios. This means that rows which can never satisfy the join condition won't clog up the state.
## How was this patch tested?
new unit tests
Author: Jose Torres <jose@databricks.com>
Closes#19452 from joseph-torres/SPARK-22136.
## What changes were proposed in this pull request?
#### before
```scala
scala> val words = spark.read.textFile("README.md").flatMap(_.split(" "))
words: org.apache.spark.sql.Dataset[String] = [value: string]
scala> val grouped = words.groupByKey(identity)
grouped: org.apache.spark.sql.KeyValueGroupedDataset[String,String] = org.apache.spark.sql.KeyValueGroupedDataset65214862
```
#### after
```scala
scala> val words = spark.read.textFile("README.md").flatMap(_.split(" "))
words: org.apache.spark.sql.Dataset[String] = [value: string]
scala> val grouped = words.groupByKey(identity)
grouped: org.apache.spark.sql.KeyValueGroupedDataset[String,String] = [key: [value: string], value: [value: string]]
```
## How was this patch tested?
existing ut
cc gatorsmile cloud-fan
Author: Kent Yao <yaooqinn@hotmail.com>
Closes#19363 from yaooqinn/minor-dataset-tostring.
## What changes were proposed in this pull request?
As pointed out in the JIRA, there is a bug which causes an exception to be thrown if `isin` is called with an empty list on a cached DataFrame. The PR fixes it.
## How was this patch tested?
Added UT.
Author: Marco Gaido <marcogaido91@gmail.com>
Closes#19494 from mgaido91/SPARK-22249.
## What changes were proposed in this pull request?
This PR aims to
- Rename `OrcRelation` to `OrcFileFormat` object.
- Replace `OrcRelation.ORC_COMPRESSION` with `org.apache.orc.OrcConf.COMPRESS`. Since [SPARK-21422](https://issues.apache.org/jira/browse/SPARK-21422), we can use `OrcConf.COMPRESS` instead of Hive's.
```scala
// The references of Hive's classes will be minimized.
val ORC_COMPRESSION = "orc.compress"
```
## How was this patch tested?
Pass the Jenkins with the existing and updated test cases.
Author: Dongjoon Hyun <dongjoon@apache.org>
Closes#19502 from dongjoon-hyun/SPARK-22282.
## What changes were proposed in this pull request?
`ObjectHashAggregateExec` should override `outputPartitioning` in order to avoid unnecessary shuffle.
## How was this patch tested?
Added Jenkins test.
Author: Liang-Chi Hsieh <viirya@gmail.com>
Closes#19501 from viirya/SPARK-22223.
## What changes were proposed in this pull request?
In EnsureStatefulOpPartitioning, we check that the inputRDD to a SparkPlan has the expected partitioning for Streaming Stateful Operators. The problem is that we are not allowed to access this information during planning.
The reason we added that check was because CoalesceExec could actually create RDDs with 0 partitions. We should fix it such that when CoalesceExec says that there is a SinglePartition, there is in fact an inputRDD of 1 partition instead of 0 partitions.
## How was this patch tested?
Regression test in StreamingQuerySuite
Author: Burak Yavuz <brkyvz@gmail.com>
Closes#19467 from brkyvz/stateful-op.
## What changes were proposed in this pull request?
When fixing schema field names using escape characters with `addReferenceMinorObj()` at [SPARK-18952](https://issues.apache.org/jira/browse/SPARK-18952) (#16361), double-quotes around the names were remained and the names become something like `"((java.lang.String) references[1])"`.
```java
/* 055 */ private int maxSteps = 2;
/* 056 */ private int numRows = 0;
/* 057 */ private org.apache.spark.sql.types.StructType keySchema = new org.apache.spark.sql.types.StructType().add("((java.lang.String) references[1])", org.apache.spark.sql.types.DataTypes.StringType);
/* 058 */ private org.apache.spark.sql.types.StructType valueSchema = new org.apache.spark.sql.types.StructType().add("((java.lang.String) references[2])", org.apache.spark.sql.types.DataTypes.LongType);
/* 059 */ private Object emptyVBase;
```
We should remove the double-quotes to refer the values in `references` properly:
```java
/* 055 */ private int maxSteps = 2;
/* 056 */ private int numRows = 0;
/* 057 */ private org.apache.spark.sql.types.StructType keySchema = new org.apache.spark.sql.types.StructType().add(((java.lang.String) references[1]), org.apache.spark.sql.types.DataTypes.StringType);
/* 058 */ private org.apache.spark.sql.types.StructType valueSchema = new org.apache.spark.sql.types.StructType().add(((java.lang.String) references[2]), org.apache.spark.sql.types.DataTypes.LongType);
/* 059 */ private Object emptyVBase;
```
## How was this patch tested?
Existing tests.
Author: Takuya UESHIN <ueshin@databricks.com>
Closes#19491 from ueshin/issues/SPARK-22273.
## What changes were proposed in this pull request?
`BasicWriteTaskStatsTracker.getFileSize()` to catch `FileNotFoundException`, log info and then return 0 as a file size.
This ensures that if a newly created file isn't visible due to the store not always having create consistency, the metric collection doesn't cause the failure.
## How was this patch tested?
New test suite included, `BasicWriteTaskStatsTrackerSuite`. This not only checks the resilience to missing files, but verifies the existing logic as to how file statistics are gathered.
Note that in the current implementation
1. if you call `Tracker..getFinalStats()` more than once, the file size count will increase by size of the last file. This could be fixed by clearing the filename field inside `getFinalStats()` itself.
2. If you pass in an empty or null string to `Tracker.newFile(path)` then IllegalArgumentException is raised, but only in `getFinalStats()`, rather than in `newFile`. There's a test for this behaviour in the new suite, as it verifies that only FNFEs get swallowed.
Author: Steve Loughran <stevel@hortonworks.com>
Closes#18979 from steveloughran/cloud/SPARK-21762-missing-files-in-metrics.
## What changes were proposed in this pull request?
This PR changes `keyWithIndexToNumValues` to `keyWithIndexToValue`.
There will be directories on HDFS named with this `keyWithIndexToNumValues`. So if we ever want to fix this, let's fix it now.
## How was this patch tested?
existing unit test cases.
Author: Liwei Lin <lwlin7@gmail.com>
Closes#19435 from lw-lin/keyWithIndex.
## What changes were proposed in this pull request?
This is an effort to reduce the difference between Hive and Spark. Spark supports case-sensitivity in columns. Especially, for Struct types, with `spark.sql.caseSensitive=true`, the following is supported.
```scala
scala> sql("select named_struct('a', 1, 'A', 2).a").show
+--------------------------+
|named_struct(a, 1, A, 2).a|
+--------------------------+
| 1|
+--------------------------+
scala> sql("select named_struct('a', 1, 'A', 2).A").show
+--------------------------+
|named_struct(a, 1, A, 2).A|
+--------------------------+
| 2|
+--------------------------+
```
And vice versa, with `spark.sql.caseSensitive=false`, the following is supported.
```scala
scala> sql("select named_struct('a', 1).A, named_struct('A', 1).a").show
+--------------------+--------------------+
|named_struct(a, 1).A|named_struct(A, 1).a|
+--------------------+--------------------+
| 1| 1|
+--------------------+--------------------+
```
However, types are considered different. For example, SET operations fail.
```scala
scala> sql("SELECT named_struct('a',1) union all (select named_struct('A',2))").show
org.apache.spark.sql.AnalysisException: Union can only be performed on tables with the compatible column types. struct<A:int> <> struct<a:int> at the first column of the second table;;
'Union
:- Project [named_struct(a, 1) AS named_struct(a, 1)#57]
: +- OneRowRelation$
+- Project [named_struct(A, 2) AS named_struct(A, 2)#58]
+- OneRowRelation$
```
This PR aims to support case-insensitive type equality. For example, in Set operation, the above operation succeed when `spark.sql.caseSensitive=false`.
```scala
scala> sql("SELECT named_struct('a',1) union all (select named_struct('A',2))").show
+------------------+
|named_struct(a, 1)|
+------------------+
| [1]|
| [2]|
+------------------+
```
## How was this patch tested?
Pass the Jenkins with a newly add test case.
Author: Dongjoon Hyun <dongjoon@apache.org>
Closes#18460 from dongjoon-hyun/SPARK-21247.
## What changes were proposed in this pull request?
Due to optimizer removing some unnecessary aliases, the logical and physical plan may have different output attribute ids. FileFormatWriter should handle this when creating the physical sort node.
## How was this patch tested?
new regression test.
Author: Wenchen Fan <wenchen@databricks.com>
Closes#19483 from cloud-fan/bug2.
## What changes were proposed in this pull request?
The method `deterministic` is frequently called in optimizer.
Refactor `deterministic` as lazy value, in order to avoid redundant computations.
## How was this patch tested?
Simple benchmark test over TPC-DS queries, run time from query string to optimized plan(continuous 20 runs, and get the average of last 5 results):
Before changes: 12601 ms
After changes: 11993ms
This is 4.8% performance improvement.
Also run test with Unit test.
Author: Wang Gengliang <ltnwgl@gmail.com>
Closes#19478 from gengliangwang/deterministicAsLazyVal.
## What changes were proposed in this pull request?
`ParquetFileFormat` to relax its requirement of output committer class from `org.apache.parquet.hadoop.ParquetOutputCommitter` or subclass thereof (and so implicitly Hadoop `FileOutputCommitter`) to any committer implementing `org.apache.hadoop.mapreduce.OutputCommitter`
This enables output committers which don't write to the filesystem the way `FileOutputCommitter` does to save parquet data from a dataframe: at present you cannot do this.
Before a committer which isn't a subclass of `ParquetOutputCommitter`, it checks to see if the context has requested summary metadata by setting `parquet.enable.summary-metadata`. If true, and the committer class isn't a parquet committer, it raises a RuntimeException with an error message.
(It could downgrade, of course, but raising an exception makes it clear there won't be an summary. It also makes the behaviour testable.)
Note that `SQLConf` already states that any `OutputCommitter` can be used, but that typically it's a subclass of ParquetOutputCommitter. That's not currently true. This patch will make the code consistent with the docs, adding tests to verify,
## How was this patch tested?
The patch includes a test suite, `ParquetCommitterSuite`, with a new committer, `MarkingFileOutputCommitter` which extends `FileOutputCommitter` and writes a marker file in the destination directory. The presence of the marker file can be used to verify the new committer was used. The tests then try the combinations of Parquet committer summary/no-summary and marking committer summary/no-summary.
| committer | summary | outcome |
|-----------|---------|---------|
| parquet | true | success |
| parquet | false | success |
| marking | false | success with marker |
| marking | true | exception |
All tests are happy.
Author: Steve Loughran <stevel@hortonworks.com>
Closes#19448 from steveloughran/cloud/SPARK-22217-committer.
## What changes were proposed in this pull request?
Adding the code for setting 'aggregate time' metric to non-codegen path in HashAggregateExec and to ObjectHashAggregateExces.
## How was this patch tested?
Tested manually.
Author: Ala Luszczak <ala@databricks.com>
Closes#19473 from ala/fix-agg-time.
## What changes were proposed in this pull request?
As we discussed in https://github.com/apache/spark/pull/19136#discussion_r137023744 , we should push down operators to data source before planning, so that data source can report statistics more accurate.
This PR also includes some cleanup for the read path.
## How was this patch tested?
existing tests.
Author: Wenchen Fan <wenchen@databricks.com>
Closes#19424 from cloud-fan/follow.
## What changes were proposed in this pull request?
In https://github.com/apache/spark/pull/18064, we allowed `RunnableCommand` to have children in order to fix some UI issues. Then we made `InsertIntoXXX` commands take the input `query` as a child, when we do the actual writing, we just pass the physical plan to the writer(`FileFormatWriter.write`).
However this is problematic. In Spark SQL, optimizer and planner are allowed to change the schema names a little bit. e.g. `ColumnPruning` rule will remove no-op `Project`s, like `Project("A", Scan("a"))`, and thus change the output schema from "<A: int>" to `<a: int>`. When it comes to writing, especially for self-description data format like parquet, we may write the wrong schema to the file and cause null values at the read path.
Fortunately, in https://github.com/apache/spark/pull/18450 , we decided to allow nested execution and one query can map to multiple executions in the UI. This releases the major restriction in #18604 , and now we don't have to take the input `query` as child of `InsertIntoXXX` commands.
So the fix is simple, this PR partially revert #18064 and make `InsertIntoXXX` commands leaf nodes again.
## How was this patch tested?
new regression test
Author: Wenchen Fan <wenchen@databricks.com>
Closes#19474 from cloud-fan/bug.
## What changes were proposed in this pull request?
Implement StreamingRelation.computeStats to fix explain
## How was this patch tested?
- unit tests: `StreamingRelation.computeStats` and `StreamingExecutionRelation.computeStats`.
- regression tests: `explain join with a normal source` and `explain join with MemoryStream`.
Author: Shixiong Zhu <zsxwing@gmail.com>
Closes#19465 from zsxwing/SPARK-21988.
## What changes were proposed in this pull request?
Currently percentile_approx never returns the first element when percentile is in (relativeError, 1/N], where relativeError default 1/10000, and N is the total number of elements. But ideally, percentiles in [0, 1/N] should all return the first element as the answer.
For example, given input data 1 to 10, if a user queries 10% (or even less) percentile, it should return 1, because the first value 1 already reaches 10%. Currently it returns 2.
Based on the paper, targetError is not rounded up, and searching index should start from 0 instead of 1. By following the paper, we should be able to fix the cases mentioned above.
## How was this patch tested?
Added a new test case and fix existing test cases.
Author: Zhenhua Wang <wzh_zju@163.com>
Closes#19438 from wzhfy/improve_percentile_approx.
This change adds a new SQL config key that is equivalent to SparkContext's
"spark.extraListeners", allowing users to register QueryExecutionListener
instances through the Spark configuration system instead of having to
explicitly do it in code.
The code used by SparkContext to implement the feature was refactored into
a helper method in the Utils class, and SQL's ExecutionListenerManager was
modified to use it to initialize listener declared in the configuration.
Unit tests were added to verify all the new functionality.
Author: Marcelo Vanzin <vanzin@cloudera.com>
Closes#19309 from vanzin/SPARK-19558.
## What changes were proposed in this pull request?
This PR adds an apply() function on df.groupby(). apply() takes a pandas udf that is a transformation on `pandas.DataFrame` -> `pandas.DataFrame`.
Static schema
-------------------
```
schema = df.schema
pandas_udf(schema)
def normalize(df):
df = df.assign(v1 = (df.v1 - df.v1.mean()) / df.v1.std()
return df
df.groupBy('id').apply(normalize)
```
Dynamic schema
-----------------------
**This use case is removed from the PR and we will discuss this as a follow up. See discussion https://github.com/apache/spark/pull/18732#pullrequestreview-66583248**
Another example to use pd.DataFrame dtypes as output schema of the udf:
```
sample_df = df.filter(df.id == 1).toPandas()
def foo(df):
ret = # Some transformation on the input pd.DataFrame
return ret
foo_udf = pandas_udf(foo, foo(sample_df).dtypes)
df.groupBy('id').apply(foo_udf)
```
In interactive use case, user usually have a sample pd.DataFrame to test function `foo` in their notebook. Having been able to use `foo(sample_df).dtypes` frees user from specifying the output schema of `foo`.
Design doc: https://github.com/icexelloss/spark/blob/pandas-udf-doc/docs/pyspark-pandas-udf.md
## How was this patch tested?
* Added GroupbyApplyTest
Author: Li Jin <ice.xelloss@gmail.com>
Author: Takuya UESHIN <ueshin@databricks.com>
Author: Bryan Cutler <cutlerb@gmail.com>
Closes#18732 from icexelloss/groupby-apply-SPARK-20396.
## What changes were proposed in this pull request?
This is a follow-up of #19384.
In the previous pr, only definitions of the config names were modified, but we also need to modify the names in runtime or tests specified as string literal.
## How was this patch tested?
Existing tests but modified the config names.
Author: Takuya UESHIN <ueshin@databricks.com>
Closes#19462 from ueshin/issues/SPARK-22159/fup1.
## What changes were proposed in this pull request?
We should not break the assumption that the length of the allocated byte array is word rounded:
https://github.com/apache/spark/blob/master/sql/catalyst/src/main/java/org/apache/spark/sql/catalyst/expressions/UnsafeRow.java#L170
So we want to use `Integer.MAX_VALUE - 15` instead of `Integer.MAX_VALUE - 8` as the upper bound of an allocated byte array.
cc: srowen gatorsmile
## How was this patch tested?
Since the Spark unit test JVM has less than 1GB heap, here we run the test code as a submit job, so it can run on a JVM has 4GB memory.
Please review http://spark.apache.org/contributing.html before opening a pull request.
Author: Feng Liu <fengliu@databricks.com>
Closes#19460 from liufengdb/fix_array_max.
## What changes were proposed in this pull request?
In state store restore, for each row, put the saved state before the row in the iterator instead of after.
This fixes an issue where agg(last('attr)) will forever return the last value of 'attr from the first microbatch.
## How was this patch tested?
new unit test
Author: Jose Torres <jose@databricks.com>
Closes#19461 from joseph-torres/SPARK-22230.
## What changes were proposed in this pull request?
This updates the broadcast join code path to lazily decompress pages and
iterate through UnsafeRows to prevent all rows from being held in memory
while the broadcast table is being built.
## How was this patch tested?
Existing tests.
Author: Ryan Blue <blue@apache.org>
Closes#19394 from rdblue/broadcast-driver-memory.
## What changes were proposed in this pull request?
When exceeding `spark.sql.codegen.hugeMethodLimit`, the runtime fallbacks to the Volcano iterator solution. This could cause an infinite loop when `FileSourceScanExec` can use the columnar batch to read the data. This PR is to fix the issue.
## How was this patch tested?
Added a test
Author: gatorsmile <gatorsmile@gmail.com>
Closes#19440 from gatorsmile/testt.
## What changes were proposed in this pull request?
Looks like `FlatMapGroupsInRExec.requiredChildDistribution` didn't consider empty grouping attributes. It should be a problem when running `EnsureRequirements` and `gapply` in R can't work on empty grouping columns.
## How was this patch tested?
Added test.
Author: Liang-Chi Hsieh <viirya@gmail.com>
Closes#19436 from viirya/fix-flatmapinr-distribution.
## What changes were proposed in this pull request?
Currently, the group state of user-defined-type is encoded as top-level columns in the UnsafeRows stores in the state store. The timeout timestamp is also saved as (when needed) as the last top-level column. Since the group state is serialized to top-level columns, you cannot save "null" as a value of state (setting null in all the top-level columns is not equivalent). So we don't let the user set the timeout without initializing the state for a key. Based on user experience, this leads to confusion.
This PR is to change the row format such that the state is saved as nested columns. This would allow the state to be set to null, and avoid these confusing corner cases.
## How was this patch tested?
Refactored tests.
Author: Tathagata Das <tathagata.das1565@gmail.com>
Closes#19416 from tdas/SPARK-22187.
## What changes were proposed in this pull request?
By definition the table name in Spark can be something like `123x`, `25a`, etc., with exceptions for literals like `12L`, `23BD`, etc. However, Spark SQL has a special byte length literal, which stops users to use digits followed by `b`, `k`, `m`, `g` as identifiers.
byte length literal is not a standard sql literal and is only used in the `tableSample` parser rule. This PR move the parsing of byte length literal from lexer to parser, so that users can use it as identifiers.
## How was this patch tested?
regression test
Author: Wenchen Fan <wenchen@databricks.com>
Closes#19392 from cloud-fan/parser-bug.
## What changes were proposed in this pull request?
This pr added code to check actual bytecode size when compiling generated code. In #18810, we added code to give up code compilation and use interpreter execution in `SparkPlan` if the line number of generated functions goes over `maxLinesPerFunction`. But, we already have code to collect metrics for compiled bytecode size in `CodeGenerator` object. So,we could easily reuse the code for this purpose.
## How was this patch tested?
Added tests in `WholeStageCodegenSuite`.
Author: Takeshi Yamamuro <yamamuro@apache.org>
Closes#19083 from maropu/SPARK-21871.
## What changes were proposed in this pull request?
This PR abstracts data compressed by `CompressibleColumnAccessor` using `ColumnVector` in batch method. When `ColumnAccessor.decompress` is called, `ColumnVector` will have uncompressed data. This batch decompress does not use `InternalRow` to reduce the number of memory accesses.
As first step of this implementation, this JIRA supports primitive data types. Another PR will support array and other data types.
This implementation decompress data in batch into uncompressed column batch, as rxin suggested at [here](https://github.com/apache/spark/pull/18468#issuecomment-316914076). Another implementation uses adapter approach [as cloud-fan suggested](https://github.com/apache/spark/pull/18468).
## How was this patch tested?
Added test suites
Author: Kazuaki Ishizaki <ishizaki@jp.ibm.com>
Closes#18704 from kiszk/SPARK-20783a.
## What changes were proposed in this pull request?
[SPARK-22193][SQL] Minor typo fix in SortMergeJoinExec. Nothing major, but it bothered me going into.Hence fixing
## How was this patch tested?
existing tests
Author: Rekha Joshi <rekhajoshm@gmail.com>
Author: rjoshi2 <rekhajoshm@gmail.com>
Closes#19422 from rekhajoshm/SPARK-22193.
## What changes were proposed in this pull request?
Allow one-sided outer joins between two streams when a watermark is defined.
## How was this patch tested?
new unit tests
Author: Jose Torres <jose@databricks.com>
Closes#19327 from joseph-torres/outerjoin.
## What changes were proposed in this pull request?
The underlying tables of persistent views are not refreshed when users issue the REFRESH TABLE command against the persistent views.
## How was this patch tested?
Added a test case
Author: gatorsmile <gatorsmile@gmail.com>
Closes#19405 from gatorsmile/refreshView.
## What changes were proposed in this pull request?
The definition of `maxRows` in `LocalLimit` operator was simply wrong. This patch introduces a new `maxRowsPerPartition` method and uses that in pruning. The patch also adds more documentation on why we need local limit vs global limit.
Note that this previously has never been a bug because the way the code is structured, but future use of the maxRows could lead to bugs.
## How was this patch tested?
Should be covered by existing test cases.
Closes#18851
Author: gatorsmile <gatorsmile@gmail.com>
Author: Reynold Xin <rxin@databricks.com>
Closes#19393 from gatorsmile/pr-18851.
## What changes were proposed in this pull request?
This pr fixed an overflow issue below in `Dataset.show`:
```
scala> Seq((1, 2), (3, 4)).toDF("a", "b").show(Int.MaxValue)
org.apache.spark.sql.AnalysisException: The limit expression must be equal to or greater than 0, but got -2147483648;;
GlobalLimit -2147483648
+- LocalLimit -2147483648
+- Project [_1#27218 AS a#27221, _2#27219 AS b#27222]
+- LocalRelation [_1#27218, _2#27219]
at org.apache.spark.sql.catalyst.analysis.CheckAnalysis$class.failAnalysis(CheckAnalysis.scala:41)
at org.apache.spark.sql.catalyst.analysis.Analyzer.failAnalysis(Analyzer.scala:89)
at org.apache.spark.sql.catalyst.analysis.CheckAnalysis$class.org$apache$spark$sql$catalyst$analysis$CheckAnalysis$$checkLimitClause(CheckAnalysis.scala:70)
at org.apache.spark.sql.catalyst.analysis.CheckAnalysis$$anonfun$checkAnalysis$1.apply(CheckAnalysis.scala:234)
at org.apache.spark.sql.catalyst.analysis.CheckAnalysis$$anonfun$checkAnalysis$1.apply(CheckAnalysis.scala:80)
at org.apache.spark.sql.catalyst.trees.TreeNode.foreachUp(TreeNode.scala:127)
```
## How was this patch tested?
Added tests in `DataFrameSuite`.
Author: Takeshi Yamamuro <yamamuro@apache.org>
Closes#19401 from maropu/MaxValueInShowString.
## What changes were proposed in this pull request?
SPARK-21690 makes one-pass `Imputer` by parallelizing the computation of all input columns. When we transform dataset with `ImputerModel`, we do `withColumn` on all input columns sequentially. We can also do this on all input columns at once by adding a `withColumns` API to `Dataset`.
The new `withColumns` API is for internal use only now.
## How was this patch tested?
Existing tests for `ImputerModel`'s change. Added tests for `withColumns` API.
Author: Liang-Chi Hsieh <viirya@gmail.com>
Closes#19229 from viirya/SPARK-22001.
## What changes were proposed in this pull request?
Since the current code ignores WITH clauses to check input relations in TPCDS queries, this leads to inaccurate per-row processing time for benchmark results. For example, in `q2`, this fix could catch all the input relations: `web_sales`, `date_dim`, and `catalog_sales` (the current code catches `date_dim` only). The one-third of the TPCDS queries uses WITH clauses, so I think it is worth fixing this.
## How was this patch tested?
Manually checked.
Author: Takeshi Yamamuro <yamamuro@apache.org>
Closes#19344 from maropu/RespectWithInTPCDSBench.
### What changes were proposed in this pull request?
`tempTables` is not right. To be consistent, we need to rename the internal variable names/comments to tempViews in SessionCatalog too.
### How was this patch tested?
N/A
Author: gatorsmile <gatorsmile@gmail.com>
Closes#19117 from gatorsmile/renameTempTablesToTempViews.
## What changes were proposed in this pull request?
Added IMPALA-modified TPCDS queries to TPC-DS query suites.
- Ref: https://github.com/cloudera/impala-tpcds-kit/tree/master/queries
## How was this patch tested?
N/A
Author: gatorsmile <gatorsmile@gmail.com>
Closes#19386 from gatorsmile/addImpalaQueries.
## What changes were proposed in this pull request?
Spark's RangePartitioner hard codes the number of sampling points per partition to be 20. This is sometimes too low. This ticket makes it configurable, via spark.sql.execution.rangeExchange.sampleSizePerPartition, and raises the default in Spark SQL to be 100.
## How was this patch tested?
Added a pretty sophisticated test based on chi square test ...
Author: Reynold Xin <rxin@databricks.com>
Closes#19387 from rxin/SPARK-22160.
## What changes were proposed in this pull request?
For some reason when we added the Exec suffix to all physical operators, we missed this one. I was looking for this physical operator today and couldn't find it, because I was looking for ExchangeExec.
## How was this patch tested?
This is a simple rename and should be covered by existing tests.
Author: Reynold Xin <rxin@databricks.com>
Closes#19376 from rxin/SPARK-22153.
## What changes were proposed in this pull request?
Now, we are not running TPC-DS queries as regular test cases. Thus, we need to add a test suite using empty tables for ensuring the new code changes will not break them. For example, optimizer/analyzer batches should not exceed the max iteration.
## How was this patch tested?
N/A
Author: gatorsmile <gatorsmile@gmail.com>
Closes#19361 from gatorsmile/tpcdsQuerySuite.
## What changes were proposed in this pull request?
`WriteableColumnVector` does not close its child column vectors. This can create memory leaks for `OffHeapColumnVector` where we do not clean up the memory allocated by a vectors children. This can be especially bad for string columns (which uses a child byte column vector).
## How was this patch tested?
I have updated the existing tests to always use both on-heap and off-heap vectors. Testing and diagnoses was done locally.
Author: Herman van Hovell <hvanhovell@databricks.com>
Closes#19367 from hvanhovell/SPARK-22143.
## What changes were proposed in this pull request?
Currently we use Arrow File format to communicate with Python worker when invoking vectorized UDF but we can use Arrow Stream format.
This pr replaces the Arrow File format with the Arrow Stream format.
## How was this patch tested?
Existing tests.
Author: Takuya UESHIN <ueshin@databricks.com>
Closes#19349 from ueshin/issues/SPARK-22125.
## What changes were proposed in this pull request?
When inferring constraints from children, Join's condition can be simplified as None.
For example,
```
val testRelation = LocalRelation('a.int)
val x = testRelation.as("x")
val y = testRelation.where($"a" === 2 && !($"a" === 2)).as("y")
x.join.where($"x.a" === $"y.a")
```
The plan will become
```
Join Inner
:- LocalRelation <empty>, [a#23]
+- LocalRelation <empty>, [a#224]
```
And the Cartesian products check will throw exception for above plan.
Propagate empty relation before checking Cartesian products, and the issue is resolved.
## How was this patch tested?
Unit test
Author: Wang Gengliang <ltnwgl@gmail.com>
Closes#19362 from gengliangwang/MoveCheckCartesianProducts.
## What changes were proposed in this pull request?
Address PR comments that appeared post-merge, to rename `addExtraCode` to `addInnerClass`,
and not count the size of the inner class to the size of the outer class.
## How was this patch tested?
YOLO.
Author: Juliusz Sompolski <julek@databricks.com>
Closes#19353 from juliuszsompolski/SPARK-22103followup.
## What changes were proposed in this pull request?
We can override `usedInputs` to claim that an operator defers input evaluation. `Sample` and `Limit` are two operators which should claim it but don't. We should do it.
## How was this patch tested?
Existing tests.
Author: Liang-Chi Hsieh <viirya@gmail.com>
Closes#19345 from viirya/SPARK-22124.
## What changes were proposed in this pull request?
This change disables the use of 0-parameter pandas_udfs due to the API being overly complex and awkward, and can easily be worked around by using an index column as an input argument. Also added doctests for pandas_udfs which revealed bugs for handling empty partitions and using the pandas_udf decorator.
## How was this patch tested?
Reworked existing 0-parameter test to verify error is raised, added doctest for pandas_udf, added new tests for empty partition and decorator usage.
Author: Bryan Cutler <cutlerb@gmail.com>
Closes#19325 from BryanCutler/arrow-pandas_udf-0-param-remove-SPARK-22106.
## What changes were proposed in this pull request?
HashAggregateExec codegen uses two paths for fast hash table and a generic one.
It generates code paths for iterating over both, and both code paths generate the consume code of the parent operator, resulting in that code being expanded twice.
This leads to a long generated function that might be an issue for the compiler (see e.g. SPARK-21603).
I propose to remove the double expansion by generating the consume code in a helper function that can just be called from both iterating loops.
An issue with separating the `consume` code to a helper function was that a number of places relied and assumed on being in the scope of an outside `produce` loop and e.g. use `continue` to jump out.
I replaced such code flows with nested scopes. It is code that should be handled the same by compiler, while getting rid of depending on assumptions that are outside of the `consume`'s own scope.
## How was this patch tested?
Existing test coverage.
Author: Juliusz Sompolski <julek@databricks.com>
Closes#19324 from juliuszsompolski/aggrconsumecodegen.
## What changes were proposed in this pull request?
The `percentile_approx` function previously accepted numeric type input and output double type results.
But since all numeric types, date and timestamp types are represented as numerics internally, `percentile_approx` can support them easily.
After this PR, it supports date type, timestamp type and numeric types as input types. The result type is also changed to be the same as the input type, which is more reasonable for percentiles.
This change is also required when we generate equi-height histograms for these types.
## How was this patch tested?
Added a new test and modified some existing tests.
Author: Zhenhua Wang <wangzhenhua@huawei.com>
Closes#19321 from wzhfy/approx_percentile_support_types.
## What changes were proposed in this pull request?
Enable Scala 2.12 REPL. Fix most remaining issues with 2.12 compilation and warnings, including:
- Selecting Kafka 0.10.1+ for Scala 2.12 and patching over a minor API difference
- Fixing lots of "eta expansion of zero arg method deprecated" warnings
- Resolving the SparkContext.sequenceFile implicits compile problem
- Fixing an odd but valid jetty-server missing dependency in hive-thriftserver
## How was this patch tested?
Existing tests
Author: Sean Owen <sowen@cloudera.com>
Closes#19307 from srowen/Scala212.
## What changes were proposed in this pull request?
The implemented `isCascadingTruncateTable` in `AggregatedDialect` is wrong. When no dialect claims cascading, once there is an unknown cascading truncate in the dialects, we should return unknown cascading, instead of false.
## How was this patch tested?
Added test.
Author: Liang-Chi Hsieh <viirya@gmail.com>
Closes#19286 from viirya/SPARK-21338-followup.
## What changes were proposed in this pull request?
This PR proposes to resolve the type conflicts in strings and timestamps in partition column values.
It looks we need to set the timezone as it needs a cast between strings and timestamps.
```scala
val df = Seq((1, "2015-01-01 00:00:00"), (2, "2014-01-01 00:00:00"), (3, "blah")).toDF("i", "str")
val path = "/tmp/test.parquet"
df.write.format("parquet").partitionBy("str").save(path)
spark.read.parquet(path).show()
```
**Before**
```
java.util.NoSuchElementException: None.get
at scala.None$.get(Option.scala:347)
at scala.None$.get(Option.scala:345)
at org.apache.spark.sql.catalyst.expressions.TimeZoneAwareExpression$class.timeZone(datetimeExpressions.scala:46)
at org.apache.spark.sql.catalyst.expressions.Cast.timeZone$lzycompute(Cast.scala:172)
at org.apache.spark.sql.catalyst.expressions.Cast.timeZone(Cast.scala:172)
at org.apache.spark.sql.catalyst.expressions.Cast$$anonfun$castToString$3$$anonfun$apply$16.apply(Cast.scala:208)
at org.apache.spark.sql.catalyst.expressions.Cast$$anonfun$castToString$3$$anonfun$apply$16.apply(Cast.scala:208)
at org.apache.spark.sql.catalyst.expressions.Cast.org$apache$spark$sql$catalyst$expressions$Cast$$buildCast(Cast.scala:201)
at org.apache.spark.sql.catalyst.expressions.Cast$$anonfun$castToString$3.apply(Cast.scala:207)
at org.apache.spark.sql.catalyst.expressions.Cast.nullSafeEval(Cast.scala:533)
at org.apache.spark.sql.catalyst.expressions.UnaryExpression.eval(Expression.scala:331)
at org.apache.spark.sql.execution.datasources.PartitioningUtils$$anonfun$org$apache$spark$sql$execution$datasources$PartitioningUtils$$resolveTypeConflicts$1.apply(PartitioningUtils.scala:481)
at org.apache.spark.sql.execution.datasources.PartitioningUtils$$anonfun$org$apache$spark$sql$execution$datasources$PartitioningUtils$$resolveTypeConflicts$1.apply(PartitioningUtils.scala:480)
at scala.collection.TraversableLike$$anonfun$map$1.apply(TraversableLike.scala:234)
at scala.collection.TraversableLike$$anonfun$map$1.apply(TraversableLike.scala:234)
at scala.collection.mutable.ResizableArray$class.foreach(ResizableArray.scala:59)
```
**After**
```
+---+-------------------+
| i| str|
+---+-------------------+
| 2|2014-01-01 00:00:00|
| 1|2015-01-01 00:00:00|
| 3| blah|
+---+-------------------+
```
## How was this patch tested?
Unit tests added in `ParquetPartitionDiscoverySuite` and manual tests.
Author: hyukjinkwon <gurwls223@gmail.com>
Closes#19331 from HyukjinKwon/SPARK-22109.
## What changes were proposed in this pull request?
Try to avoid allocating an array bigger than Integer.MAX_VALUE - 8, which is the actual max size on some JVMs, in several places
## How was this patch tested?
Existing tests
Author: Sean Owen <sowen@cloudera.com>
Closes#19266 from srowen/SPARK-22033.
## What changes were proposed in this pull request?
`OffHeapColumnVector.reserveInternal()` will only copy already inserted values during reallocation if `data != null`. In vectors containing arrays or structs this is incorrect, since there field `data` is not used at all. We need to check `nulls` instead.
## How was this patch tested?
Adds new tests to `ColumnVectorSuite` that reproduce the errors.
Author: Ala Luszczak <ala@databricks.com>
Closes#19308 from ala/vector-realloc.
This PR adds vectorized UDFs to the Python API
**Proposed API**
Introduce a flag to turn on vectorization for a defined UDF, for example:
```
pandas_udf(DoubleType())
def plus(a, b)
return a + b
```
or
```
plus = pandas_udf(lambda a, b: a + b, DoubleType())
```
Usage is the same as normal UDFs
0-parameter UDFs
pandas_udf functions can declare an optional `**kwargs` and when evaluated, will contain a key "size" that will give the required length of the output. For example:
```
pandas_udf(LongType())
def f0(**kwargs):
return pd.Series(1).repeat(kwargs["size"])
df.select(f0())
```
Added new unit tests in pyspark.sql that are enabled if pyarrow and Pandas are available.
- [x] Fix support for promoted types with null values
- [ ] Discuss 0-param UDF API (use of kwargs)
- [x] Add tests for chained UDFs
- [ ] Discuss behavior when pyarrow not installed / enabled
- [ ] Cleanup pydoc and add user docs
Author: Bryan Cutler <cutlerb@gmail.com>
Author: Takuya UESHIN <ueshin@databricks.com>
Closes#18659 from BryanCutler/arrow-vectorized-udfs-SPARK-21404.
## What changes were proposed in this pull request?
Right now the calculation of SortMergeJoinExec's outputOrdering relies on the fact that its children have already been sorted on the join keys, while this is often not true until EnsureRequirements has been applied. So we ended up not getting the correct outputOrdering during physical planning stage before Sort nodes are added to the children.
For example, J = {A join B on key1 = key2}
1. if A is NOT ordered on key1 ASC, J's outputOrdering should include "key1 ASC"
2. if A is ordered on key1 ASC, J's outputOrdering should include "key1 ASC"
3. if A is ordered on key1 ASC, with sameOrderExp=c1, J's outputOrdering should include "key1 ASC, sameOrderExp=c1"
So to fix this I changed the behavior of <code>getKeyOrdering(keys, childOutputOrdering)</code> to:
1. If the childOutputOrdering satisfies (is a superset of) the required child ordering => childOutputOrdering
2. Otherwise => required child ordering
In addition, I organized the logic for deciding the relationship between two orderings into SparkPlan, so that it can be reused by EnsureRequirements and SortMergeJoinExec, and potentially other classes.
## How was this patch tested?
Added new test cases.
Passed all integration tests.
Author: maryannxue <maryann.xue@gmail.com>
Closes#19281 from maryannxue/spark-21998.
## What changes were proposed in this pull request?
`processAllAvailable` should also check the query state and if the query is stopped, it should return.
## How was this patch tested?
The new unit test.
Author: Shixiong Zhu <zsxwing@gmail.com>
Closes#19314 from zsxwing/SPARK-22094.
## What changes were proposed in this pull request?
#### Architecture
This PR implements stream-stream inner join using a two-way symmetric hash join. At a high level, we want to do the following.
1. For each stream, we maintain the past rows as state in State Store.
- For each joining key, there can be multiple rows that have been received.
- So, we have to effectively maintain a key-to-list-of-values multimap as state for each stream.
2. In each batch, for each input row in each stream
- Look up the other streams state to see if there are matching rows, and output them if they satisfy the joining condition
- Add the input row to corresponding stream’s state.
- If the data has a timestamp/window column with watermark, then we will use that to calculate the threshold for keys that are required to buffered for future matches and drop the rest from the state.
Cleaning up old unnecessary state rows depends completely on whether watermark has been defined and what are join conditions. We definitely want to support state clean up two types of queries that are likely to be common.
- Queries to time range conditions - E.g. `SELECT * FROM leftTable, rightTable ON leftKey = rightKey AND leftTime > rightTime - INTERVAL 8 MINUTES AND leftTime < rightTime + INTERVAL 1 HOUR`
- Queries with windows as the matching key - E.g. `SELECT * FROM leftTable, rightTable ON leftKey = rightKey AND window(leftTime, "1 hour") = window(rightTime, "1 hour")` (pseudo-SQL)
#### Implementation
The stream-stream join is primarily implemented in three classes
- `StreamingSymmetricHashJoinExec` implements the above symmetric join algorithm.
- `SymmetricsHashJoinStateManagers` manages the streaming state for the join. This essentially is a fault-tolerant key-to-list-of-values multimap built on the StateStore APIs. `StreamingSymmetricHashJoinExec` instantiates two such managers, one for each join side.
- `StreamingSymmetricHashJoinExecHelper` is a helper class to extract threshold for the state based on the join conditions and the event watermark.
Refer to the scaladocs class for more implementation details.
Besides the implementation of stream-stream inner join SparkPlan. Some additional changes are
- Allowed inner join in append mode in UnsupportedOperationChecker
- Prevented stream-stream join on an empty batch dataframe to be collapsed by the optimizer
## How was this patch tested?
- New tests in StreamingJoinSuite
- Updated tests UnsupportedOperationSuite
Author: Tathagata Das <tathagata.das1565@gmail.com>
Closes#19271 from tdas/SPARK-22053.
## What changes were proposed in this pull request?
I test on a dataset of about 13M instances, and found that using `treeAggregate` give a speedup in following algs:
|Algs| SpeedUp |
|------|-----------|
|OneHotEncoder| 5% |
|StatFunctions.calculateCov| 7% |
|StatFunctions.multipleApproxQuantiles| 9% |
|RegressionEvaluator| 8% |
## How was this patch tested?
existing tests
Author: Zheng RuiFeng <ruifengz@foxmail.com>
Closes#19232 from zhengruifeng/use_treeAggregate.
## What changes were proposed in this pull request?
Adjust EnsureStatefulOpPartitioningSuite to use scalatest lifecycle normally instead of constructor; fixes:
```
*** RUN ABORTED ***
org.apache.spark.SparkException: Only one SparkContext may be running in this JVM (see SPARK-2243). To ignore this error, set spark.driver.allowMultipleContexts = true. The currently running SparkContext was created at:
org.apache.spark.sql.streaming.EnsureStatefulOpPartitioningSuite.<init>(EnsureStatefulOpPartitioningSuite.scala:35)
```
## How was this patch tested?
Existing tests
Author: Sean Owen <sowen@cloudera.com>
Closes#19306 from srowen/SPARK-21977.2.
## What changes were proposed in this pull request?
Clarify behavior of to_utc_timestamp/from_utc_timestamp with an example
## How was this patch tested?
Doc only change / existing tests
Author: Sean Owen <sowen@cloudera.com>
Closes#19276 from srowen/SPARK-22049.
## What changes were proposed in this pull request?
Update plugins, including scala-maven-plugin, to latest versions. Update checkstyle to 8.2. Remove bogus checkstyle config and enable it. Fix existing and new Java checkstyle errors.
## How was this patch tested?
Existing tests
Author: Sean Owen <sowen@cloudera.com>
Closes#19282 from srowen/SPARK-22066.
## What changes were proposed in this pull request?
This is a bit hard to explain as there are several issues here, I'll try my best. Here are the requirements:
1. A StructuredStreaming Source that can generate empty RDDs with 0 partitions
2. A StructuredStreaming query that uses the above source, performs a stateful aggregation
(mapGroupsWithState, groupBy.count, ...), and coalesce's by 1
The crux of the problem is that when a dataset has a `coalesce(1)` call, it receives a `SinglePartition` partitioning scheme. This scheme satisfies most required distributions used for aggregations such as HashAggregateExec. This causes a world of problems:
Symptom 1. If the input RDD has 0 partitions, the whole lineage will receive 0 partitions, nothing will be executed, the state store will not create any delta files. When this happens, the next trigger fails, because the StateStore fails to load the delta file for the previous trigger
Symptom 2. Let's say that there was data. Then in this case, if you stop your stream, and change `coalesce(1)` with `coalesce(2)`, then restart your stream, your stream will fail, because `spark.sql.shuffle.partitions - 1` number of StateStores will fail to find its delta files.
To fix the issues above, we must check that the partitioning of the child of a `StatefulOperator` satisfies:
If the grouping expressions are empty:
a) AllTuple distribution
b) Single physical partition
If the grouping expressions are non empty:
a) Clustered distribution
b) spark.sql.shuffle.partition # of partitions
whether or not `coalesce(1)` exists in the plan, and whether or not the input RDD for the trigger has any data.
Once you fix the above problem by adding an Exchange to the plan, you come across the following bug:
If you call `coalesce(1).groupBy().count()` on a Streaming DataFrame, and if you have a trigger with no data, `StateStoreRestoreExec` doesn't return the prior state. However, for this specific aggregation, `HashAggregateExec` after the restore returns a (0, 0) row, since we're performing a count, and there is no data. Then this data gets stored in `StateStoreSaveExec` causing the previous counts to be overwritten and lost.
## How was this patch tested?
Regression tests
Author: Burak Yavuz <brkyvz@gmail.com>
Closes#19196 from brkyvz/sa-0.
This change modifies the live listener bus so that all listeners are
added to queues; each queue has its own thread to dispatch events,
making it possible to separate slow listeners from other more
performance-sensitive ones.
The public API has not changed - all listeners added with the existing
"addListener" method, which after this change mostly means all
user-defined listeners, end up in a default queue. Internally, there's
an API allowing listeners to be added to specific queues, and that API
is used to separate the internal Spark listeners into 3 categories:
application status listeners (e.g. UI), executor management (e.g. dynamic
allocation), and the event log.
The queueing logic, while abstracted away in a separate class, is kept
as much as possible hidden away from consumers. Aside from choosing their
queue, there's no code change needed to take advantage of queues.
Test coverage relies on existing tests; a few tests had to be tweaked
because they relied on `LiveListenerBus.postToAll` being synchronous,
and the change makes that method asynchronous. Other tests were simplified
not to use the asynchronous LiveListenerBus.
Author: Marcelo Vanzin <vanzin@cloudera.com>
Closes#19211 from vanzin/SPARK-18838.
