## What changes were proposed in this pull request?
When a null in a nested field in struct, casting from the struct throws error, currently.
```scala
scala> sql("select cast(struct(1, null) as struct<a:int,b:int>)").show
scala.MatchError: NullType (of class org.apache.spark.sql.types.NullType$)
at org.apache.spark.sql.catalyst.expressions.Cast.castToInt(Cast.scala:447)
at org.apache.spark.sql.catalyst.expressions.Cast.cast(Cast.scala:635)
at org.apache.spark.sql.catalyst.expressions.Cast.$anonfun$castStruct$1(Cast.scala:603)
```
Similarly, inline table, which casts null in nested field under the hood, also throws an error.
```scala
scala> sql("select * FROM VALUES (('a', (10, null))), (('b', (10, 50))), (('c', null)) AS tab(x, y)").show
org.apache.spark.sql.AnalysisException: failed to evaluate expression named_struct('col1', 10, 'col2', NULL): NullType (of class org.apache.spark.sql.t
ypes.NullType$); line 1 pos 14
at org.apache.spark.sql.catalyst.analysis.package$AnalysisErrorAt.failAnalysis(package.scala:47)
at org.apache.spark.sql.catalyst.analysis.ResolveInlineTables.$anonfun$convert$6(ResolveInlineTables.scala:106)
```
This fixes the issue.
## How was this patch tested?
Added tests.
Closes#24576 from viirya/cast-null.
Authored-by: Liang-Chi Hsieh <viirya@gmail.com>
Signed-off-by: Dongjoon Hyun <dhyun@apple.com>
## What changes were proposed in this pull request?
This pr fix hadoop-3.2 test issues(except the `hive-thriftserver` module):
1. Add `hive.metastore.schema.verification` and `datanucleus.schema.autoCreateAll` to HiveConf.
2. hadoop-3.2 support access the Hive metastore from 0.12 to 2.2
After [SPARK-27176](https://issues.apache.org/jira/browse/SPARK-27176) and this PR, we upgraded the built-in Hive to 2.3 when enabling the Hadoop 3.2+ profile. This upgrade fixes the following issues:
- [HIVE-6727](https://issues.apache.org/jira/browse/HIVE-6727): Table level stats for external tables are set incorrectly.
- [HIVE-15653](https://issues.apache.org/jira/browse/HIVE-15653): Some ALTER TABLE commands drop table stats.
- [SPARK-12014](https://issues.apache.org/jira/browse/SPARK-12014): Spark SQL query containing semicolon is broken in Beeline.
- [SPARK-25193](https://issues.apache.org/jira/browse/SPARK-25193): insert overwrite doesn't throw exception when drop old data fails.
- [SPARK-25919](https://issues.apache.org/jira/browse/SPARK-25919): Date value corrupts when tables are "ParquetHiveSerDe" formatted and target table is Partitioned.
- [SPARK-26332](https://issues.apache.org/jira/browse/SPARK-26332): Spark sql write orc table on viewFS throws exception.
- [SPARK-26437](https://issues.apache.org/jira/browse/SPARK-26437): Decimal data becomes bigint to query, unable to query.
## How was this patch tested?
This pr test Spark’s Hadoop 3.2 profile on jenkins and #24591 test Spark’s Hadoop 2.7 profile on jenkins
This PR close#24591Closes#24391 from wangyum/SPARK-27402.
Authored-by: Yuming Wang <yumwang@ebay.com>
Signed-off-by: gatorsmile <gatorsmile@gmail.com>
## What changes were proposed in this pull request?
This PR goes to add `max_by()` and `min_by()` SQL aggregate functions.
Quoting from the [Presto docs](https://prestodb.github.io/docs/current/functions/aggregate.html#max_by)
> max_by(x, y) → [same as x]
> Returns the value of x associated with the maximum value of y over all input values.
`min_by()` works similarly.
## How was this patch tested?
Added tests.
Closes#24557 from viirya/SPARK-27653.
Authored-by: Liang-Chi Hsieh <viirya@gmail.com>
Signed-off-by: Wenchen Fan <wenchen@databricks.com>
## What changes were proposed in this pull request?
Currently,thread number of broadcast-exchange thread pool is fixed and keepAliveSeconds is also fixed as 60s.
```
object BroadcastExchangeExec {
private[execution] val executionContext = ExecutionContext.fromExecutorService(
ThreadUtils.newDaemonCachedThreadPool("broadcast-exchange", 128))
}
/**
* Create a cached thread pool whose max number of threads is `maxThreadNumber`. Thread names
* are formatted as prefix-ID, where ID is a unique, sequentially assigned integer.
*/
def newDaemonCachedThreadPool(
prefix: String, maxThreadNumber: Int, keepAliveSeconds: Int = 60): ThreadPoolExecutor = {
val threadFactory = namedThreadFactory(prefix)
val threadPool = new ThreadPoolExecutor(
maxThreadNumber, // corePoolSize: the max number of threads to create before queuing the tasks
maxThreadNumber, // maximumPoolSize: because we use LinkedBlockingDeque, this one is not used
keepAliveSeconds,
TimeUnit.SECONDS,
new LinkedBlockingQueue[Runnable],
threadFactory)
threadPool.allowCoreThreadTimeOut(true)
threadPool
}
```
But some times, if the Thead object do not GC quickly it may caused server(driver) OOM. In such case,we need to make this thread pool configurable.
A case has described in https://issues.apache.org/jira/browse/SPARK-26601
## How was this patch tested?
UT
Closes#23670 from caneGuy/zhoukang/make-broadcat-config.
Authored-by: zhoukang <zhoukang199191@gmail.com>
Signed-off-by: HyukjinKwon <gurwls223@apache.org>
## What changes were proposed in this pull request?
Remove the dead code of Spark Repl in Scala 2.11. We only support Scala 2.12
## How was this patch tested?
N/A
Closes#24588 from gatorsmile/removeDeadCode.
Authored-by: gatorsmile <gatorsmile@gmail.com>
Signed-off-by: Wenchen Fan <wenchen@databricks.com>
## What changes were proposed in this pull request?
In File source V1, the statistics of `HadoopFsRelation` is `compressionFactor * sizeInBytesOfAllFiles`.
To follow it, we can implement the interface SupportsReportStatistics in FileScan and report the same statistics.
## How was this patch tested?
Unit test
Closes#24571 from gengliangwang/stats.
Authored-by: Gengliang Wang <gengliang.wang@databricks.com>
Signed-off-by: Wenchen Fan <wenchen@databricks.com>
## What changes were proposed in this pull request?
[SPARK-27343](https://issues.apache.org/jira/projects/SPARK/issues/SPARK-27343)
Based on the previous PR: https://github.com/apache/spark/pull/24270
Extracting parameters , building the objects of ConfigEntry.
For example:
for the parameter "spark.kafka.producer.cache.timeout",we build
```
private[kafka010] val PRODUCER_CACHE_TIMEOUT =
ConfigBuilder("spark.kafka.producer.cache.timeout")
.doc("The expire time to remove the unused producers.")
.timeConf(TimeUnit.MILLISECONDS)
.createWithDefaultString("10m")
```
Closes#24574 from hehuiyuan/hehuiyuan-patch-9.
Authored-by: hehuiyuan <hehuiyuan@ZBMAC-C02WD3K5H.local>
Signed-off-by: Sean Owen <sean.owen@databricks.com>
## What changes were proposed in this pull request?
To move DS v2 API to the catalyst module, we can't refer to an internal class (`MutableColumnarRow`) in `ColumnarBatch`.
This PR creates a read-only version of `MutableColumnarRow`, and use it in `ColumnarBatch`.
close https://github.com/apache/spark/pull/24546
## How was this patch tested?
existing tests
Closes#24581 from cloud-fan/mutable-row.
Authored-by: Wenchen Fan <wenchen@databricks.com>
Signed-off-by: HyukjinKwon <gurwls223@apache.org>
## What changes were proposed in this pull request?
added the missing config for structured streaming when using file source.
from the code we have
```
/**
* Maximum age of a file that can be found in this directory, before it is ignored. For the
* first batch all files will be considered valid. If `latestFirst` is set to `true` and
* `maxFilesPerTrigger` is set, then this parameter will be ignored, because old files that are
* valid, and should be processed, may be ignored. Please refer to SPARK-19813 for details.
*
* The max age is specified with respect to the timestamp of the latest file, and not the
* timestamp of the current system. That this means if the last file has timestamp 1000, and the
* current system time is 2000, and max age is 200, the system will purge files older than
* 800 (rather than 1800) from the internal state.
*
* Default to a week.
*/
val maxFileAgeMs: Long =
Utils.timeStringAsMs(parameters.getOrElse("maxFileAge", "7d"))
```
which is not documented.
also the file processing order was not mentioned but in the code we specifically select the file list based on file mtime:
```scala
private val fileSortOrder = if (sourceOptions.latestFirst) {
logWarning(
"""'latestFirst' is true. New files will be processed first, which may affect the watermark
|value. In addition, 'maxFileAge' will be ignored.""".stripMargin)
implicitly[Ordering[Long]].reverse
} else {
implicitly[Ordering[Long]]
}
val files = allFiles.sortBy(_.getModificationTime)(fileSortOrder).map { status =>
(status.getPath.toUri.toString, status.getModificationTime)
}
```
---------
---------
Closes#24548 from linehrr/master.
Lead-authored-by: ryne.yang <ryne.yang@acuityads.com>
Co-authored-by: Ryne Yang <ryne.yang@acuityads.com>
Co-authored-by: linehrr <linehrr@gmail.com>
Signed-off-by: Dongjoon Hyun <dhyun@apple.com>
## What changes were proposed in this pull request?
This patch fixes a bug where `--supervised` Spark jobs would retry multiple times whenever an agent would crash, come back, and re-register even when those jobs had already relaunched on a different agent.
That is:
```
- supervised driver is running on agent1
- agent1 crashes
- driver is relaunched on another agent as `<task-id>-retry-1`
- agent1 comes back online and re-registers with scheduler
- spark relaunches the same job as `<task-id>-retry-2`
- now there are two jobs running simultaneously
```
This is because when an agent would come back and re-register it would send a status update `TASK_FAILED` for its old driver-task. Previous logic would indiscriminately remove the `submissionId` from Zookeeper's `launchedDrivers` node and add it to `retryList` node. Then, when a new offer came in, it would relaunch another `-retry-` task even though one was previously running.
For example logs, scroll to bottom
## How was this patch tested?
- Added a unit test to simulate behavior described above
- Tested manually on a DC/OS cluster by
```
- launching a --supervised spark job
- dcos node ssh <to the agent with the running spark-driver>
- systemctl stop dcos-mesos-slave
- docker kill <driver-container-id>
- [ wait until spark job is relaunched ]
- systemctl start dcos-mesos-slave
- [ observe spark driver is not relaunched as `-retry-2` ]
```
Log snippets included below. Notice the `-retry-1` task is running when status update for the old task comes in afterward:
```
19/01/15 19:21:38 TRACE MesosClusterScheduler: Received offers from Mesos:
... [offers] ...
19/01/15 19:21:39 TRACE MesosClusterScheduler: Using offer 5d421001-0630-4214-9ecb-d5838a2ec149-O2532 to launch driver driver-20190115192138-0001 with taskId: value: "driver-20190115192138-0001"
...
19/01/15 19:21:42 INFO MesosClusterScheduler: Received status update: taskId=driver-20190115192138-0001 state=TASK_STARTING message=''
19/01/15 19:21:43 INFO MesosClusterScheduler: Received status update: taskId=driver-20190115192138-0001 state=TASK_RUNNING message=''
...
19/01/15 19:29:12 INFO MesosClusterScheduler: Received status update: taskId=driver-20190115192138-0001 state=TASK_LOST message='health check timed out' reason=REASON_SLAVE_REMOVED
...
19/01/15 19:31:12 TRACE MesosClusterScheduler: Using offer 5d421001-0630-4214-9ecb-d5838a2ec149-O2681 to launch driver driver-20190115192138-0001 with taskId: value: "driver-20190115192138-0001-retry-1"
...
19/01/15 19:31:15 INFO MesosClusterScheduler: Received status update: taskId=driver-20190115192138-0001-retry-1 state=TASK_STARTING message=''
19/01/15 19:31:16 INFO MesosClusterScheduler: Received status update: taskId=driver-20190115192138-0001-retry-1 state=TASK_RUNNING message=''
...
19/01/15 19:33:45 INFO MesosClusterScheduler: Received status update: taskId=driver-20190115192138-0001 state=TASK_FAILED message='Unreachable agent re-reregistered'
...
19/01/15 19:33:45 INFO MesosClusterScheduler: Received status update: taskId=driver-20190115192138-0001 state=TASK_FAILED message='Abnormal executor termination: unknown container' reason=REASON_EXECUTOR_TERMINATED
19/01/15 19:33:45 ERROR MesosClusterScheduler: Unable to find driver with driver-20190115192138-0001 in status update
...
19/01/15 19:33:47 TRACE MesosClusterScheduler: Using offer 5d421001-0630-4214-9ecb-d5838a2ec149-O2729 to launch driver driver-20190115192138-0001 with taskId: value: "driver-20190115192138-0001-retry-2"
...
19/01/15 19:33:50 INFO MesosClusterScheduler: Received status update: taskId=driver-20190115192138-0001-retry-2 state=TASK_STARTING message=''
19/01/15 19:33:51 INFO MesosClusterScheduler: Received status update: taskId=driver-20190115192138-0001-retry-2 state=TASK_RUNNING message=''
```
Closes#24276 from samvantran/SPARK-27347-duplicate-retries.
Authored-by: Sam Tran <stran@mesosphere.com>
Signed-off-by: Dongjoon Hyun <dhyun@apple.com>
## What changes were proposed in this pull request?
For the below three thread configuration items applied to both driver and executor,
spark.rpc.io.serverThreads
spark.rpc.io.clientThreads
spark.rpc.netty.dispatcher.numThreads,
we separate them to driver specifics and executor specifics.
spark.driver.rpc.io.serverThreads < - > spark.executor.rpc.io.serverThreads
spark.driver.rpc.io.clientThreads < - > spark.executor.rpc.io.clientThreads
spark.driver.rpc.netty.dispatcher.numThreads < - > spark.executor.rpc.netty.dispatcher.numThreads
Spark reads these specifics first and fall back to the common configurations.
## How was this patch tested?
We ran the SimpleMap app without shuffle for benchmark purpose to test Spark's scalability in HPC with omini-path NIC which has higher bandwidth than normal ethernet NIC.
Spark's base version is 2.4.0.
Spark ran in the Standalone mode. Driver was in a standalone node.
After the separation, the performance is improved a lot in 256 nodes and 512 nodes. see below test results of SimpleMapTask before and after the enhancement. You can view the tables in the [JIRA](https://issues.apache.org/jira/browse/SPARK-26632) too.
ds: spark.driver.rpc.io.serverThreads
dc: spark.driver.rpc.io.clientThreads
dd: spark.driver.rpc.netty.dispatcher.numThreads
ed: spark.executor.rpc.netty.dispatcher.numThreads
time: Overall Time (s)
old time: Overall Time without Separation (s)
**Before:**
nodes | ds | dc | dd | ed | time
-- |-- | -- | -- | -- | --
128 nodes | 8 | 8 | 8 | 8 | 108
256 nodes | 8 | 8 | 8 | 8 | 196
512 nodes | 8 | 8 | 8 | 8 | 377
**After:**
nodes | ds | dc | dd | ed | time | improvement
-- | -- | -- | -- | -- | -- | --
128 nodes | 15 | 15 | 10 | 30 | 107 | 0.9%
256 nodes | 12 | 15 | 10 | 30 | 159 | 18.8%
512 nodes | 12 | 15 | 10 | 30 | 283 | 24.9%
Closes#23560 from zjf2012/thread_conf_separation.
Authored-by: jiafu.zhang@intel.com <jiafu.zhang@intel.com>
Signed-off-by: Marcelo Vanzin <vanzin@cloudera.com>
## What changes were proposed in this pull request?
We should add since info to all expressions.
SPARK-7886 Rand / Randn
af3746ce0d RLike, Like (I manually checked that it exists from 1.0.0)
SPARK-8262 Split
SPARK-8256 RegExpReplace
SPARK-8255 RegExpExtract
9aadcffabd Coalesce / IsNull / IsNotNull (I manually checked that it exists from 1.0.0)
SPARK-14541 IfNull / NullIf / Nvl / Nvl2
SPARK-9080 IsNaN
SPARK-9168 NaNvl
## How was this patch tested?
N/A
Closes#24579 from HyukjinKwon/SPARK-27673.
Authored-by: HyukjinKwon <gurwls223@apache.org>
Signed-off-by: Dongjoon Hyun <dhyun@apple.com>
## What changes were proposed in this pull request?
- the accumulator warning is too verbose
- when a test fails with schema mismatch, you never see the error message / exception
Closes#24549 from ericl/test-nits.
Lead-authored-by: Eric Liang <ekl@databricks.com>
Co-authored-by: Eric Liang <ekhliang@gmail.com>
Signed-off-by: Wenchen Fan <wenchen@databricks.com>
## What changes were proposed in this pull request?
If a type is annotated, `ScalaReflection` can fail if the datatype is an `Option`, a `Seq`, a `Map` and other similar types. This is because it assumes we are dealing with `TypeRef`, while types with annotations are `AnnotatedType`.
The PR deals with the case the annotation is present.
## How was this patch tested?
added UT
Closes#24564 from mgaido91/SPARK-27625.
Authored-by: Marco Gaido <marcogaido91@gmail.com>
Signed-off-by: Wenchen Fan <wenchen@databricks.com>
We updated our website a long time ago to describe Spark as the unified analytics engine, which is also how Spark is described in the community now. But our README and docs page still use the same description from 2011 ... This patch updates them.
The patch also updates the README example to use more modern APIs, and refer to Structured Streaming rather than Spark Streaming.
Closes#24573 from rxin/consistent-message.
Authored-by: Reynold Xin <rxin@databricks.com>
Signed-off-by: HyukjinKwon <gurwls223@apache.org>
## What changes were proposed in this pull request?
Test steps to reproduce:
1) bin/spark-shell
```
val dataset = spark.createDataFrame(Seq(
(0L, 1L, 1.0),
(1L,2L,1.0),
(3L, 4L,1.0),
(4L,0L,0.1))).toDF("src", "dst", "weight")
val model = new PowerIterationClustering().
setMaxIter(10).
setInitMode("degree").
setWeightCol("weight")
val prediction = model.assignClusters(dataset).select("id", "cluster")
```
2) Open storage tab of the UI. We can see many RDD block cached, even after running the PIC.
In this PR, basically materializes the new graph before unpersisting the old ones.
## How was this patch tested?
Manually tested and existing UTs.
Before patch:
After patch:
Closes#24531 from shahidki31/SPARK-27636.
Authored-by: Shahid <shahidki31@gmail.com>
Signed-off-by: Sean Owen <sean.owen@databricks.com>
## What changes were proposed in this pull request?
Added method 'meanAveragePrecisionAt' k to RankingMetrics.
This branch is rebased with squashed commits from https://github.com/apache/spark/pull/24458
## How was this patch tested?
Added code in the existing test RankingMetricsSuite.
Please review http://spark.apache.org/contributing.html before opening a pull request.
Closes#24543 from qb-tarushg/SPARK-27540-REBASE.
Authored-by: qb-tarushg <tarush.grover@quantumblack.com>
Signed-off-by: Sean Owen <sean.owen@databricks.com>
…rtBasedAggregate
Normally, the aggregate operations that are invoked for an aggregation buffer for User Defined Aggregate Functions(UDAF) follow the order like initialize(), update(), eval() OR initialize(), merge(), eval(). However, after a certain threshold configurable by spark.sql.objectHashAggregate.sortBased.fallbackThreshold is reached, ObjectHashAggregate falls back to SortBasedAggregator which invokes the merge or update operation without calling initialize() on the aggregate buffer.
## What changes were proposed in this pull request?
The fix here is to initialize aggregate buffers again when fallback to SortBasedAggregate operator happens.
## How was this patch tested?
The patch was tested as part of [SPARK-24935](https://issues.apache.org/jira/browse/SPARK-24935) as documented in PR https://github.com/apache/spark/pull/23778.
Closes#24149 from pgandhi999/SPARK-27207.
Authored-by: pgandhi <pgandhi@verizonmedia.com>
Signed-off-by: Wenchen Fan <wenchen@databricks.com>
## What changes were proposed in this pull request?
### Background:
The data source option `pathGlobFilter` is introduced for Binary file format: https://github.com/apache/spark/pull/24354 , which can be used for filtering file names, e.g. reading `.png` files only while there is `.json` files in the same directory.
### Proposal:
Make the option `pathGlobFilter` as a general option for all file sources. The path filtering should happen in the path globbing on Driver.
### Motivation:
Filtering the file path names in file scan tasks on executors is kind of ugly.
### Impact:
1. The splitting of file partitions will be more balanced.
2. The metrics of file scan will be more accurate.
3. Users can use the option for reading other file sources.
## How was this patch tested?
Unit tests
Closes#24518 from gengliangwang/globFilter.
Authored-by: Gengliang Wang <gengliang.wang@databricks.com>
Signed-off-by: HyukjinKwon <gurwls223@apache.org>
## What changes were proposed in this pull request?
The mothod `populateStartOffsets` exists a inappropriate identifier `secondLatestBatchId`.
I think `secondLatestBatchId = latestBatchId - 1` and `offsetLog.get(latestBatchId - 1)` is a offset.
So I change the identifier as follows:
`secondLatestOffsets = offsetLog.get(latestBatchId - 1)`
## How was this patch tested?
Exists UT.
Closes#24550 from beliefer/fix-inappropriate-identifier.
Authored-by: gengjiaan <gengjiaan@360.cn>
Signed-off-by: Dongjoon Hyun <dhyun@apple.com>
## What changes were proposed in this pull request?
When following the example for using `spark.streams().awaitAnyTermination()`
a valid pyspark code will output the following error:
```
Traceback (most recent call last):
File "pyspark_app.py", line 182, in <module>
spark.streams().awaitAnyTermination()
TypeError: 'StreamingQueryManager' object is not callable
```
Docs URL: https://spark.apache.org/docs/latest/structured-streaming-programming-guide.html#managing-streaming-queries
This changes the documentation line to properly call the method under the StreamingQueryManager Class
https://spark.apache.org/docs/latest/api/python/pyspark.sql.html#pyspark.sql.streaming.StreamingQueryManager
## How was this patch tested?
After changing the syntax, error no longer occurs and pyspark application works
This is only docs change
Closes#24547 from asaf400/patch-1.
Authored-by: Asaf Levy <asaf400@gmail.com>
Signed-off-by: HyukjinKwon <gurwls223@apache.org>
## What changes were proposed in this pull request?
To move DS v2 to the catalyst module, we can't make v2 offset rely on v1 offset, as v1 offset is in sql/core.
## How was this patch tested?
existing tests
Closes#24538 from cloud-fan/offset.
Authored-by: Wenchen Fan <wenchen@databricks.com>
Signed-off-by: gatorsmile <gatorsmile@gmail.com>
## What changes were proposed in this pull request?
Python uses a prefetch approach to read the result from upstream and serve them in another thread, thus it's possible that if the children operator doesn't consume all the data then the Task cleanup may happen before Python side read process finishes, this in turn create a race condition that the block read locks are freed during Task cleanup and then the reader try to release the read lock it holds and find it has been released, in this case we shall hit a AssertionError.
We shall catch the AssertionError in PythonRunner and prevent this kill the Executor.
## How was this patch tested?
Hard to write a unit test case for this case, manually verified with failed job.
Closes#24542 from jiangxb1987/pyError.
Authored-by: Xingbo Jiang <xingbo.jiang@databricks.com>
Signed-off-by: HyukjinKwon <gurwls223@apache.org>
## What changes were proposed in this pull request?
For historical reasons, structured streaming still has some old way of use
`spark.network.timeout`
, even though
`org.apache.spark.internal.config.Network.NETWORK_TIMEOUT`
is now available.
## How was this patch tested?
Exists UT.
Closes#24545 from beliefer/unify-spark-network-timeout.
Authored-by: gengjiaan <gengjiaan@360.cn>
Signed-off-by: Wenchen Fan <wenchen@databricks.com>
## What changes were proposed in this pull request?
Add a legacy flag to restore the old session init behavior, where SparkConf defaults take precedence over configs in a parent session.
Closes#24540 from jose-torres/oss.
Authored-by: Jose Torres <torres.joseph.f+github@gmail.com>
Signed-off-by: Dongjoon Hyun <dhyun@apple.com>
## What changes were proposed in this pull request?
This adds the TableCatalog API proposed in the [Table Metadata API SPIP](https://docs.google.com/document/d/1zLFiA1VuaWeVxeTDXNg8bL6GP3BVoOZBkewFtEnjEoo/edit#heading=h.m45webtwxf2d).
For `TableCatalog` to use `Table`, it needed to be moved into the catalyst module where the v2 catalog API is located. This also required moving `TableCapability`. Most of the files touched by this PR are import changes needed by this move.
## How was this patch tested?
This adds a test implementation and contract tests.
Closes#24246 from rdblue/SPARK-24252-add-table-catalog-api.
Authored-by: Ryan Blue <blue@apache.org>
Signed-off-by: Wenchen Fan <wenchen@databricks.com>
## What changes were proposed in this pull request?
This fixes an error when a PySpark local iterator, for both RDD and DataFrames, goes out of scope and the connection is closed before fully consuming the iterator. The error occurs on the JVM in the serving thread, when Python closes the local socket while the JVM is writing to it. This usually happens when there is enough data to fill the socket read buffer, causing the write call to block.
Additionally, this fixes a problem when an error occurs in the Python worker and the collect job is cancelled with an exception. Previously, the Python driver was never notified of the error so the user could get a partial result (iteration until the error) and the application will continue. With this change, an error in the worker is sent to the Python iterator and is then raised.
The change here introduces a protocol for PySpark local iterators that work as follows:
1) The local socket connection is made when the iterator is created
2) When iterating, Python first sends a request for partition data as a non-zero integer
3) While the JVM local iterator over partitions has next, it triggers a job to collect the next partition
4) The JVM sends a nonzero response to indicate it has the next partition to send
5) The next partition is sent to Python and read by the PySpark deserializer
6) After sending the entire partition, an `END_OF_DATA_SECTION` is sent to Python which stops the deserializer and allows to make another request
7) When the JVM gets a request from Python but has already consumed it's local iterator, it will send a zero response to Python and both will close the socket cleanly
8) If an error occurs in the worker, a negative response is sent to Python followed by the error message. Python will then raise a RuntimeError with the message, stopping iteration.
9) When the PySpark local iterator is garbage-collected, it will read any remaining data from the current partition (this is data that has already been collected) and send a request of zero to tell the JVM to stop collection jobs and close the connection.
Steps 1, 3, 5, 6 are the same as before. Step 8 was completely missing before because errors in the worker were never communicated back to Python. The other steps add synchronization to allow for a clean closing of the socket, with a small trade-off in performance for each partition. This is mainly because the JVM does not start collecting partition data until it receives a request to do so, where before it would eagerly write all data until the socket receive buffer is full.
## How was this patch tested?
Added new unit tests for DataFrame and RDD `toLocalIterator` and tested not fully consuming the iterator. Manual tests with Python 2.7 and 3.6.
Closes#24070 from BryanCutler/pyspark-toLocalIterator-clean-stop-SPARK-23961.
Authored-by: Bryan Cutler <cutlerb@gmail.com>
Signed-off-by: Bryan Cutler <cutlerb@gmail.com>
## What changes were proposed in this pull request?
The actual implementation doesn't support multi-cluster Kafka connection with delegation token. In this PR I've added this functionality.
What this PR contains:
* New way of configuration
* Multiple delegation token obtain/store/use functionality
* Documentation
* The change works on DStreams also
## How was this patch tested?
Existing + additional unit tests.
Additionally tested on cluster.
Test scenario:
* 2 * 4 node clusters
* The 4-4 nodes are in different kerberos realms
* Cross-Realm trust between the 2 realms
* Yarn
* Kafka broker version 2.1.0
* security.protocol = SASL_SSL
* sasl.mechanism = SCRAM-SHA-512
* Artificial exceptions during processing
* Source reads from realm1 sink writes to realm2
Kafka broker settings:
* delegation.token.expiry.time.ms=600000 (10 min)
* delegation.token.max.lifetime.ms=1200000 (20 min)
* delegation.token.expiry.check.interval.ms=300000 (5 min)
Closes#24305 from gaborgsomogyi/SPARK-27294.
Authored-by: Gabor Somogyi <gabor.g.somogyi@gmail.com>
Signed-off-by: Marcelo Vanzin <vanzin@cloudera.com>
## What changes were proposed in this pull request?
If the interval is `0`, it doesn't show both the value `0` and the unit at all. For example, this happens in the explain plans and Spark Web UI on `EventTimeWatermark` diagram.
**BEFORE**
```scala
scala> spark.readStream.schema("ts timestamp").parquet("/tmp/t").withWatermark("ts", "1 microsecond").explain
== Physical Plan ==
EventTimeWatermark ts#0: timestamp, interval 1 microseconds
+- StreamingRelation FileSource[/tmp/t], [ts#0]
scala> spark.readStream.schema("ts timestamp").parquet("/tmp/t").withWatermark("ts", "0 microsecond").explain
== Physical Plan ==
EventTimeWatermark ts#3: timestamp, interval
+- StreamingRelation FileSource[/tmp/t], [ts#3]
```
**AFTER**
```scala
scala> spark.readStream.schema("ts timestamp").parquet("/tmp/t").withWatermark("ts", "1 microsecond").explain
== Physical Plan ==
EventTimeWatermark ts#0: timestamp, interval 1 microseconds
+- StreamingRelation FileSource[/tmp/t], [ts#0]
scala> spark.readStream.schema("ts timestamp").parquet("/tmp/t").withWatermark("ts", "0 microsecond").explain
== Physical Plan ==
EventTimeWatermark ts#3: timestamp, interval 0 microseconds
+- StreamingRelation FileSource[/tmp/t], [ts#3]
```
## How was this patch tested?
Pass the Jenkins with the updated test case.
Closes#24516 from dongjoon-hyun/SPARK-27624.
Authored-by: Dongjoon Hyun <dhyun@apple.com>
Signed-off-by: Dongjoon Hyun <dhyun@apple.com>
## What changes were proposed in this pull request?
Fixed the `spark-<version>-yarn-shuffle.jar` artifact packaging to shade the native netty libraries:
- shade the `META-INF/native/libnetty_*` native libraries when packagin
the yarn shuffle service jar. This is required as netty library loader
derives that based on shaded package name.
- updated the `org/spark_project` shade package prefix to `org/sparkproject`
(i.e. removed underscore) as the former breaks the netty native lib loading.
This was causing the yarn external shuffle service to fail
when spark.shuffle.io.mode=EPOLL
## How was this patch tested?
Manual tests
Closes#24502 from amuraru/SPARK-27610_master.
Authored-by: Adi Muraru <amuraru@adobe.com>
Signed-off-by: Marcelo Vanzin <vanzin@cloudera.com>
## What changes were proposed in this pull request?
This is a followup of https://github.com/apache/spark/pull/24375
When `TaskSetManager` skips a task because its corresponding partition is already completed by other `TaskSetManager`s, we should not consider the duration of the task that is finished by other `TaskSetManager`s to schedule the speculative tasks of this `TaskSetManager`.
## How was this patch tested?
updated test case
Closes#24485 from cloud-fan/minor.
Authored-by: Wenchen Fan <wenchen@databricks.com>
Signed-off-by: Imran Rashid <irashid@cloudera.com>
## What changes were proposed in this pull request?
Choose the last record in chunks when calculating metrics with downsampling in `BinaryClassificationMetrics`.
## How was this patch tested?
A new unit test is added to verify thresholds from downsampled records.
Closes#24470 from shishaochen/spark-mllib-binary-metrics.
Authored-by: Shaochen Shi <shishaochen@bytedance.com>
Signed-off-by: Sean Owen <sean.owen@databricks.com>
This is PR is meant to replace #20503, which lay dormant for a while. The solution in the original PR is still valid, so this is just that patch rebased onto the current master.
Original summary follows.
## What changes were proposed in this pull request?
Fix `__repr__` behaviour for Rows.
Rows `__repr__` assumes data is a string when column name is missing.
Examples,
```
>>> from pyspark.sql.types import Row
>>> Row ("Alice", "11")
<Row(Alice, 11)>
>>> Row (name="Alice", age=11)
Row(age=11, name='Alice')
>>> Row ("Alice", 11)
<snip stack trace>
TypeError: sequence item 1: expected string, int found
```
This is because Row () when called without column names assumes everything is a string.
## How was this patch tested?
Manually tested and a unit test was added to `python/pyspark/sql/tests/test_types.py`.
Closes#24448 from tbcs/SPARK-23299.
Lead-authored-by: Tibor Csögör <tibi@tiborius.net>
Co-authored-by: Shashwat Anand <me@shashwat.me>
Signed-off-by: Holden Karau <holden@pigscanfly.ca>
## What changes were proposed in this pull request?
`BaseStreamingSource` and `BaseStreamingSink` is used to unify v1 and v2 streaming data source API in some code paths.
This PR removes these 2 interfaces, and let the v1 API extend v2 API to keep API compatibility.
The motivation is https://github.com/apache/spark/pull/24416 . We want to move data source v2 to catalyst module, but `BaseStreamingSource` and `BaseStreamingSink` are in sql/core.
## How was this patch tested?
existing tests
Closes#24471 from cloud-fan/streaming.
Authored-by: Wenchen Fan <wenchen@databricks.com>
Signed-off-by: Wenchen Fan <wenchen@databricks.com>
## What changes were proposed in this pull request?
Because a review is resolved during analysis when we create a dataset, the content of the view is determined when the dataset is created, not when it is evaluated. Now the explain result of a dataset is not correctly consistent with the collected result of it, because we use pre-analyzed logical plan of the dataset in explain command. The explain command will analyzed the logical plan passed in. So if a view is changed after the dataset was created, the plans shown by explain command aren't the same with the plan of the dataset.
```scala
scala> spark.range(10).createOrReplaceTempView("test")
scala> spark.range(5).createOrReplaceTempView("test2")
scala> spark.sql("select * from test").createOrReplaceTempView("tmp001")
scala> val df = spark.sql("select * from tmp001")
scala> spark.sql("select * from test2").createOrReplaceTempView("tmp001")
scala> df.show
+---+
| id|
+---+
| 0|
| 1|
| 2|
| 3|
| 4|
| 5|
| 6|
| 7|
| 8|
| 9|
+---+
scala> df.explain(true)
```
Before:
```scala
== Parsed Logical Plan ==
'Project [*]
+- 'UnresolvedRelation `tmp001`
== Analyzed Logical Plan ==
id: bigint
Project [id#2L]
+- SubqueryAlias `tmp001`
+- Project [id#2L]
+- SubqueryAlias `test2`
+- Range (0, 5, step=1, splits=Some(12))
== Optimized Logical Plan ==
Range (0, 5, step=1, splits=Some(12))
== Physical Plan ==
*(1) Range (0, 5, step=1, splits=12)
```
After:
```scala
== Parsed Logical Plan ==
'Project [*]
+- 'UnresolvedRelation `tmp001`
== Analyzed Logical Plan ==
id: bigint
Project [id#0L]
+- SubqueryAlias `tmp001`
+- Project [id#0L]
+- SubqueryAlias `test`
+- Range (0, 10, step=1, splits=Some(12))
== Optimized Logical Plan ==
Range (0, 10, step=1, splits=Some(12))
== Physical Plan ==
*(1) Range (0, 10, step=1, splits=12)
```
To fix it, this passes query execution of Dataset when explaining it. The query execution contains pre-analyzed plan which is consistent with Dataset's result.
## How was this patch tested?
Manually test and unit test.
Closes#24464 from viirya/SPARK-27439-2.
Authored-by: Liang-Chi Hsieh <viirya@gmail.com>
Signed-off-by: Dongjoon Hyun <dhyun@apple.com>
## What changes were proposed in this pull request?
**Description from JIRA**
For the JDBC option `query`, we use the identifier name to start with underscore: s"(${subquery}) _SPARK_GEN_JDBC_SUBQUERY_NAME${curId.getAndIncrement()}". This is not supported by Oracle.
The Oracle doesn't seem to support identifier name to start with non-alphabet character (unless it is quoted) and has length restrictions as well. [link](https://docs.oracle.com/cd/B19306_01/server.102/b14200/sql_elements008.htm)
In this PR, the generated alias name 'SPARK_GEN_JDBC_SUBQUERY_NAME<int value>' is fixed to remove "_" prefix and also the alias name is shortened to not exceed the identifier length limit.
## How was this patch tested?
Tests are added for MySql, Postgress, Oracle and DB2 to ensure enough coverage.
Closes#24532 from dilipbiswal/SPARK-27596.
Authored-by: Dilip Biswal <dbiswal@us.ibm.com>
Signed-off-by: gatorsmile <gatorsmile@gmail.com>
## What changes were proposed in this pull request?
In SPARK-27612, one correctness issue was reported. When protocol 4 is used to pickle Python objects, we found that unpickled objects were wrong. A temporary fix was proposed by not using highest protocol.
It was found that Opcodes.MEMOIZE was appeared in the opcodes in protocol 4. It is suspect to this issue.
A deeper dive found that Opcodes.MEMOIZE stores objects into internal map of Unpickler object. We use single Unpickler object to unpickle serialized Python bytes. Stored objects intervenes next round of unpickling, if the map is not cleared.
We has two options:
1. Continues to reuse Unpickler, but calls its close after each unpickling.
2. Not to reuse Unpickler and create new Unpickler object in each unpickling.
This patch takes option 1.
## How was this patch tested?
Passing the test added in SPARK-27612 (#24519).
Closes#24521 from viirya/SPARK-27629.
Authored-by: Liang-Chi Hsieh <viirya@gmail.com>
Signed-off-by: HyukjinKwon <gurwls223@apache.org>
## What changes were proposed in this pull request?
`date_trunc` argument order was flipped, phrasing was awkward.
## How was this patch tested?
Documentation-only.
Closes#24522 from mojodna/patch-2.
Authored-by: Seth Fitzsimmons <seth@mojodna.net>
Signed-off-by: HyukjinKwon <gurwls223@apache.org>
## What changes were proposed in this pull request?
SQLConf does not load hive-site.xml.So HiveSerDe should fall back to hadoopconf if hive.default.fileformat is not found in SQLConf
## How was this patch tested?
Tested manually.
Added UT
Closes#24489 from sandeep-katta/spark-27555.
Authored-by: sandeep katta <sandeep.katta2007@gmail.com>
Signed-off-by: HyukjinKwon <gurwls223@apache.org>
## What changes were proposed in this pull request?
When transform(...) method is called on a LinearRegressionModel created directly with the coefficients and intercepts, the following exception is encountered.
```
java.util.NoSuchElementException: Failed to find a default value for loss
at org.apache.spark.ml.param.Params$$anonfun$getOrDefault$2.apply(params.scala:780)
at org.apache.spark.ml.param.Params$$anonfun$getOrDefault$2.apply(params.scala:780)
at scala.Option.getOrElse(Option.scala:121)
at org.apache.spark.ml.param.Params$class.getOrDefault(params.scala:779)
at org.apache.spark.ml.PipelineStage.getOrDefault(Pipeline.scala:42)
at org.apache.spark.ml.param.Params$class.$(params.scala:786)
at org.apache.spark.ml.PipelineStage.$(Pipeline.scala:42)
at org.apache.spark.ml.regression.LinearRegressionParams$class.validateAndTransformSchema(LinearRegression.scala:111)
at org.apache.spark.ml.regression.LinearRegressionModel.validateAndTransformSchema(LinearRegression.scala:637)
at org.apache.spark.ml.PredictionModel.transformSchema(Predictor.scala:192)
at org.apache.spark.ml.PipelineModel$$anonfun$transformSchema$5.apply(Pipeline.scala:311)
at org.apache.spark.ml.PipelineModel$$anonfun$transformSchema$5.apply(Pipeline.scala:311)
at scala.collection.IndexedSeqOptimized$class.foldl(IndexedSeqOptimized.scala:57)
at scala.collection.IndexedSeqOptimized$class.foldLeft(IndexedSeqOptimized.scala:66)
at scala.collection.mutable.ArrayOps$ofRef.foldLeft(ArrayOps.scala:186)
at org.apache.spark.ml.PipelineModel.transformSchema(Pipeline.scala:311)
at org.apache.spark.ml.PipelineStage.transformSchema(Pipeline.scala:74)
at org.apache.spark.ml.PipelineModel.transform(Pipeline.scala:305)
```
This is because validateAndTransformSchema() is called both during training and scoring phases, but the checks against the training related params like loss should really be performed during training phase only, I think, please correct me if I'm missing anything :)
This issue was first reported for mleap (https://github.com/combust/mleap/issues/455) because basically when we serialize the Spark transformers for mleap, we only serialize the params that are relevant for scoring. We do have the option to de-serialize the serialized transformers back into Spark for scoring again, but in that case, we no longer have all the training params.
## How was this patch tested?
Added a unit test to check this scenario.
Please let me know if there's anything additional required, this is the first PR that I've raised in this project.
Closes#24509 from ancasarb/linear_regression_params_fix.
Authored-by: asarb <asarb@expedia.com>
Signed-off-by: Sean Owen <sean.owen@databricks.com>
## What changes were proposed in this pull request?
This is a long standing issue which I met before and I've seen other people got trouble with it:
[test cases stuck on "local-cluster mode" of ReplSuite?](http://apache-spark-developers-list.1001551.n3.nabble.com/test-cases-stuck-on-quot-local-cluster-mode-quot-of-ReplSuite-td3086.html)
[Spark tests hang on local machine due to "testGuavaOptional" in JavaAPISuite](http://apache-spark-developers-list.1001551.n3.nabble.com/Spark-tests-hang-on-local-machine-due-to-quot-testGuavaOptional-quot-in-JavaAPISuite-tc10999.html)
When running test under local-cluster mode with wrong SPARK_HOME(spark.test.home), test just get stuck and no response forever. After looking into SPARK_WORKER_DIR, I found there's endless executor directories under it. So, this explains what happens during test getting stuck.
The whole process looks like:
1. Driver submits an app to Master and asks for N executors
2. Master inits executor state with LAUNCHING and sends `LaunchExecutor` to Worker
3. Worker receives `LaunchExecutor`, launches ExecutorRunner asynchronously and sends `ExecutorStateChanged(state=RUNNING)` to Mater immediately
4. Master receives `ExecutorStateChanged(state=RUNNING)` and reset `_retyCount` to 0.
5. ExecutorRunner throws exception during executor launching, sends `ExecutorStateChanged(state=FAILED)` to Worker, Worker forwards the msg to Master
6. Master receives `ExecutorStateChanged(state=FAILED)`. Since Master always reset `_retyCount` when it receives RUNNING msg, so, event if a Worker fails to launch executor for continuous many times, ` _retryCount` would never exceed `maxExecutorRetries`. So, Master continue to launch executor and fall into the dead loop.
The problem exists in step 3. Worker sends `ExecutorStateChanged(state=RUNNING)` to Master immediately while executor is still launching. And, when Master receive that msg, it believes the executor has launched successfully, and reset `_retryCount` subsequently. However, that's not true.
This pr suggests to remove step 3 and requires Worker only send `ExecutorStateChanged(state=RUNNING)` after executor has really launched successfully.
## How was this patch tested?
Tested Manually.
Closes#24408 from Ngone51/fix-dead-loop.
Authored-by: wuyi <ngone_5451@163.com>
Signed-off-by: Xingbo Jiang <xingbo.jiang@databricks.com>
## What changes were proposed in this pull request?
Although we use shebang `#!/usr/bin/env bash`, `minikube docker-env` returns invalid commands in `non-bash` environment and causes failures at `eval` because it only recognizes the default shell. We had better add `--shell bash` option explicitly in our `bash` script.
```bash
$ bash -c 'eval $(minikube docker-env)'
bash: line 0: set: -g: invalid option
set: usage: set [-abefhkmnptuvxBCHP] [-o option-name] [--] [arg ...]
bash: line 0: set: -g: invalid option
set: usage: set [-abefhkmnptuvxBCHP] [-o option-name] [--] [arg ...]
bash: line 0: set: -g: invalid option
set: usage: set [-abefhkmnptuvxBCHP] [-o option-name] [--] [arg ...]
bash: line 0: set: -g: invalid option
set: usage: set [-abefhkmnptuvxBCHP] [-o option-name] [--] [arg ...]
$ bash -c 'eval $(minikube docker-env --shell bash)'
```
## How was this patch tested?
Manual. Run the script with non-bash shell environment.
```
bin/docker-image-tool.sh -m -t testing build
```
Closes#24517 from dongjoon-hyun/SPARK-27626.
Authored-by: Dongjoon Hyun <dhyun@apple.com>
Signed-off-by: Dongjoon Hyun <dhyun@apple.com>
## What changes were proposed in this pull request?
This pr add missing options for `sql-migration-guide-hive-compatibility.md`.
## How was this patch tested?
N/A
Closes#24520 from wangyum/SPARK-24360.
Authored-by: Yuming Wang <yumwang@ebay.com>
Signed-off-by: gatorsmile <gatorsmile@gmail.com>
## What changes were proposed in this pull request?
This PR partially reverts https://github.com/apache/spark/pull/20691
After we changed the Python protocol to highest ones, seems like it introduced a correctness bug. This potentially affects all Python related code paths.
I suspect a bug related to Pryolite (maybe opcodes `MEMOIZE`, `FRAME` and/or our `RowPickler`). I would like to stick to default protocol for now and investigate the issue separately.
I will separately investigate later to bring highest protocol back.
## How was this patch tested?
Unittest was added.
```bash
./run-tests --python-executables=python3.7 --testname "pyspark.sql.tests.test_serde SerdeTests.test_int_array_serialization"
```
Closes#24519 from HyukjinKwon/SPARK-27612.
Authored-by: HyukjinKwon <gurwls223@apache.org>
Signed-off-by: HyukjinKwon <gurwls223@apache.org>
## What changes were proposed in this pull request?
This PR replaces for-comprehension if statement with while loop to gain better performance in `TypeUtils.compareBinary`.
## How was this patch tested?
Add UT to test old version and new version comparison result
Closes#24494 from woudygao/opt_binary_compare.
Authored-by: gaoweikang <gaoweikang@bytedance.com>
Signed-off-by: Dongjoon Hyun <dhyun@apple.com>
## What changes were proposed in this pull request?
This PR aims to upgrade Maven to 3.6.1 to bring JDK9+ related patches like [MNG-6506](https://issues.apache.org/jira/browse/MNG-6506). For the full release note, please see the following.
- https://maven.apache.org/docs/3.6.1/release-notes.html
This was committed and reverted due to AppVeyor failure. It turns out that the root cause is `PATH` issue. With the updated AppVeyor script, it passed.
https://ci.appveyor.com/project/ApacheSoftwareFoundation/spark/builds/24273412
## How was this patch tested?
Pass the Jenkins and AppVoyer
Closes#24481 from dongjoon-hyun/SPARK-R.
Lead-authored-by: Dongjoon Hyun <dhyun@apple.com>
Co-authored-by: Yuming Wang <yumwang@ebay.com>
Signed-off-by: Dongjoon Hyun <dhyun@apple.com>
