forked from apache/spark
-
Notifications
You must be signed in to change notification settings - Fork 0
/
SparkBuild.scala
1764 lines (1564 loc) · 75.1 KB
/
SparkBuild.scala
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import java.io._
import java.nio.charset.StandardCharsets.UTF_8
import java.nio.file.{Files, Paths, StandardCopyOption}
import java.util.Locale
import scala.io.Source
import scala.util.Properties
import scala.collection.mutable.ListBuffer
import scala.jdk.CollectionConverters._
import sbt._
import sbt.Classpaths.publishOrSkip
import sbt.Keys._
import sbt.librarymanagement.{ VersionNumber, SemanticSelector }
import com.etsy.sbt.checkstyle.CheckstylePlugin.autoImport._
import com.simplytyped.Antlr4Plugin._
import sbtpomreader.{PomBuild, SbtPomKeys}
import com.typesafe.tools.mima.plugin.MimaKeys
import org.scalastyle.sbt.ScalastylePlugin.autoImport._
import org.scalastyle.sbt.Tasks
import sbtassembly.AssemblyPlugin.autoImport._
import spray.revolver.RevolverPlugin._
import sbtprotoc.ProtocPlugin.autoImport._
object BuildCommons {
private val buildLocation = file(".").getAbsoluteFile.getParentFile
val sqlProjects@Seq(sqlApi, catalyst, sql, hive, hiveThriftServer, tokenProviderKafka010, sqlKafka010, avro, protobuf) =
Seq("sql-api", "catalyst", "sql", "hive", "hive-thriftserver", "token-provider-kafka-0-10",
"sql-kafka-0-10", "avro", "protobuf").map(ProjectRef(buildLocation, _))
val streamingProjects@Seq(streaming, streamingKafka010) =
Seq("streaming", "streaming-kafka-0-10").map(ProjectRef(buildLocation, _))
val connectProjects@Seq(connectCommon, connect, connectClient, connectShims) =
Seq("connect-common", "connect", "connect-client-jvm", "connect-shims")
.map(ProjectRef(buildLocation, _))
val allProjects@Seq(
core, graphx, mllib, mllibLocal, repl, networkCommon, networkShuffle, launcher, unsafe, tags, sketch, kvstore,
commonUtils, variant, _*
) = Seq(
"core", "graphx", "mllib", "mllib-local", "repl", "network-common", "network-shuffle", "launcher", "unsafe",
"tags", "sketch", "kvstore", "common-utils", "variant"
).map(ProjectRef(buildLocation, _)) ++ sqlProjects ++ streamingProjects ++ connectProjects
val optionallyEnabledProjects@Seq(kubernetes, yarn,
sparkGangliaLgpl, streamingKinesisAsl,
dockerIntegrationTests, hadoopCloud, kubernetesIntegrationTests) =
Seq("kubernetes", "yarn",
"ganglia-lgpl", "streaming-kinesis-asl",
"docker-integration-tests", "hadoop-cloud", "kubernetes-integration-tests").map(ProjectRef(buildLocation, _))
val assemblyProjects@Seq(networkYarn, streamingKafka010Assembly, streamingKinesisAslAssembly) =
Seq("network-yarn", "streaming-kafka-0-10-assembly", "streaming-kinesis-asl-assembly")
.map(ProjectRef(buildLocation, _))
val copyJarsProjects@Seq(assembly, examples) = Seq("assembly", "examples")
.map(ProjectRef(buildLocation, _))
val tools = ProjectRef(buildLocation, "tools")
// Root project.
val spark = ProjectRef(buildLocation, "spark")
val sparkHome = buildLocation
val testTempDir = s"$sparkHome/target/tmp"
val javaVersion = settingKey[String]("source and target JVM version for javac and scalac")
// Google Protobuf version used for generating the protobuf.
// SPARK-41247: needs to be consistent with `protobuf.version` in `pom.xml`.
val protoVersion = "4.29.1"
// GRPC version used for Spark Connect.
val grpcVersion = "1.67.1"
}
object SparkBuild extends PomBuild {
import BuildCommons._
import sbtunidoc.GenJavadocPlugin
import sbtunidoc.GenJavadocPlugin.autoImport._
import scala.collection.mutable.Map
val projectsMap: Map[String, Seq[Setting[_]]] = Map.empty
override val profiles = {
val profiles = Properties.envOrNone("SBT_MAVEN_PROFILES")
.orElse(Properties.propOrNone("sbt.maven.profiles")) match {
case None => Seq("sbt")
case Some(v) =>
v.split("(\\s+|,)").filterNot(_.isEmpty).map(_.trim.replaceAll("-P", "")).toSeq
}
if (profiles.contains("jdwp-test-debug")) {
sys.props.put("test.jdwp.enabled", "true")
}
if (profiles.contains("user-defined-protoc")) {
val sparkProtocExecPath = Properties.envOrNone("SPARK_PROTOC_EXEC_PATH")
val connectPluginExecPath = Properties.envOrNone("CONNECT_PLUGIN_EXEC_PATH")
if (sparkProtocExecPath.isDefined) {
sys.props.put("spark.protoc.executable.path", sparkProtocExecPath.get)
}
if (connectPluginExecPath.isDefined) {
sys.props.put("connect.plugin.executable.path", connectPluginExecPath.get)
}
}
profiles
}
Properties.envOrNone("SBT_MAVEN_PROPERTIES") match {
case Some(v) =>
v.split("(\\s+|,)").filterNot(_.isEmpty).map(_.split("=")).foreach(x => System.setProperty(x(0), x(1)))
case _ =>
}
override val userPropertiesMap = System.getProperties.asScala.toMap
lazy val MavenCompile = config("m2r") extend(Compile)
lazy val SbtCompile = config("sbt") extend(Compile)
lazy val sparkGenjavadocSettings: Seq[sbt.Def.Setting[_]] = GenJavadocPlugin.projectSettings ++ Seq(
scalacOptions ++= Seq(
"-P:genjavadoc:strictVisibility=true" // hide package private types
)
)
lazy val scalaStyleRules = Project("scalaStyleRules", file("scalastyle"))
.settings(
libraryDependencies += "org.scalastyle" %% "scalastyle" % "1.0.0"
)
lazy val scalaStyleOnCompile = taskKey[Unit]("scalaStyleOnCompile")
lazy val scalaStyleOnTest = taskKey[Unit]("scalaStyleOnTest")
// We special case the 'println' lint rule to only be a warning on compile, because adding
// printlns for debugging is a common use case and is easy to remember to remove.
val scalaStyleOnCompileConfig: String = {
val in = "scalastyle-config.xml"
val out = "scalastyle-on-compile.generated.xml"
val replacements = Map(
"""customId="println" level="error"""" -> """customId="println" level="warn""""
)
val source = Source.fromFile(in)
try {
var contents = source.getLines.mkString("\n")
for ((k, v) <- replacements) {
require(contents.contains(k), s"Could not rewrite '$k' in original scalastyle config.")
contents = contents.replace(k, v)
}
new PrintWriter(out) {
write(contents)
close()
}
out
} finally {
source.close()
}
}
// Return a cached scalastyle task for a given configuration (usually Compile or Test)
private def cachedScalaStyle(config: Configuration) = Def.task {
val logger = streams.value.log
// We need a different cache dir per Configuration, otherwise they collide
val cacheDir = target.value / s"scalastyle-cache-${config.name}"
val cachedFun = FileFunction.cached(cacheDir, FilesInfo.lastModified, FilesInfo.exists) {
(inFiles: Set[File]) => {
val args: Seq[String] = Seq.empty
val scalaSourceV = Seq(file((config / scalaSource).value.getAbsolutePath))
val configV = (ThisBuild / baseDirectory).value / scalaStyleOnCompileConfig
val configUrlV = (config / scalastyleConfigUrl).value
val streamsV = ((config / streams).value: @sbtUnchecked)
val failOnErrorV = true
val failOnWarningV = false
val scalastyleTargetV = (config / scalastyleTarget).value
val configRefreshHoursV = (config / scalastyleConfigRefreshHours).value
val targetV = (config / target).value
val configCacheFileV = (config / scalastyleConfigUrlCacheFile).value
logger.info(s"Running scalastyle on ${name.value} in ${config.name}")
Tasks.doScalastyle(args, configV, configUrlV, failOnErrorV, failOnWarningV, scalaSourceV,
scalastyleTargetV, streamsV, configRefreshHoursV, targetV, configCacheFileV)
Set.empty
}
}
cachedFun(findFiles((config / scalaSource).value))
}
private def findFiles(file: File): Set[File] = if (file.isDirectory) {
file.listFiles().toSet.flatMap(findFiles) + file
} else {
Set(file)
}
def enableScalaStyle: Seq[sbt.Def.Setting[_]] = Seq(
scalaStyleOnCompile := cachedScalaStyle(Compile).value,
scalaStyleOnTest := cachedScalaStyle(Test).value,
(scalaStyleOnCompile / logLevel) := Level.Warn,
(scalaStyleOnTest / logLevel) := Level.Warn,
(Compile / compile) := {
scalaStyleOnCompile.value
(Compile / compile).value
},
(Test / compile) := {
scalaStyleOnTest.value
(Test / compile).value
}
)
lazy val compilerWarningSettings: Seq[sbt.Def.Setting[_]] = Seq(
(Compile / scalacOptions) ++= {
Seq(
// replace -Xfatal-warnings with fine-grained configuration, since 2.13.2
// verbose warning on deprecation, error on all others
// see `scalac -Wconf:help` for details
// since 2.13.15, "-Wconf:cat=deprecation:wv,any:e" no longer takes effect and needs to
// be changed to "-Wconf:any:e", "-Wconf:cat=deprecation:wv",
// please refer to the details: https://github.com/scala/scala/pull/10708
"-Wconf:any:e",
"-Wconf:cat=deprecation:wv",
// 2.13-specific warning hits to be muted (as narrowly as possible) and addressed separately
"-Wunused:imports",
"-Wconf:msg=^(?=.*?method|value|type|object|trait|inheritance)(?=.*?deprecated)(?=.*?since 2.13).+$:e",
"-Wconf:msg=^(?=.*?Widening conversion from)(?=.*?is deprecated because it loses precision).+$:e",
// SPARK-45610 Convert "Auto-application to `()` is deprecated" to compile error, as it will become a compile error in Scala 3.
"-Wconf:cat=deprecation&msg=Auto-application to \\`\\(\\)\\` is deprecated:e",
// SPARK-35574 Prevent the recurrence of compilation warnings related to `procedure syntax is deprecated`
"-Wconf:cat=deprecation&msg=procedure syntax is deprecated:e",
// SPARK-45627 Symbol literals are deprecated in Scala 2.13 and it's a compile error in Scala 3.
"-Wconf:cat=deprecation&msg=symbol literal is deprecated:e",
// SPARK-45627 `enum`, `export` and `given` will become keywords in Scala 3,
// so they are prohibited from being used as variable names in Scala 2.13 to
// reduce the cost of migration in subsequent versions.
"-Wconf:cat=deprecation&msg=it will become a keyword in Scala 3:e",
// SPARK-46938 to prevent enum scan on pmml-model, under spark-mllib module.
"-Wconf:cat=other&site=org.dmg.pmml.*:w",
// SPARK-49937 ban call the method `SparkThrowable#getErrorClass`
"-Wconf:cat=deprecation&msg=method getErrorClass in trait SparkThrowable is deprecated:e"
)
}
)
val noLintOnCompile = sys.env.contains("NOLINT_ON_COMPILE") &&
!sys.env.get("NOLINT_ON_COMPILE").contains("false")
lazy val sharedSettings = sparkGenjavadocSettings ++
compilerWarningSettings ++
(if (noLintOnCompile) Nil else enableScalaStyle) ++ Seq(
(Compile / exportJars) := true,
(Test / exportJars) := false,
javaHome := sys.env.get("JAVA_HOME")
.orElse(sys.props.get("java.home").map { p => new File(p).getParentFile().getAbsolutePath() })
.map(file),
publishMavenStyle := true,
unidocGenjavadocVersion := "0.19",
// Override SBT's default resolvers:
resolvers := Seq(
// Google Mirror of Maven Central, placed first so that it's used instead of flaky Maven Central.
// See https://storage-download.googleapis.com/maven-central/index.html for more info.
"gcs-maven-central-mirror" at "https://maven-central.storage-download.googleapis.com/maven2/",
DefaultMavenRepository,
Resolver.mavenLocal,
Resolver.file("ivyLocal", file(Path.userHome.absolutePath + "/.ivy2/local"))(Resolver.ivyStylePatterns)
),
externalResolvers := resolvers.value,
otherResolvers := SbtPomKeys.mvnLocalRepository(dotM2 => Seq(Resolver.file("dotM2", dotM2))).value,
(MavenCompile / publishLocalConfiguration) := PublishConfiguration()
.withResolverName("dotM2")
.withArtifacts(packagedArtifacts.value.toVector)
.withLogging(ivyLoggingLevel.value),
(SbtCompile / publishLocalConfiguration) := PublishConfiguration()
.withResolverName("ivyLocal")
.withArtifacts(packagedArtifacts.value.toVector)
.withLogging(ivyLoggingLevel.value),
(MavenCompile / publishMavenStyle) := true,
(SbtCompile / publishMavenStyle) := false,
(MavenCompile / publishLocal) := publishOrSkip((MavenCompile / publishLocalConfiguration),
(publishLocal / skip)).value,
(SbtCompile / publishLocal) := publishOrSkip((SbtCompile / publishLocalConfiguration),
(publishLocal / skip)).value,
publishLocal := Seq((MavenCompile / publishLocal), (SbtCompile / publishLocal)).dependOn.value,
javaOptions ++= {
// for `dev.ludovic.netlib.blas` which implements such hardware-accelerated BLAS operations
Seq("--add-modules=jdk.incubator.vector")
},
(Compile / doc / javacOptions) ++= {
Seq("-Xdoclint:all", "-Xdoclint:-missing")
},
javaVersion := SbtPomKeys.effectivePom.value.getProperties.get("java.version").asInstanceOf[String],
(Compile / javacOptions) ++= Seq(
"-encoding", UTF_8.name(),
"-g",
"--release", javaVersion.value
),
// This -target and Xlint:unchecked options cannot be set in the Compile configuration scope since
// `javadoc` doesn't play nicely with them; see https://github.com/sbt/sbt/issues/355#issuecomment-3817629
// for additional discussion and explanation.
(Compile / compile / javacOptions) ++= Seq(
"-Xlint:unchecked"
),
(Compile / scalacOptions) ++= Seq(
"-release", javaVersion.value,
"-sourcepath", (ThisBuild / baseDirectory).value.getAbsolutePath // Required for relative source links in scaladoc
),
SbtPomKeys.profiles := profiles,
// Remove certain packages from Scaladoc
(Compile / doc / scalacOptions) := Seq(
"-groups",
"-skip-packages", Seq(
"org.apache.spark.api.python",
"org.apache.spark.network",
"org.apache.spark.deploy",
"org.apache.spark.util.collection",
"org.apache.spark.sql.scripting"
).mkString(":"),
"-doc-title", "Spark " + version.value.replaceAll("-SNAPSHOT", "") + " ScalaDoc"
),
// disable Mima check for all modules,
// to be enabled in specific ones that have previous artifacts
MimaKeys.mimaFailOnNoPrevious := false,
// Setting version for the protobuf compiler. This has to be propagated to every sub-project
// even if the project is not using it.
PB.protocVersion := protoVersion,
)
def enable(settings: Seq[Setting[_]])(projectRef: ProjectRef) = {
val existingSettings = projectsMap.getOrElse(projectRef.project, Seq[Setting[_]]())
projectsMap += (projectRef.project -> (existingSettings ++ settings))
}
// Note ordering of these settings matter.
/* Enable shared settings on all projects */
(allProjects ++ optionallyEnabledProjects ++ assemblyProjects ++ copyJarsProjects ++ Seq(spark, tools))
.foreach(enable(sharedSettings ++ DependencyOverrides.settings ++
ExcludedDependencies.settings ++ Checkstyle.settings ++ ExcludeShims.settings))
/* Enable tests settings for all projects except examples, assembly and tools */
(allProjects ++ optionallyEnabledProjects).foreach(enable(TestSettings.settings))
val mimaProjects = allProjects.filterNot { x =>
Seq(
spark, hive, hiveThriftServer, repl, networkCommon, networkShuffle, networkYarn,
unsafe, tags, tokenProviderKafka010, sqlKafka010, connectCommon, connect, connectClient,
variant, connectShims
).contains(x)
}
mimaProjects.foreach { x =>
enable(MimaBuild.mimaSettings(sparkHome, x))(x)
}
/* Generate and pick the spark build info from extra-resources */
enable(CommonUtils.settings)(commonUtils)
enable(Core.settings)(core)
/* Unsafe settings */
enable(Unsafe.settings)(unsafe)
/*
* Set up tasks to copy dependencies during packaging. This step can be disabled in the command
* line, so that dev/mima can run without trying to copy these files again and potentially
* causing issues.
*/
if (!"false".equals(System.getProperty("copyDependencies"))) {
copyJarsProjects.foreach(enable(CopyDependencies.settings))
}
/* Enable Assembly for all assembly projects */
assemblyProjects.foreach(enable(Assembly.settings))
/* Package pyspark artifacts in a separate zip file for YARN. */
enable(PySparkAssembly.settings)(assembly)
/* Enable unidoc only for the root spark project */
enable(SparkUnidoc.settings)(spark)
/* Enable unidoc only for the root spark connect client project */
// enable(SparkConnectClientUnidoc.settings)(connectClient)
/* Sql-api ANTLR generation settings */
enable(SqlApi.settings)(sqlApi)
/* Spark SQL Core settings */
enable(SQL.settings)(sql)
/* Hive console settings */
enable(Hive.settings)(hive)
enable(SparkConnectCommon.settings)(connectCommon)
enable(SparkConnect.settings)(connect)
enable(SparkConnectClient.settings)(connectClient)
/* Protobuf settings */
enable(SparkProtobuf.settings)(protobuf)
enable(DockerIntegrationTests.settings)(dockerIntegrationTests)
enable(KubernetesIntegrationTests.settings)(kubernetesIntegrationTests)
enable(YARN.settings)(yarn)
if (profiles.contains("sparkr")) {
enable(SparkR.settings)(core)
}
/**
* Adds the ability to run the spark shell directly from SBT without building an assembly
* jar.
*
* Usage: `build/sbt sparkShell`
*/
val sparkShell = taskKey[Unit]("start a spark-shell.")
val sparkPackage = inputKey[Unit](
s"""
|Download and run a spark package.
|Usage `builds/sbt "sparkPackage <group:artifact:version> <MainClass> [args]
""".stripMargin)
val sparkSql = taskKey[Unit]("starts the spark sql CLI.")
enable(Seq(
(run / connectInput) := true,
fork := true,
(run / outputStrategy) := Some (StdoutOutput),
javaOptions += "-Xmx2g",
sparkShell := {
(Compile / runMain).toTask(" org.apache.spark.repl.Main -usejavacp").value
},
sparkPackage := {
import complete.DefaultParsers._
val packages :: className :: otherArgs = spaceDelimited("<group:artifact:version> <MainClass> [args]").parsed.toList
val scalaRun = (run / runner).value
val classpath = (Runtime / fullClasspath).value
val args = Seq("--packages", packages, "--class", className, (LocalProject("core") / Compile / Keys.`package`)
.value.getCanonicalPath) ++ otherArgs
println(args)
scalaRun.run("org.apache.spark.deploy.SparkSubmit", classpath.map(_.data), args, streams.value.log)
},
(Compile / javaOptions) += "-Dspark.master=local",
sparkSql := {
(Compile / runMain).toTask(" org.apache.spark.sql.hive.thriftserver.SparkSQLCLIDriver").value
}
))(assembly)
enable(Seq(sparkShell := (LocalProject("assembly") / sparkShell).value))(spark)
// TODO: move this to its upstream project.
override def projectDefinitions(baseDirectory: File): Seq[Project] = {
super.projectDefinitions(baseDirectory).map { x =>
if (projectsMap.exists(_._1 == x.id)) x.settings(projectsMap(x.id): _*)
else x.settings(Seq[Setting[_]](): _*)
} ++ Seq[Project](OldDeps.project)
}
if (!sys.env.contains("SERIAL_SBT_TESTS")) {
allProjects.foreach(enable(SparkParallelTestGrouping.settings))
}
}
object SparkParallelTestGrouping {
// Settings for parallelizing tests. The basic strategy here is to run the slowest suites (or
// collections of suites) in their own forked JVMs, allowing us to gain parallelism within a
// SBT project. Here, we take an opt-in approach where the default behavior is to run all
// tests sequentially in a single JVM, requiring us to manually opt-in to the extra parallelism.
//
// There are a reasons why such an opt-in approach is good:
//
// 1. Launching one JVM per suite adds significant overhead for short-running suites. In
// addition to JVM startup time and JIT warmup, it appears that initialization of Derby
// metastores can be very slow so creating a fresh warehouse per suite is inefficient.
//
// 2. When parallelizing within a project we need to give each forked JVM a different tmpdir
// so that the metastore warehouses do not collide. Unfortunately, it seems that there are
// some tests which have an overly tight dependency on the default tmpdir, so those fragile
// tests need to continue re-running in the default configuration (or need to be rewritten).
// Fixing that problem would be a huge amount of work for limited payoff in most cases
// because most test suites are short-running.
//
private val testsWhichShouldRunInTheirOwnDedicatedJvm = Set(
"org.apache.spark.DistributedSuite",
"org.apache.spark.scheduler.HealthTrackerIntegrationSuite",
"org.apache.spark.sql.catalyst.expressions.DateExpressionsSuite",
"org.apache.spark.sql.catalyst.expressions.HashExpressionsSuite",
"org.apache.spark.sql.catalyst.expressions.CastSuite",
"org.apache.spark.sql.catalyst.expressions.MathExpressionsSuite",
"org.apache.spark.sql.hive.HiveExternalCatalogSuite",
"org.apache.spark.sql.hive.StatisticsSuite",
"org.apache.spark.sql.hive.client.HiveClientVersions",
"org.apache.spark.sql.hive.HiveExternalCatalogVersionsSuite",
"org.apache.spark.ml.classification.LogisticRegressionSuite",
"org.apache.spark.ml.classification.LinearSVCSuite",
"org.apache.spark.sql.SQLQueryTestSuite",
"org.apache.spark.sql.hive.client.HadoopVersionInfoSuite",
"org.apache.spark.sql.hive.thriftserver.SparkExecuteStatementOperationSuite",
"org.apache.spark.sql.hive.thriftserver.ThriftServerQueryTestSuite",
"org.apache.spark.sql.hive.thriftserver.SparkSQLEnvSuite",
"org.apache.spark.sql.hive.thriftserver.ui.ThriftServerPageSuite",
"org.apache.spark.sql.hive.thriftserver.ui.HiveThriftServer2ListenerSuite",
"org.apache.spark.sql.kafka010.KafkaDelegationTokenSuite",
"org.apache.spark.sql.streaming.RocksDBStateStoreStreamingAggregationSuite",
"org.apache.spark.shuffle.KubernetesLocalDiskShuffleDataIOSuite",
"org.apache.spark.sql.hive.HiveScalaReflectionSuite"
) ++ sys.env.get("DEDICATED_JVM_SBT_TESTS").map(_.split(",")).getOrElse(Array.empty).toSet
private val DEFAULT_TEST_GROUP = "default_test_group"
private val HIVE_EXECUTION_TEST_GROUP = "hive_execution_test_group"
private def testNameToTestGroup(name: String): String = name match {
case _ if testsWhichShouldRunInTheirOwnDedicatedJvm.contains(name) => name
// Different with the cases in testsWhichShouldRunInTheirOwnDedicatedJvm, here we are grouping
// all suites of `org.apache.spark.sql.hive.execution.*` into a single group, instead of
// launching one JVM per suite.
case _ if name.contains("org.apache.spark.sql.hive.execution") => HIVE_EXECUTION_TEST_GROUP
case _ => DEFAULT_TEST_GROUP
}
lazy val settings = Seq(
(Test / testGrouping) := {
val tests: Seq[TestDefinition] = (Test / definedTests).value
val defaultForkOptions = ForkOptions(
javaHome = javaHome.value,
outputStrategy = outputStrategy.value,
bootJars = Vector.empty[java.io.File],
workingDirectory = Some(baseDirectory.value),
runJVMOptions = (Test / javaOptions).value.toVector,
connectInput = connectInput.value,
envVars = (Test / envVars).value
)
tests.groupBy(test => testNameToTestGroup(test.name)).map { case (groupName, groupTests) =>
val forkOptions = {
if (groupName == DEFAULT_TEST_GROUP) {
defaultForkOptions
} else {
defaultForkOptions.withRunJVMOptions(defaultForkOptions.runJVMOptions ++
Seq(s"-Djava.io.tmpdir=${baseDirectory.value}/target/tmp/$groupName"))
}
}
new Tests.Group(
name = groupName,
tests = groupTests,
runPolicy = Tests.SubProcess(forkOptions))
}
}.toSeq
)
}
object CommonUtils {
import scala.sys.process.Process
def buildenv = Process(Seq("uname")).!!.trim.replaceFirst("[^A-Za-z0-9].*", "").toLowerCase
def bashpath = Process(Seq("where", "bash")).!!.split("[\r\n]+").head.replace('\\', '/')
lazy val settings = Seq(
(Compile / resourceGenerators) += Def.task {
val buildScript = baseDirectory.value + "/../../build/spark-build-info"
val targetDir = baseDirectory.value + "/target/extra-resources/"
// support Windows build under cygwin/mingw64, etc
val bash = buildenv match {
case "cygwin" | "msys2" | "mingw64" | "clang64" => bashpath
case _ => "bash"
}
val command = Seq(bash, buildScript, targetDir, version.value)
Process(command).!!
val propsFile = baseDirectory.value / "target" / "extra-resources" / "spark-version-info.properties"
Seq(propsFile)
}.taskValue
)
}
object Core {
import BuildCommons.protoVersion
lazy val settings = Seq(
// Setting version for the protobuf compiler. This has to be propagated to every sub-project
// even if the project is not using it.
PB.protocVersion := BuildCommons.protoVersion,
// For some reason the resolution from the imported Maven build does not work for some
// of these dependendencies that we need to shade later on.
libraryDependencies ++= {
Seq(
"com.google.protobuf" % "protobuf-java" % protoVersion % "protobuf"
)
},
(Compile / PB.targets) := Seq(
PB.gens.java -> (Compile / sourceManaged).value
)
) ++ {
val sparkProtocExecPath = sys.props.get("spark.protoc.executable.path")
if (sparkProtocExecPath.isDefined) {
Seq(
PB.protocExecutable := file(sparkProtocExecPath.get)
)
} else {
Seq.empty
}
}
}
object SparkConnectCommon {
import BuildCommons.protoVersion
lazy val settings = Seq(
// Setting version for the protobuf compiler. This has to be propagated to every sub-project
// even if the project is not using it.
PB.protocVersion := BuildCommons.protoVersion,
// For some reason the resolution from the imported Maven build does not work for some
// of these dependendencies that we need to shade later on.
libraryDependencies ++= {
val guavaVersion =
SbtPomKeys.effectivePom.value.getProperties.get(
"connect.guava.version").asInstanceOf[String]
val guavaFailureaccessVersion =
SbtPomKeys.effectivePom.value.getProperties.get(
"guava.failureaccess.version").asInstanceOf[String]
Seq(
"io.grpc" % "protoc-gen-grpc-java" % BuildCommons.grpcVersion asProtocPlugin(),
"com.google.guava" % "guava" % guavaVersion,
"com.google.guava" % "failureaccess" % guavaFailureaccessVersion,
"com.google.protobuf" % "protobuf-java" % protoVersion % "protobuf"
)
},
dependencyOverrides ++= {
val guavaVersion =
SbtPomKeys.effectivePom.value.getProperties.get(
"connect.guava.version").asInstanceOf[String]
val guavaFailureaccessVersion =
SbtPomKeys.effectivePom.value.getProperties.get(
"guava.failureaccess.version").asInstanceOf[String]
Seq(
"com.google.guava" % "guava" % guavaVersion,
"com.google.guava" % "failureaccess" % guavaFailureaccessVersion,
"com.google.protobuf" % "protobuf-java" % protoVersion
)
},
(assembly / test) := { },
(assembly / logLevel) := Level.Info,
// Exclude `scala-library` from assembly.
(assembly / assemblyPackageScala / assembleArtifact) := false,
// Exclude `pmml-model-*.jar`, `scala-collection-compat_*.jar`,`jsr305-*.jar` and
// `netty-*.jar` and `unused-1.0.0.jar` from assembly.
(assembly / assemblyExcludedJars) := {
val cp = (assembly / fullClasspath).value
cp filter { v =>
val name = v.data.getName
name.startsWith("pmml-model-") || name.startsWith("scala-collection-compat_") ||
name.startsWith("jsr305-") || name.startsWith("netty-") || name == "unused-1.0.0.jar"
}
},
(assembly / assemblyMergeStrategy) := {
case m if m.toLowerCase(Locale.ROOT).endsWith("manifest.mf") => MergeStrategy.discard
// Drop all proto files that are not needed as artifacts of the build.
case m if m.toLowerCase(Locale.ROOT).endsWith(".proto") => MergeStrategy.discard
case _ => MergeStrategy.first
}
) ++ {
val sparkProtocExecPath = sys.props.get("spark.protoc.executable.path")
val connectPluginExecPath = sys.props.get("connect.plugin.executable.path")
if (sparkProtocExecPath.isDefined && connectPluginExecPath.isDefined) {
Seq(
(Compile / PB.targets) := Seq(
PB.gens.java -> (Compile / sourceManaged).value,
PB.gens.plugin(name = "grpc-java", path = connectPluginExecPath.get) -> (Compile / sourceManaged).value
),
PB.protocExecutable := file(sparkProtocExecPath.get)
)
} else {
Seq(
(Compile / PB.targets) := Seq(
PB.gens.java -> (Compile / sourceManaged).value,
PB.gens.plugin("grpc-java") -> (Compile / sourceManaged).value
)
)
}
}
}
object SparkConnect {
import BuildCommons.protoVersion
lazy val settings = Seq(
// For some reason the resolution from the imported Maven build does not work for some
// of these dependendencies that we need to shade later on.
libraryDependencies ++= {
val guavaVersion =
SbtPomKeys.effectivePom.value.getProperties.get(
"connect.guava.version").asInstanceOf[String]
val guavaFailureaccessVersion =
SbtPomKeys.effectivePom.value.getProperties.get(
"guava.failureaccess.version").asInstanceOf[String]
Seq(
"com.google.guava" % "guava" % guavaVersion,
"com.google.guava" % "failureaccess" % guavaFailureaccessVersion,
"com.google.protobuf" % "protobuf-java" % protoVersion % "protobuf"
)
},
dependencyOverrides ++= {
val guavaVersion =
SbtPomKeys.effectivePom.value.getProperties.get(
"connect.guava.version").asInstanceOf[String]
val guavaFailureaccessVersion =
SbtPomKeys.effectivePom.value.getProperties.get(
"guava.failureaccess.version").asInstanceOf[String]
Seq(
"com.google.guava" % "guava" % guavaVersion,
"com.google.guava" % "failureaccess" % guavaFailureaccessVersion,
"com.google.protobuf" % "protobuf-java" % protoVersion
)
},
(assembly / test) := { },
(assembly / logLevel) := Level.Info,
// Exclude `scala-library` from assembly.
(assembly / assemblyPackageScala / assembleArtifact) := false,
// SPARK-46733: Include `spark-connect-*.jar`, `unused-*.jar`,`guava-*.jar`,
// `failureaccess-*.jar`, `annotations-*.jar`, `grpc-*.jar`, `protobuf-*.jar`,
// `gson-*.jar`, `error_prone_annotations-*.jar`, `j2objc-annotations-*.jar`,
// `animal-sniffer-annotations-*.jar`, `perfmark-api-*.jar`,
// `proto-google-common-protos-*.jar` in assembly.
// This needs to be consistent with the content of `maven-shade-plugin`.
(assembly / assemblyExcludedJars) := {
val cp = (assembly / fullClasspath).value
val validPrefixes = Set("spark-connect", "unused-", "guava-", "failureaccess-",
"annotations-", "grpc-", "protobuf-", "gson", "error_prone_annotations",
"j2objc-annotations", "animal-sniffer-annotations", "perfmark-api",
"proto-google-common-protos")
cp filterNot { v =>
validPrefixes.exists(v.data.getName.startsWith)
}
},
(assembly / assemblyShadeRules) := Seq(
ShadeRule.rename("io.grpc.**" -> "org.sparkproject.connect.grpc.@0").inAll,
ShadeRule.rename("com.google.common.**" -> "org.sparkproject.connect.guava.@1").inAll,
ShadeRule.rename("com.google.thirdparty.**" -> "org.sparkproject.connect.guava.@1").inAll,
ShadeRule.rename("com.google.protobuf.**" -> "org.sparkproject.connect.protobuf.@1").inAll,
ShadeRule.rename("android.annotation.**" -> "org.sparkproject.connect.android_annotation.@1").inAll,
ShadeRule.rename("io.perfmark.**" -> "org.sparkproject.connect.io_perfmark.@1").inAll,
ShadeRule.rename("org.codehaus.mojo.animal_sniffer.**" -> "org.sparkproject.connect.animal_sniffer.@1").inAll,
ShadeRule.rename("com.google.j2objc.annotations.**" -> "org.sparkproject.connect.j2objc_annotations.@1").inAll,
ShadeRule.rename("com.google.errorprone.annotations.**" -> "org.sparkproject.connect.errorprone_annotations.@1").inAll,
ShadeRule.rename("org.checkerframework.**" -> "org.sparkproject.connect.checkerframework.@1").inAll,
ShadeRule.rename("com.google.gson.**" -> "org.sparkproject.connect.gson.@1").inAll,
ShadeRule.rename("com.google.api.**" -> "org.sparkproject.connect.google_protos.api.@1").inAll,
ShadeRule.rename("com.google.cloud.**" -> "org.sparkproject.connect.google_protos.cloud.@1").inAll,
ShadeRule.rename("com.google.geo.**" -> "org.sparkproject.connect.google_protos.geo.@1").inAll,
ShadeRule.rename("com.google.logging.**" -> "org.sparkproject.connect.google_protos.logging.@1").inAll,
ShadeRule.rename("com.google.longrunning.**" -> "org.sparkproject.connect.google_protos.longrunning.@1").inAll,
ShadeRule.rename("com.google.rpc.**" -> "org.sparkproject.connect.google_protos.rpc.@1").inAll,
ShadeRule.rename("com.google.type.**" -> "org.sparkproject.connect.google_protos.type.@1").inAll
),
(assembly / assemblyMergeStrategy) := {
case m if m.toLowerCase(Locale.ROOT).endsWith("manifest.mf") => MergeStrategy.discard
// Drop all proto files that are not needed as artifacts of the build.
case m if m.toLowerCase(Locale.ROOT).endsWith(".proto") => MergeStrategy.discard
case _ => MergeStrategy.first
}
)
}
object SparkConnectClient {
import BuildCommons.protoVersion
val buildTestDeps = TaskKey[Unit]("buildTestDeps", "Build needed dependencies for test.")
lazy val settings = Seq(
// For some reason the resolution from the imported Maven build does not work for some
// of these dependendencies that we need to shade later on.
libraryDependencies ++= {
val guavaVersion =
SbtPomKeys.effectivePom.value.getProperties.get(
"connect.guava.version").asInstanceOf[String]
Seq(
"com.google.guava" % "guava" % guavaVersion,
"com.google.protobuf" % "protobuf-java" % protoVersion % "protobuf"
)
},
dependencyOverrides ++= {
val guavaVersion =
SbtPomKeys.effectivePom.value.getProperties.get(
"connect.guava.version").asInstanceOf[String]
Seq(
"com.google.guava" % "guava" % guavaVersion,
"com.google.protobuf" % "protobuf-java" % protoVersion
)
},
buildTestDeps := {
(LocalProject("assembly") / Compile / Keys.`package`).value
(LocalProject("catalyst") / Test / Keys.`package`).value
},
// SPARK-42538: Make sure the `${SPARK_HOME}/assembly/target/scala-$SPARK_SCALA_VERSION/jars` is available for testing.
// At the same time, the build of `connect`, `connect-client-jvm` and `sql` will be triggered by `assembly` build,
// so no additional configuration is required.
test := ((Test / test) dependsOn (buildTestDeps)).value,
testOnly := ((Test / testOnly) dependsOn (buildTestDeps)).evaluated,
(assembly / test) := { },
(assembly / logLevel) := Level.Info,
// Exclude `scala-library` from assembly.
(assembly / assemblyPackageScala / assembleArtifact) := false,
// Exclude `pmml-model-*.jar`, `scala-collection-compat_*.jar`,`jsr305-*.jar` and
// `netty-*.jar` and `unused-1.0.0.jar` from assembly.
(assembly / assemblyExcludedJars) := {
val cp = (assembly / fullClasspath).value
cp filter { v =>
val name = v.data.getName
name.startsWith("pmml-model-") || name.startsWith("scala-collection-compat_") ||
name.startsWith("jsr305-") || name == "unused-1.0.0.jar"
}
},
(assembly / assemblyShadeRules) := Seq(
ShadeRule.rename("io.grpc.**" -> "org.sparkproject.connect.client.io.grpc.@1").inAll,
ShadeRule.rename("com.google.**" -> "org.sparkproject.connect.client.com.google.@1").inAll,
ShadeRule.rename("io.netty.**" -> "org.sparkproject.connect.client.io.netty.@1").inAll,
ShadeRule.rename("org.checkerframework.**" -> "org.sparkproject.connect.client.org.checkerframework.@1").inAll,
ShadeRule.rename("javax.annotation.**" -> "org.sparkproject.connect.client.javax.annotation.@1").inAll,
ShadeRule.rename("io.perfmark.**" -> "org.sparkproject.connect.client.io.perfmark.@1").inAll,
ShadeRule.rename("org.codehaus.**" -> "org.sparkproject.connect.client.org.codehaus.@1").inAll,
ShadeRule.rename("android.annotation.**" -> "org.sparkproject.connect.client.android.annotation.@1").inAll
),
(assembly / assemblyMergeStrategy) := {
case m if m.toLowerCase(Locale.ROOT).endsWith("manifest.mf") => MergeStrategy.discard
// Drop all proto files that are not needed as artifacts of the build.
case m if m.toLowerCase(Locale.ROOT).endsWith(".proto") => MergeStrategy.discard
case _ => MergeStrategy.first
}
)
}
object SparkProtobuf {
import BuildCommons.protoVersion
lazy val settings = Seq(
// Setting version for the protobuf compiler. This has to be propagated to every sub-project
// even if the project is not using it.
PB.protocVersion := BuildCommons.protoVersion,
// For some reason the resolution from the imported Maven build does not work for some
// of these dependendencies that we need to shade later on.
libraryDependencies += "com.google.protobuf" % "protobuf-java" % protoVersion % "protobuf",
dependencyOverrides += "com.google.protobuf" % "protobuf-java" % protoVersion,
(Test / PB.protoSources) += (Test / sourceDirectory).value / "resources" / "protobuf",
(Test / PB.protocOptions) += "--include_imports",
(Test / PB.targets) := Seq(
PB.gens.java -> target.value / "generated-test-sources",
PB.gens.descriptorSet -> target.value / "generated-test-sources/descriptor-set-sbt.desc",
// The above creates single descriptor file with all the proto files. This is different from
// Maven, which create one descriptor file for each proto file.
),
(assembly / test) := { },
(assembly / logLevel) := Level.Info,
// Exclude `scala-library` from assembly.
(assembly / assemblyPackageScala / assembleArtifact) := false,
// Exclude `pmml-model-*.jar`, `scala-collection-compat_*.jar`,
// `spark-tags_*.jar`, "guava-*.jar" and `unused-1.0.0.jar` from assembly.
(assembly / assemblyExcludedJars) := {
val cp = (assembly / fullClasspath).value
cp filter { v =>
val name = v.data.getName
name.startsWith("pmml-model-") || name.startsWith("scala-collection-compat_") ||
name.startsWith("spark-tags_") || name.startsWith("guava-") || name == "unused-1.0.0.jar"
}
},
(assembly / assemblyShadeRules) := Seq(
ShadeRule.rename("com.google.protobuf.**" -> "org.sparkproject.spark_protobuf.protobuf.@1").inAll,
),
(assembly / assemblyMergeStrategy) := {
case m if m.toLowerCase(Locale.ROOT).endsWith("manifest.mf") => MergeStrategy.discard
// Drop all proto files that are not needed as artifacts of the build.
case m if m.toLowerCase(Locale.ROOT).endsWith(".proto") => MergeStrategy.discard
case _ => MergeStrategy.first
},
) ++ {
val sparkProtocExecPath = sys.props.get("spark.protoc.executable.path")
if (sparkProtocExecPath.isDefined) {
Seq(
PB.protocExecutable := file(sparkProtocExecPath.get)
)
} else {
Seq.empty
}
}
}
object Unsafe {
lazy val settings = Seq(
// This option is needed to suppress warnings from sun.misc.Unsafe usage
(Compile / javacOptions) += "-XDignore.symbol.file"
)
}
object DockerIntegrationTests {
// This serves to override the override specified in DependencyOverrides:
lazy val settings = Seq(
dependencyOverrides += "com.google.guava" % "guava" % "33.1.0-jre"
)
}
/**
* These settings run the Kubernetes integration tests.
* Docker images will have the "dev" tag, and will be overwritten every time the
* integration tests are run. The integration tests are actually bound to the "test" phase,
* so running "test" on this module will run the integration tests.
*
* There are two ways to run the tests:
* - the "tests" task builds docker images and runs the test, so it's a little slow.
* - the "run-its" task just runs the tests on a pre-built set of images.
*
* Note that this does not use the shell scripts that the maven build uses, which are more
* configurable. This is meant as a quick way for developers to run these tests against their
* local changes.
*/
object KubernetesIntegrationTests {
import BuildCommons._
import scala.sys.process.Process
val dockerBuild = TaskKey[Unit]("docker-imgs", "Build the docker images for ITs.")
val runITs = TaskKey[Unit]("run-its", "Only run ITs, skip image build.")
val imageRepo = sys.props.getOrElse("spark.kubernetes.test.imageRepo", "docker.io/kubespark")
val imageTag = sys.props.get("spark.kubernetes.test.imageTag")
val namespace = sys.props.get("spark.kubernetes.test.namespace")
val deployMode = sys.props.get("spark.kubernetes.test.deployMode")
// Hack: this variable is used to control whether to build docker images. It's updated by
// the tasks below in a non-obvious way, so that you get the functionality described in
// the scaladoc above.
private var shouldBuildImage = true
lazy val settings = Seq(
dockerBuild := {
if (shouldBuildImage) {
val dockerTool = s"$sparkHome/bin/docker-image-tool.sh"
val bindingsDir = s"$sparkHome/resource-managers/kubernetes/docker/src/main/dockerfiles/spark/bindings"
val javaImageTag = sys.props.get("spark.kubernetes.test.javaImageTag")
val dockerFile = sys.props.getOrElse("spark.kubernetes.test.dockerFile",
s"$sparkHome/resource-managers/kubernetes/docker/src/main/dockerfiles/spark/Dockerfile")
val pyDockerFile = sys.props.getOrElse("spark.kubernetes.test.pyDockerFile",
s"$bindingsDir/python/Dockerfile")