Skip to content

#765 Allow skipping 'msck repair table' queries explicitly when creating Hive tables #767

New issue

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

Sign up for GitHub

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

Already on GitHub? Sign in to your account

Merged
yruslan merged 1 commit into from
Jun 26, 2026
Merged

#765 Allow skipping 'msck repair table' queries explicitly when creating Hive tables #767

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 3 additions & 3 deletions pramen/core/src/main/scala/za/co/absa/pramen/core/metastore/MetastoreImpl.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -206,7 +206,7 @@ class MetastoreImpl(appConfig: Config,

if (recreate) {
log.info(s"Recreating Hive table '$fullTableName'")
hiveHelper.createOrUpdateHiveTable(effectivePath, format, effectiveSchema, Seq(mt.infoDateColumn), mt.hiveConfig.database, hiveTable, neverRepairPartitions)
hiveHelper.createOrUpdateHiveTable(effectivePath, format, effectiveSchema, Seq(mt.infoDateColumn), mt.hiveConfig.database, hiveTable, !neverRepairPartitions)
} else {
if (hiveHelper.doesTableExist(mt.hiveConfig.database, hiveTable)) {
if (updateSchema) {
Expand All @@ -218,15 +218,15 @@ class MetastoreImpl(appConfig: Config,
} catch {
case NonFatal(ex) =>
log.warn(s"Could not update Hive schema via ${hiveHelper.getClass.getName}. Recreating Hive table '$fullTableName'", ex)
hiveHelper.createOrUpdateHiveTable(effectivePath, format, effectiveSchema, Seq(mt.infoDateColumn), mt.hiveConfig.database, hiveTable, neverRepairPartitions)
hiveHelper.createOrUpdateHiveTable(effectivePath, format, effectiveSchema, Seq(mt.infoDateColumn), mt.hiveConfig.database, hiveTable, !neverRepairPartitions)
}
} else {
// Schema didn't change, but we need to add the new partition
needAddPartition = true
}
} else {
log.info(s"The table '$fullTableName' does not exist. Creating it.")
hiveHelper.createOrUpdateHiveTable(effectivePath, format, effectiveSchema, Seq(mt.infoDateColumn), mt.hiveConfig.database, hiveTable, neverRepairPartitions)
hiveHelper.createOrUpdateHiveTable(effectivePath, format, effectiveSchema, Seq(mt.infoDateColumn), mt.hiveConfig.database, hiveTable, !neverRepairPartitions)
}
}

Expand Down
4 changes: 2 additions & 2 deletions pramen/core/src/main/scala/za/co/absa/pramen/core/utils/hive/HiveHelper.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -29,15 +29,15 @@ abstract class HiveHelper {
partitionBy: Seq[String],
databaseName: Option[String],
tableName: String,
neverRepairPartitions: Boolean = false): Unit
autoRepairPartitions: Boolean = true): Unit

def createOrUpdateHiveTable(path: String,
format: HiveFormat,
schema: StructType,
partitionBy: Seq[String],
databaseName: Option[String],
tableName: String,
neverRepairPartitions: Boolean = false): Unit
autoRepairPartitions: Boolean = true): Unit

def replaceHiveTableSchema(schema: StructType,
partitionBy: Seq[String],
Expand Down
12 changes: 6 additions & 6 deletions pramen/core/src/main/scala/za/co/absa/pramen/core/utils/hive/HiveHelperSparkCatalog.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -33,18 +33,18 @@ class HiveHelperSparkCatalog(spark: SparkSession) extends HiveHelper {
partitionBy: Seq[String],
databaseName: Option[String],
tableName: String,
neverRepairPartitions: Boolean): Unit = {
autoRepairPartitions: Boolean): Unit = {
val fullTableName = HiveHelper.getFullTable(databaseName, tableName)

createCatalogTable(fullTableName, path, format)

if (partitionBy.nonEmpty && !neverRepairPartitions) {
if (partitionBy.nonEmpty && autoRepairPartitions) {
repairHiveTable(databaseName, tableName, format)
} else {
if (partitionBy.isEmpty)
log.info(s"Skipping repairing partition for $fullTableName because the table is not partitioned.")
else
log.info(s"Skipping repairing partition for $fullTableName because repairing partitions is disabled.")
log.info(s"Skipping repairing partition for $fullTableName as requested.")
}

if (!doesTableExist(databaseName, tableName)) {
Expand All @@ -58,7 +58,7 @@ class HiveHelperSparkCatalog(spark: SparkSession) extends HiveHelper {
partitionBy: Seq[String],
databaseName: Option[String],
tableName: String,
neverRepairPartitions: Boolean): Unit = {
autoRepairPartitions: Boolean): Unit = {
val fullTableName = HiveHelper.getFullTable(databaseName, tableName)

if (doesTableExist(databaseName, tableName)) {
Expand All @@ -68,13 +68,13 @@ class HiveHelperSparkCatalog(spark: SparkSession) extends HiveHelper {

createCatalogTable(fullTableName, path, format)

if (partitionBy.nonEmpty && !neverRepairPartitions) {
if (partitionBy.nonEmpty && autoRepairPartitions) {
repairHiveTable(databaseName, tableName, format)
} else {
if (partitionBy.isEmpty)
log.info(s"Skipping repairing partition for $fullTableName because the table is not partitioned.")
else
log.info(s"Skipping repairing partition for $fullTableName because repairing partitions is disabled.")
log.info(s"Skipping repairing partition for $fullTableName as requested.")
}

if (!doesTableExist(databaseName, tableName)) {
Expand Down
12 changes: 6 additions & 6 deletions pramen/core/src/main/scala/za/co/absa/pramen/core/utils/hive/HiveHelperSql.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -31,17 +31,17 @@ class HiveHelperSql(val queryExecutor: QueryExecutor,
partitionBy: Seq[String],
databaseName: Option[String],
tableName: String,
neverRepairPartitions: Boolean): Unit = {
autoRepairPartitions: Boolean): Unit = {
val fullTableName = HiveHelper.getFullTable(databaseName, tableName)

createHiveTable(fullTableName, path, format, schema, partitionBy, failIfExists = true)
if (partitionBy.nonEmpty && !neverRepairPartitions) {
if (partitionBy.nonEmpty && autoRepairPartitions) {
repairHiveTable(fullTableName)
} else {
if (partitionBy.isEmpty)
log.info(s"Skipping repairing partition for $fullTableName because the table is not partitioned.")
else
log.info(s"Skipping repairing partition for $fullTableName because repairing partitions is disabled.")
log.info(s"Skipping repairing partition for $fullTableName as requested.")
}
}

Expand All @@ -51,18 +51,18 @@ class HiveHelperSql(val queryExecutor: QueryExecutor,
partitionBy: Seq[String],
databaseName: Option[String],
tableName: String,
neverRepairPartitions: Boolean): Unit = {
autoRepairPartitions: Boolean): Unit = {
val fullTableName = HiveHelper.getFullTable(databaseName, tableName)

dropHiveTable(fullTableName)
createHiveTable(fullTableName, path, format, schema, partitionBy, failIfExists = false)
if (partitionBy.nonEmpty && !neverRepairPartitions) {
if (partitionBy.nonEmpty && autoRepairPartitions) {
repairHiveTable(fullTableName)
} else {
if (partitionBy.isEmpty)
log.info(s"Skipping repairing partition for $fullTableName because the table is not partitioned.")
else
log.info(s"Skipping repairing partition for $fullTableName because repairing partitions is disabled.")
log.info(s"Skipping repairing partition for $fullTableName as requested.")
}
}

Expand Down
24 changes: 12 additions & 12 deletions .../src/test/scala/za/co/absa/pramen/core/tests/utils/hive/HiveHelperSparkCatalogSuite.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -38,12 +38,12 @@ class HiveHelperSparkCatalogSuite extends AnyWordSpec with SparkTestBase with Te
val hiveHelper = new HiveHelperSparkCatalog(spark)
val schema = spark.read.parquet(path).schema

hiveHelper.createHiveTable(path, HiveFormat.Parquet, schema, Nil, Some("default"), "tbl1", neverRepairPartitions = false)
hiveHelper.createHiveTable(path, HiveFormat.Parquet, schema, Nil, Some("default"), "tbl1")
assert(spark.catalog.tableExists("default.tbl1"))

// If the table exists an exception should be thrown
assertThrows[AnalysisException] {
hiveHelper.createHiveTable(path, HiveFormat.Parquet, schema, Nil, Some("default"), "tbl1", neverRepairPartitions = false)
hiveHelper.createHiveTable(path, HiveFormat.Parquet, schema, Nil, Some("default"), "tbl1")
}
}
}
Expand All @@ -55,13 +55,13 @@ class HiveHelperSparkCatalogSuite extends AnyWordSpec with SparkTestBase with Te
val hiveHelper = new HiveHelperSparkCatalog(spark)
val schema = spark.read.parquet(path).withColumn("b", lit(1)).schema

hiveHelper.createHiveTable(path, HiveFormat.Parquet, schema, "a" :: "b" :: Nil, Some("default"), "tbl2", neverRepairPartitions = false)
hiveHelper.createHiveTable(path, HiveFormat.Parquet, schema, "a" :: "b" :: Nil, Some("default"), "tbl2")
assert(hiveHelper.doesTableExist(Some("default"), "tbl2"))

spark.sql(s"DROP TABLE default.tbl2").collect()
assert(!hiveHelper.doesTableExist(Some("default"), "tbl2"))

hiveHelper.createHiveTable(path, HiveFormat.Parquet, schema, "a" :: "b" :: Nil, Some("default"), "tbl2", neverRepairPartitions = false)
hiveHelper.createHiveTable(path, HiveFormat.Parquet, schema, "a" :: "b" :: Nil, Some("default"), "tbl2")
assert(hiveHelper.doesTableExist(Some("default"), "tbl2"))
}
}
Expand All @@ -73,7 +73,7 @@ class HiveHelperSparkCatalogSuite extends AnyWordSpec with SparkTestBase with Te
val hiveHelper = new HiveHelperSparkCatalog(spark)
val schema = spark.read.parquet(path).withColumn("b", lit(1)).schema

hiveHelper.createHiveTable(path, HiveFormat.Parquet, schema, "a" :: "b" :: Nil, Some("default"), "tbl3", neverRepairPartitions = false)
hiveHelper.createHiveTable(path, HiveFormat.Parquet, schema, "a" :: "b" :: Nil, Some("default"), "tbl3")
assert(hiveHelper.doesTableExist(Some("default"), "tbl3"))

assertThrows[AnalysisException] {
Expand All @@ -91,11 +91,11 @@ class HiveHelperSparkCatalogSuite extends AnyWordSpec with SparkTestBase with Te
val hiveHelper = new HiveHelperSparkCatalog(spark)
val schema = spark.read.parquet(path).schema

hiveHelper.createOrUpdateHiveTable(path, HiveFormat.Parquet, schema, Nil, Some("default"), "tbl1", neverRepairPartitions = false)
hiveHelper.createOrUpdateHiveTable(path, HiveFormat.Parquet, schema, Nil, Some("default"), "tbl1")
assert(spark.catalog.tableExists("default.tbl1"))

// If the table exists it should be re-created
hiveHelper.createOrUpdateHiveTable(path, HiveFormat.Parquet, schema, Nil, Some("default"), "tbl1", neverRepairPartitions = false)
hiveHelper.createOrUpdateHiveTable(path, HiveFormat.Parquet, schema, Nil, Some("default"), "tbl1")
assert(spark.catalog.tableExists("default.tbl1"))
}
}
Expand All @@ -107,13 +107,13 @@ class HiveHelperSparkCatalogSuite extends AnyWordSpec with SparkTestBase with Te
val hiveHelper = new HiveHelperSparkCatalog(spark)
val schema = spark.read.parquet(path).withColumn("b", lit(1)).schema

hiveHelper.createOrUpdateHiveTable(path, HiveFormat.Parquet, schema, "a" :: "b" :: Nil, Some("default"), "tbl2", neverRepairPartitions = false)
hiveHelper.createOrUpdateHiveTable(path, HiveFormat.Parquet, schema, "a" :: "b" :: Nil, Some("default"), "tbl2")
assert(hiveHelper.doesTableExist(Some("default"),"tbl2"))

spark.sql(s"DROP TABLE default.tbl2").collect()
assert(!hiveHelper.doesTableExist(Some("default"),"tbl2"))

hiveHelper.createOrUpdateHiveTable(path, HiveFormat.Parquet, schema, "a" :: "b" :: Nil, Some("default"), "tbl2", neverRepairPartitions = false)
hiveHelper.createOrUpdateHiveTable(path, HiveFormat.Parquet, schema, "a" :: "b" :: Nil, Some("default"), "tbl2")
assert(hiveHelper.doesTableExist(Some("default"),"tbl2"))
}
}
Expand All @@ -125,7 +125,7 @@ class HiveHelperSparkCatalogSuite extends AnyWordSpec with SparkTestBase with Te
val hiveHelper = new HiveHelperSparkCatalog(spark)
val schema = spark.read.format("delta").load(path).withColumn("b", lit(1)).schema

hiveHelper.createOrUpdateHiveTable(path, HiveFormat.Delta, schema, "b" :: Nil, Some("default"), "tbl3", neverRepairPartitions = false)
hiveHelper.createOrUpdateHiveTable(path, HiveFormat.Delta, schema, "b" :: Nil, Some("default"), "tbl3")
assert(hiveHelper.doesTableExist(Some("default"), "tbl3"))

assert(spark.table("default.tbl3").count() == 3)
Expand All @@ -139,7 +139,7 @@ class HiveHelperSparkCatalogSuite extends AnyWordSpec with SparkTestBase with Te
val hiveHelper = new HiveHelperSparkCatalog(spark)
val schema = spark.read.parquet(path).withColumn("b", lit(1)).schema

hiveHelper.createOrUpdateHiveTable(path, HiveFormat.Parquet, schema, "a" :: "b" :: Nil, Some("default"), "tbl4", neverRepairPartitions = true)
hiveHelper.createOrUpdateHiveTable(path, HiveFormat.Parquet, schema, "a" :: "b" :: Nil, Some("default"), "tbl4")
assert(hiveHelper.doesTableExist(Some("default"), "tbl4"))

hiveHelper.dropTable(Some("default"), "tbl4")
Expand All @@ -155,7 +155,7 @@ class HiveHelperSparkCatalogSuite extends AnyWordSpec with SparkTestBase with Te
val hiveHelper = new HiveHelperSparkCatalog(spark)
val schema = spark.read.parquet(path).withColumn("b", lit(1)).schema

hiveHelper.createOrUpdateHiveTable(path, HiveFormat.Parquet, schema, "a" :: "b" :: Nil, Some("default"), "tbl5", neverRepairPartitions = false)
hiveHelper.createOrUpdateHiveTable(path, HiveFormat.Parquet, schema, "a" :: "b" :: Nil, Some("default"), "tbl5")
assert(hiveHelper.doesTableExist(Some("default"),"tbl5"))

val df = List(("D", 40)).toDF("a", "c")
Expand Down
8 changes: 4 additions & 4 deletions pramen/core/src/test/scala/za/co/absa/pramen/core/tests/utils/hive/HiveHelperSqlSuite.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -55,7 +55,7 @@ class HiveHelperSqlSuite extends AnyWordSpec with SparkTestBase with TempDirFixt
val hiveHelper = new HiveHelperSql(qe, defaultHiveConfig, false)
val schema = spark.read.parquet(path).schema

hiveHelper.createOrUpdateHiveTable(path, HiveFormat.Parquet, schema, Nil, Some("db"), "tbl", neverRepairPartitions = false)
hiveHelper.createOrUpdateHiveTable(path, HiveFormat.Parquet, schema, Nil, Some("db"), "tbl", autoRepairPartitions = false)

qe.close()

Expand Down Expand Up @@ -86,7 +86,7 @@ class HiveHelperSqlSuite extends AnyWordSpec with SparkTestBase with TempDirFixt
val hiveHelper = new HiveHelperSql(qe, defaultHiveConfig, true)
val schema = spark.read.parquet(path).withColumn("b", lit(1)).schema

hiveHelper.createOrUpdateHiveTable(path, HiveFormat.Parquet, schema, "a" :: "b" :: Nil, Some("db"), "tbl", neverRepairPartitions = false)
hiveHelper.createOrUpdateHiveTable(path, HiveFormat.Parquet, schema, "a" :: "b" :: Nil, Some("db"), "tbl")

val actual = qe.queries.mkString("\n")

Expand All @@ -113,7 +113,7 @@ class HiveHelperSqlSuite extends AnyWordSpec with SparkTestBase with TempDirFixt
val hiveHelper = new HiveHelperSql(qe, defaultHiveConfig, true)
val schema = spark.read.parquet(path).withColumn("b", lit(1)).schema

hiveHelper.createOrUpdateHiveTable(path, HiveFormat.Parquet, schema, "a" :: "b" :: Nil, Some("db"), "tbl", neverRepairPartitions = true)
hiveHelper.createOrUpdateHiveTable(path, HiveFormat.Parquet, schema, "a" :: "b" :: Nil, Some("db"), "tbl", autoRepairPartitions = false)

val actual = qe.queries.mkString("\n")

Expand Down Expand Up @@ -141,7 +141,7 @@ class HiveHelperSqlSuite extends AnyWordSpec with SparkTestBase with TempDirFixt
val hiveHelper = new HiveHelperSql(qe, defaultHiveConfig, true)
val schema = spark.read.parquet(path).withColumn("b", lit(1)).schema

hiveHelper.createHiveTable(path, HiveFormat.Parquet, schema, "a" :: "b" :: Nil, Some("db"), "tbl", neverRepairPartitions = false)
hiveHelper.createHiveTable(path, HiveFormat.Parquet, schema, "a" :: "b" :: Nil, Some("db"), "tbl")

val actual = qe.queries.mkString("\n")

Expand Down
2 changes: 1 addition & 1 deletion pramen/extras/src/main/scala/za/co/absa/pramen/extras/sink/EnceladusSink.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -451,7 +451,7 @@ class EnceladusSink(sinkConfig: Config,

val schema = dfForHiveSchema.schema

hiveHelper.createOrUpdateHiveTable(publishBase, HiveFormat.Parquet, schema, Seq("enceladus_info_date", "enceladus_info_version"), enceladusConfig.hiveDatabase, hiveTable, neverRepairPartitions)
hiveHelper.createOrUpdateHiveTable(publishBase, HiveFormat.Parquet, schema, Seq("enceladus_info_date", "enceladus_info_version"), enceladusConfig.hiveDatabase, hiveTable, !neverRepairPartitions)
}
}

Expand Down
3 changes: 1 addition & 2 deletions pramen/extras/src/main/scala/za/co/absa/pramen/extras/sink/StandardizationSink.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -234,8 +234,7 @@ class StandardizationSink(sinkConfig: Config,
fullSchema,
partitionBy,
standardizationConfig.hiveDatabase,
hiveTable,
neverRepairPartitions = false)
hiveTable)
(Seq.empty[String], Seq(fullTableName))
} catch {
case NonFatal(ex) =>
Expand Down
Loading