Supporting special chars in column names by using back ticks. Closed https://github.com/apache/spark/pull/2804 and created this PR as it has merge conflicts
Author: ravipesala <ravindra.pesala@huawei.com>
Closes#2927 from ravipesala/SPARK-3483-NEW and squashes the following commits:
f6329f3 [ravipesala] Rebased with master
Please refer to added tests for cases that can trigger the bug.
JIRA: https://issues.apache.org/jira/browse/SPARK-4068
Author: Yin Huai <huai@cse.ohio-state.edu>
Closes#2918 from yhuai/SPARK-4068 and squashes the following commits:
d360eae [Yin Huai] Handle nulls when building key paths from elements of an array.
Please check https://issues.apache.org/jira/browse/SPARK-4052 for cases triggering this bug.
Author: Yin Huai <huai@cse.ohio-state.edu>
Closes#2899 from yhuai/SPARK-4052 and squashes the following commits:
1188f70 [Yin Huai] Address liancheng's comments.
b6712be [Yin Huai] Use scala.collection.Map instead of Predef.Map (scala.collection.immutable.Map).
This PR refines in-memory columnar table statistics:
1. adds 2 more statistics for in-memory table columns: `count` and `sizeInBytes`
1. adds filter pushdown support for `IS NULL` and `IS NOT NULL`.
1. caches and propagates statistics in `InMemoryRelation` once the underlying cached RDD is materialized.
Statistics are collected to driver side with an accumulator.
This PR also fixes SPARK-3914 by properly propagating in-memory statistics.
Author: Cheng Lian <lian@databricks.com>
Closes#2860 from liancheng/propagates-in-mem-stats and squashes the following commits:
0cc5271 [Cheng Lian] Restricts visibility of o.a.s.s.c.p.l.Statistics
c5ff904 [Cheng Lian] Fixes test table name conflict
a8c818d [Cheng Lian] Refines tests
1d01074 [Cheng Lian] Bug fix: shouldn't call STRING.actualSize on null string value
7dc6a34 [Cheng Lian] Adds more in-memory table statistics and propagates them properly
This follows https://github.com/apache/spark/pull/2893 , but does not completely fix SPARK-3359 either. This fixes minor scaladoc/javadoc issues that Javadoc 8 will treat as errors.
Author: Sean Owen <sowen@cloudera.com>
Closes#2909 from srowen/SPARK-3359 and squashes the following commits:
f62c347 [Sean Owen] Fix some javadoc issues that javadoc 8 considers errors. This is not all of the errors turned up when javadoc 8 runs on output of genjavadoc.
As part of the upgrade I also copy the newest version of the query tests, and whitelist a bunch of new ones that are now passing.
Author: Michael Armbrust <michael@databricks.com>
Closes#2936 from marmbrus/fix13tests and squashes the following commits:
d9cbdab [Michael Armbrust] Remove user specific tests
65801cd [Michael Armbrust] style and rat
8f6b09a [Michael Armbrust] Update test harness to work with both Hive 12 and 13.
f044843 [Michael Armbrust] Update Hive query tests and golden files to 0.13
Previously cached data was found by `sameResult` plan matching on optimized plans. This technique however fails to locate the cached data when a temporary table with a projection is queried with a further reduced projection. The failure is due to the fact that optimization will collapse the projections, producing a plan that no longer produces the sameResult as the cached data (though the cached data still subsumes the desired data). For example consider the following previously failing test case.
```scala
sql("CACHE TABLE tempTable AS SELECT key FROM testData")
assertCached(sql("SELECT COUNT(*) FROM tempTable"))
```
In this PR I change the matching to occur after analysis instead of optimization, so that in the case of temporary tables, the plans will always match. I think this should work generally, however, this error does raise questions about the need to do more thorough subsumption checking when locating cached data.
Another question is what sort of semantics we want to provide when uncaching data from temporary tables. For example consider the following sequence of commands:
```scala
testData.select('key).registerTempTable("tempTable1")
testData.select('key).registerTempTable("tempTable2")
cacheTable("tempTable1")
// This obviously works.
assertCached(sql("SELECT COUNT(*) FROM tempTable1"))
// It seems good that this works ...
assertCached(sql("SELECT COUNT(*) FROM tempTable2"))
// ... but is this valid?
uncacheTable("tempTable2")
// Should this still be cached?
assertCached(sql("SELECT COUNT(*) FROM tempTable1"), 0)
```
Author: Michael Armbrust <michael@databricks.com>
Closes#2912 from marmbrus/cachingBug and squashes the following commits:
9c822d4 [Michael Armbrust] remove commented out code
5c72fb7 [Michael Armbrust] Add a test case / question about uncaching semantics.
63a23e4 [Michael Armbrust] Perform caching on analyzed instead of optimized plan.
03f1cfe [Michael Armbrust] Clean-up / add tests to SameResult suite.
Some developers want to replace `Optimizer` to fit their projects but can't do so because currently `Optimizer` is an `object`.
Author: Takuya UESHIN <ueshin@happy-camper.st>
Closes#2825 from ueshin/issues/SPARK-3969 and squashes the following commits:
abbc53c [Takuya UESHIN] Re-rename Optimizer object.
4d2e1bc [Takuya UESHIN] Rename Optimizer object.
9547a23 [Takuya UESHIN] Extract abstract class from Optimizer for developers to be able to replace Optimizer.
Author: Michael Armbrust <michael@databricks.com>
Closes#2658 from marmbrus/nestedAggs and squashes the following commits:
862b763 [Michael Armbrust] Merge remote-tracking branch 'origin/master' into nestedAggs
3234521 [Michael Armbrust] Merge remote-tracking branch 'origin/master' into nestedAggs
8b06fdc [Michael Armbrust] possible fix for grouping on nested fields
Author: Cheng Lian <lian.cs.zju@gmail.com>
Closes#2767 from liancheng/multi-join and squashes the following commits:
9dc0d18 [Cheng Lian] Adds multiple join support for SQLContext
Make JavaPairRDD.collectAsMap result Serializable since Java Maps generally are
Author: Sean Owen <sowen@cloudera.com>
Closes#2805 from srowen/SPARK-3926 and squashes the following commits:
ecb78ee [Sean Owen] Fix conflict between java.io.Serializable and use of Scala's Serializable
f4717f9 [Sean Owen] Oops, fix compile problem
ae1b36f [Sean Owen] Expand to cover Maps returned from other Java API methods as well
51c26c2 [Sean Owen] Make JavaPairRDD.collectAsMap result Serializable since Java Maps generally are
In the current implementation it was possible for the reference to change after analysis.
Author: Michael Armbrust <michael@databricks.com>
Closes#2717 from marmbrus/pythonUdfResults and squashes the following commits:
da14879 [Michael Armbrust] Fix test
6343bcb [Michael Armbrust] add test
9533286 [Michael Armbrust] Correctly preserve the result attribute of python UDFs though transformations
This is a small number of clean-up changes on top of #2782. Closes#2782.
Author: Prashant Sharma <prashant.s@imaginea.com>
Author: Patrick Wendell <pwendell@gmail.com>
Closes#2803 from pwendell/pr-2782 and squashes the following commits:
56d5b7a [Patrick Wendell] Minor clean-up
44089ec [Patrick Wendell] Clean-up the TaskContext API.
ed551ce [Prashant Sharma] Fixed a typo
df261d0 [Prashant Sharma] Josh's suggestion
facf3b1 [Prashant Sharma] Fixed the mima issue.
7ecc2fe [Prashant Sharma] CR, Moved implementations to TaskContextImpl
bbd9e05 [Prashant Sharma] adding missed out files to git.
ef633f5 [Prashant Sharma] SPARK-3874, Provide stable TaskContext API
Adds some functions that were very useful when trying to track down the bug from #2656. This change also changes the tree output for query plans to include the `'` prefix to unresolved nodes and `!` prefix to nodes that refer to non-existent attributes.
Author: Michael Armbrust <michael@databricks.com>
Closes#2657 from marmbrus/debugging and squashes the following commits:
654b926 [Michael Armbrust] Clean-up, add tests
763af15 [Michael Armbrust] Add typeChecking debugging functions
8c69303 [Michael Armbrust] Add inputSet, references to QueryPlan. Improve tree string with a prefix to denote invalid or unresolved nodes.
fbeab54 [Michael Armbrust] Better toString, factories for AttributeSet.
Original problem is [SPARK-3764](https://issues.apache.org/jira/browse/SPARK-3764).
`AppendingParquetOutputFormat` uses a binary-incompatible method `context.getTaskAttemptID`.
This causes binary-incompatible of Spark itself, i.e. if Spark itself is built against hadoop-1, the artifact is for only hadoop-1, and vice versa.
Author: Takuya UESHIN <ueshin@happy-camper.st>
Closes#2638 from ueshin/issues/SPARK-3771 and squashes the following commits:
efd3784 [Takuya UESHIN] Add a comment to explain the reason to use reflection.
ec213c1 [Takuya UESHIN] Use reflection to prevent breaking binary-compatibility.
This PR adds a new rule `CheckAggregation` to the analyzer to provide better error message for non-aggregate attributes with aggregation.
Author: Cheng Lian <lian.cs.zju@gmail.com>
Closes#2774 from liancheng/non-aggregate-attr and squashes the following commits:
5246004 [Cheng Lian] Passes test suites
bf1878d [Cheng Lian] Adds checks for non-aggregate attributes with aggregation
Author: Daoyuan Wang <daoyuan.wang@intel.com>
Closes#2344 from adrian-wang/date and squashes the following commits:
f15074a [Daoyuan Wang] remove outdated lines
2038085 [Daoyuan Wang] update return type
00fe81f [Daoyuan Wang] address lian cheng's comments
0df6ea1 [Daoyuan Wang] rebase and remove simple string
bb1b1ef [Daoyuan Wang] remove failing test
aa96735 [Daoyuan Wang] not cast for same type compare
30bf48b [Daoyuan Wang] resolve rebase conflict
617d1a8 [Daoyuan Wang] add date_udf case to white list
c37e848 [Daoyuan Wang] comment update
5429212 [Daoyuan Wang] change to long
f8f219f [Daoyuan Wang] revise according to Cheng Hao
0e0a4f5 [Daoyuan Wang] minor format
4ddcb92 [Daoyuan Wang] add java api for date
0e3110e [Daoyuan Wang] try to fix timezone issue
17fda35 [Daoyuan Wang] set test list
2dfbb5b [Daoyuan Wang] support date type
Author: Reynold Xin <rxin@apache.org>
Closes#2727 from rxin/SPARK-3861-broadcast-hash-2 and squashes the following commits:
9c7b1a2 [Reynold Xin] Revert "Reuse CompactBuffer in UniqueKeyHashedRelation."
97626a1 [Reynold Xin] Reuse CompactBuffer in UniqueKeyHashedRelation.
7fcffb5 [Reynold Xin] Make UniqueKeyHashedRelation private[joins].
18eb214 [Reynold Xin] Merge branch 'SPARK-3861-broadcast-hash' into SPARK-3861-broadcast-hash-1
4b9d0c9 [Reynold Xin] UniqueKeyHashedRelation.get should return null if the value is null.
e0ebdd1 [Reynold Xin] Added a test case.
90b58c0 [Reynold Xin] [SPARK-3861] Avoid rebuilding hash tables on each partition
0c0082b [Reynold Xin] Fix line length.
cbc664c [Reynold Xin] Rename join -> joins package.
a070d44 [Reynold Xin] Fix line length in HashJoin
a39be8c [Reynold Xin] [SPARK-3857] Create a join package for various join operators.
Using `MEMORY_AND_DISK` as default storage level for in-memory table caching. Due to the in-memory columnar representation, recomputing an in-memory cached table partitions can be very expensive.
Author: Cheng Lian <lian.cs.zju@gmail.com>
Closes#2686 from liancheng/spark-3824 and squashes the following commits:
35d2ed0 [Cheng Lian] Removes extra space
1ab7967 [Cheng Lian] Reduces test data size to fit DiskStore.getBytes()
ba565f0 [Cheng Lian] Maks CachedBatch serializable
07f0204 [Cheng Lian] Sets in-memory table default storage level to MEMORY_AND_DISK
This PR is a follow up of #2590, and tries to introduce a top level SQL parser entry point for all SQL dialects supported by Spark SQL.
A top level parser `SparkSQLParser` is introduced to handle the syntaxes that all SQL dialects should recognize (e.g. `CACHE TABLE`, `UNCACHE TABLE` and `SET`, etc.). For all the syntaxes this parser doesn't recognize directly, it fallbacks to a specified function that tries to parse arbitrary input to a `LogicalPlan`. This function is typically another parser combinator like `SqlParser`. DDL syntaxes introduced in #2475 can be moved to here.
The `ExtendedHiveQlParser` now only handle Hive specific extensions.
Also took the chance to refactor/reformat `SqlParser` for better readability.
Author: Cheng Lian <lian.cs.zju@gmail.com>
Closes#2698 from liancheng/gen-sql-parser and squashes the following commits:
ceada76 [Cheng Lian] Minor styling fixes
9738934 [Cheng Lian] Minor refactoring, removes optional trailing ";" in the parser
bb2ab12 [Cheng Lian] SET property value can be empty string
ce8860b [Cheng Lian] Passes test suites
e86968e [Cheng Lian] Removes debugging code
8bcace5 [Cheng Lian] Replaces digit.+ to rep1(digit) (Scala style checking doesn't like it)
d15d54f [Cheng Lian] Unifies SQL and HiveQL parsers
This prevents it from changing during serialization, leading to corrupted results.
Author: Michael Armbrust <michael@databricks.com>
Closes#2656 from marmbrus/generateBug and squashes the following commits:
efa32eb [Michael Armbrust] Store the output of a generator in a val. This prevents it from changing during serialization.
"case when" conditional function is already supported in Spark SQL but there is no support in SqlParser. So added parser support to it.
Author : ravipesala ravindra.pesalahuawei.com
Author: ravipesala <ravindra.pesala@huawei.com>
Closes#2678 from ravipesala/SPARK-3813 and squashes the following commits:
70c75a7 [ravipesala] Fixed styles
713ea84 [ravipesala] Updated as per admin comments
709684f [ravipesala] Changed parser to support case when function.
The alias parameter is being ignored, which makes it more difficult to specify a qualifier for Generator expressions.
Author: Nathan Howell <nhowell@godaddy.com>
Closes#2721 from NathanHowell/SPARK-3858 and squashes the following commits:
8aa0f43 [Nathan Howell] [SPARK-3858][SQL] Pass the generator alias into logical plan node
This PR aims to provide a way to skip/query corrupt JSON records. To do so, we introduce an internal column to hold corrupt records (the default name is `_corrupt_record`. This name can be changed by setting the value of `spark.sql.columnNameOfCorruptRecord`). When there is a parsing error, we will put the corrupt record in its unparsed format to the internal column. Users can skip/query this column through SQL.
* To query those corrupt records
```
-- For Hive parser
SELECT `_corrupt_record`
FROM jsonTable
WHERE `_corrupt_record` IS NOT NULL
-- For our SQL parser
SELECT _corrupt_record
FROM jsonTable
WHERE _corrupt_record IS NOT NULL
```
* To skip corrupt records and query regular records
```
-- For Hive parser
SELECT field1, field2
FROM jsonTable
WHERE `_corrupt_record` IS NULL
-- For our SQL parser
SELECT field1, field2
FROM jsonTable
WHERE _corrupt_record IS NULL
```
Generally, it is not recommended to change the name of the internal column. If the name has to be changed to avoid possible name conflicts, you can use `sqlContext.setConf(SQLConf.COLUMN_NAME_OF_CORRUPT_RECORD, <new column name>)` or `sqlContext.sql(SET spark.sql.columnNameOfCorruptRecord=<new column name>)`.
Author: Yin Huai <huai@cse.ohio-state.edu>
Closes#2680 from yhuai/corruptJsonRecord and squashes the following commits:
4c9828e [Yin Huai] Merge remote-tracking branch 'upstream/master' into corruptJsonRecord
309616a [Yin Huai] Change the default name of corrupt record to "_corrupt_record".
b4a3632 [Yin Huai] Merge remote-tracking branch 'upstream/master' into corruptJsonRecord
9375ae9 [Yin Huai] Set the column name of corrupt json record back to the default one after the unit test.
ee584c0 [Yin Huai] Provide a way to query corrupt json records as unparsed strings.
In JSONRDD.scala, add 'case TimestampType' in the enforceCorrectType function and a toTimestamp function.
Author: Mike Timper <mike@aurorafeint.com>
Closes#2720 from mtimper/master and squashes the following commits:
9386ab8 [Mike Timper] Fix and tests for SPARK-3853
Author: Reynold Xin <rxin@apache.org>
Closes#2719 from rxin/sql-join-break and squashes the following commits:
0c0082b [Reynold Xin] Fix line length.
cbc664c [Reynold Xin] Rename join -> joins package.
a070d44 [Reynold Xin] Fix line length in HashJoin
a39be8c [Reynold Xin] [SPARK-3857] Create a join package for various join operators.
This PR uses JSON instead of `toString` to serialize `DataType`s. The latter is not only hard to parse but also flaky in many cases.
Since we already write schema information to Parquet metadata in the old style, we have to reserve the old `DataType` parser and ensure downward compatibility. The old parser is now renamed to `CaseClassStringParser` and moved into `object DataType`.
JoshRosen davies Please help review PySpark related changes, thanks!
Author: Cheng Lian <lian.cs.zju@gmail.com>
Closes#2563 from liancheng/datatype-to-json and squashes the following commits:
fc92eb3 [Cheng Lian] Reverts debugging code, simplifies primitive type JSON representation
438c75f [Cheng Lian] Refactors PySpark DataType JSON SerDe per comments
6b6387b [Cheng Lian] Removes debugging code
6a3ee3a [Cheng Lian] Addresses per review comments
dc158b5 [Cheng Lian] Addresses PEP8 issues
99ab4ee [Cheng Lian] Adds compatibility est case for Parquet type conversion
a983a6c [Cheng Lian] Adds PySpark support
f608c6e [Cheng Lian] De/serializes DataType objects from/to JSON
If we write the filter which is always FALSE like
SELECT * from person WHERE FALSE;
200 tasks will run. I think, 1 task is enough.
And current optimizer cannot optimize the case NOT is duplicated like
SELECT * from person WHERE NOT ( NOT (age > 30));
The filter rule above should be simplified
Author: Kousuke Saruta <sarutak@oss.nttdata.co.jp>
Closes#2692 from sarutak/SPARK-3831 and squashes the following commits:
25f3e20 [Kousuke Saruta] Merge branch 'master' of git://git.apache.org/spark into SPARK-3831
23c750c [Kousuke Saruta] Improved unsupported predicate test case
a11b9f3 [Kousuke Saruta] Modified NOT predicate test case in PartitionBatchPruningSuite
8ea872b [Kousuke Saruta] Fixed the number of tasks when the data of LocalRelation is empty.
Although lazy caching for in-memory table seems consistent with the `RDD.cache()` API, it's relatively confusing for users who mainly work with SQL and not familiar with Spark internals. The `CACHE TABLE t; SELECT COUNT(*) FROM t;` pattern is also commonly seen just to ensure predictable performance.
This PR makes both the `CACHE TABLE t [AS SELECT ...]` statement and the `SQLContext.cacheTable()` API eager by default, and adds a new `CACHE LAZY TABLE t [AS SELECT ...]` syntax to provide lazy in-memory table caching.
Also, took the chance to make some refactoring: `CacheCommand` and `CacheTableAsSelectCommand` are now merged and renamed to `CacheTableCommand` since the former is strictly a special case of the latter. A new `UncacheTableCommand` is added for the `UNCACHE TABLE t` statement.
Author: Cheng Lian <lian.cs.zju@gmail.com>
Closes#2513 from liancheng/eager-caching and squashes the following commits:
fe92287 [Cheng Lian] Makes table caching eager by default and adds syntax for lazy caching
_Also addresses: SPARK-1671, SPARK-1379 and SPARK-3641_
This PR introduces a new trait, `CacheManger`, which replaces the previous temporary table based caching system. Instead of creating a temporary table that shadows an existing table with and equivalent cached representation, the cached manager maintains a separate list of logical plans and their cached data. After optimization, this list is searched for any matching plan fragments. When a matching plan fragment is found it is replaced with the cached data.
There are several advantages to this approach:
- Calling .cache() on a SchemaRDD now works as you would expect, and uses the more efficient columnar representation.
- Its now possible to provide a list of temporary tables, without having to decide if a given table is actually just a cached persistent table. (To be done in a follow-up PR)
- In some cases it is possible that cached data will be used, even if a cached table was not explicitly requested. This is because we now look at the logical structure instead of the table name.
- We now correctly invalidate when data is inserted into a hive table.
Author: Michael Armbrust <michael@databricks.com>
Closes#2501 from marmbrus/caching and squashes the following commits:
63fbc2c [Michael Armbrust] Merge remote-tracking branch 'origin/master' into caching.
0ea889e [Michael Armbrust] Address comments.
1e23287 [Michael Armbrust] Add support for cache invalidation for hive inserts.
65ed04a [Michael Armbrust] fix tests.
bdf9a3f [Michael Armbrust] Merge remote-tracking branch 'origin/master' into caching
b4b77f2 [Michael Armbrust] Address comments
6923c9d [Michael Armbrust] More comments / tests
80f26ac [Michael Armbrust] First draft of improved semantics for Spark SQL caching.
The following code gives error.
```
sqlContext.registerFunction("len", (s: String) => s.length)
sqlContext.sql("select len(foo) as a, count(1) from t1 group by len(foo)").collect()
```
Because SQl parser creates the aliases to the functions in grouping expressions with generated alias names. So if user gives the alias names to the functions inside projection then it does not match the generated alias name of grouping expression.
This kind of queries are working in Hive.
So the fix I have given that if user provides alias to the function in projection then don't generate alias in grouping expression,use the same alias.
Author: ravipesala <ravindra.pesala@huawei.com>
Closes#2511 from ravipesala/SPARK-3371 and squashes the following commits:
9fb973f [ravipesala] Removed aliases to grouping expressions.
f8ace79 [ravipesala] Fixed the testcase issue
bad2fd0 [ravipesala] SPARK-3371 : Fixed Renaming a function expression with group by gives error
Considering `Command.executeCollect()` simply delegates to `Command.sideEffectResult`, we no longer need to leave the latter `protected[sql]`.
Author: Cheng Lian <lian.cs.zju@gmail.com>
Closes#2431 from liancheng/narrow-scope and squashes the following commits:
1bfc16a [Cheng Lian] Made Command.sideEffectResult protected
BinaryType is derived from NativeType and added Ordering support.
Author: Venkata Ramana G <ramana.gollamudihuawei.com>
Author: Venkata Ramana Gollamudi <ramana.gollamudi@huawei.com>
Closes#2617 from gvramana/binarytype_sort and squashes the following commits:
1cf26f3 [Venkata Ramana Gollamudi] Supported Sorting of BinaryType
Thread names are useful for correlating failures.
Author: Reynold Xin <rxin@apache.org>
Closes#2600 from rxin/log4j and squashes the following commits:
83ffe88 [Reynold Xin] [SPARK-3748] Log thread name in unit test logs
Author: Reynold Xin <rxin@apache.org>
Closes#2560 from rxin/TaskContext and squashes the following commits:
9eff95a [Reynold Xin] [SPARK-3543] remaining cleanup work.
This will allow us to take advantage of things like the spark.defaults file.
Author: Michael Armbrust <michael@databricks.com>
Closes#2493 from marmbrus/copySparkConf and squashes the following commits:
0bd1377 [Michael Armbrust] Copy SQL configuration from SparkConf when a SQLContext is created.
It returns null metadata from parquet if querying on empty parquet file while calculating splits.So added null check and returns the empty splits.
Author : ravipesala ravindra.pesalahuawei.com
Author: ravipesala <ravindra.pesala@huawei.com>
Closes#2456 from ravipesala/SPARK-3536 and squashes the following commits:
1e81a50 [ravipesala] Fixed the issue when querying on empty parquet file.
**This PR introduces a subtle change in semantics for HiveContext when using the results in Python or Scala. Specifically, while resolution remains case insensitive, it is now case preserving.**
_This PR is a follow up to #2293 (and to a lesser extent #2262#2334)._
In #2293 the catalog was changed to store analyzed logical plans instead of unresolved ones. While this change fixed the reported bug (which was caused by yet another instance of us forgetting to put in a `LowerCaseSchema` operator) it had the consequence of breaking assumptions made by `MultiInstanceRelation`. Specifically, we can't replace swap out leaf operators in a tree without rewriting changed expression ids (which happens when you self join the same RDD that has been registered as a temp table).
In this PR, I instead remove the need to insert `LowerCaseSchema` operators at all, by moving the concern of matching up identifiers completely into analysis. Doing so allows the test cases from both #2293 and #2262 to pass at the same time (and likely fixes a slew of other "unknown unknown" bugs).
While it is rolled back in this PR, storing the analyzed plan might actually be a good idea. For instance, it is kind of confusing if you register a temporary table, change the case sensitivity of resolution and now you can't query that table anymore. This can be addressed in a follow up PR.
Follow-ups:
- Configurable case sensitivity
- Consider storing analyzed plans for temp tables
Author: Michael Armbrust <michael@databricks.com>
Closes#2382 from marmbrus/lowercase and squashes the following commits:
c21171e [Michael Armbrust] Ensure the resolver is used for field lookups and ensure that case insensitive resolution is still case preserving.
d4320f1 [Michael Armbrust] Merge remote-tracking branch 'origin/master' into lowercase
2de881e [Michael Armbrust] Address comments.
219805a [Michael Armbrust] style
5b93711 [Michael Armbrust] Replace LowerCaseSchema with Resolver.
This helps to replace shuffled hash joins with broadcast hash joins in some cases.
Author: Cheng Lian <lian.cs.zju@gmail.com>
Closes#2468 from liancheng/more-stats and squashes the following commits:
32687dc [Cheng Lian] Moved the test case to PlannerSuite
5595a91 [Cheng Lian] Removes debugging code
73faf69 [Cheng Lian] Test case for auto choosing broadcast hash join
f30fe1d [Cheng Lian] Adds sizeInBytes estimation for Limit when all output types are native types
Author: Sandy Ryza <sandy@cloudera.com>
Closes#2460 from sryza/sandy-spark-3605 and squashes the following commits:
09d940b [Sandy Ryza] SPARK-3605. Fix typo in SchemaRDD.
This feature allows user to add cache table from the select query.
Example : ```CACHE TABLE testCacheTable AS SELECT * FROM TEST_TABLE```
Spark takes this type of SQL as command and it does lazy caching just like ```SQLContext.cacheTable```, ```CACHE TABLE <name>``` does.
It can be executed from both SQLContext and HiveContext.
Recreated the pull request after rebasing with master.And fixed all the comments raised in previous pull requests.
https://github.com/apache/spark/pull/2381https://github.com/apache/spark/pull/2390
Author : ravipesala ravindra.pesalahuawei.com
Author: ravipesala <ravindra.pesala@huawei.com>
Closes#2397 from ravipesala/SPARK-2594 and squashes the following commits:
a5f0beb [ravipesala] Simplified the code as per Admin comment.
8059cd2 [ravipesala] Changed the behaviour from eager caching to lazy caching.
d6e469d [ravipesala] Code review comments by Admin are handled.
c18aa38 [ravipesala] Merge remote-tracking branch 'remotes/ravipesala/Add-Cache-table-as' into SPARK-2594
394d5ca [ravipesala] Changed style
fb1759b [ravipesala] Updated as per Admin comments
8c9993c [ravipesala] Changed the style
d8b37b2 [ravipesala] Updated as per the comments by Admin
bc0bffc [ravipesala] Merge remote-tracking branch 'ravipesala/Add-Cache-table-as' into Add-Cache-table-as
e3265d0 [ravipesala] Updated the code as per the comments by Admin in pull request.
724b9db [ravipesala] Changed style
aaf5b59 [ravipesala] Added comment
dc33895 [ravipesala] Updated parser to support add cache table command
b5276b2 [ravipesala] Updated parser to support add cache table command
eebc0c1 [ravipesala] Add CACHE TABLE <name> AS SELECT ...
6758f80 [ravipesala] Changed style
7459ce3 [ravipesala] Added comment
13c8e27 [ravipesala] Updated parser to support add cache table command
4e858d8 [ravipesala] Updated parser to support add cache table command
b803fc8 [ravipesala] Add CACHE TABLE <name> AS SELECT ...
SchemaRDD overrides RDD functions, including collect, count, and take, with optimized versions making use of the query optimizer. The java and python interface classes wrapping SchemaRDD need to ensure the optimized versions are called as well. This patch overrides relevant calls in the python and java interfaces with optimized versions.
Adds a new Row serialization pathway between python and java, based on JList[Array[Byte]] versus the existing RDD[Array[Byte]]. I wasn’t overjoyed about doing this, but I noticed that some QueryPlans implement optimizations in executeCollect(), which outputs an Array[Row] rather than the typical RDD[Row] that can be shipped to python using the existing serialization code. To me it made sense to ship the Array[Row] over to python directly instead of converting it back to an RDD[Row] just for the purpose of sending the Rows to python using the existing serialization code.
Author: Aaron Staple <aaron.staple@gmail.com>
Closes#1592 from staple/SPARK-2314 and squashes the following commits:
89ff550 [Aaron Staple] Merge with master.
6bb7b6c [Aaron Staple] Fix typo.
b56d0ac [Aaron Staple] [SPARK-2314][SQL] Override count in JavaSchemaRDD, forwarding to SchemaRDD's count.
0fc9d40 [Aaron Staple] Fix comment typos.
f03cdfa [Aaron Staple] [SPARK-2314][SQL] Override collect and take in sql.py, forwarding to SchemaRDD's collect.
This PR aims to support reading top level JSON arrays and take every element in such an array as a row (an empty array will not generate a row).
JIRA: https://issues.apache.org/jira/browse/SPARK-3308
Author: Yin Huai <huai@cse.ohio-state.edu>
Closes#2400 from yhuai/SPARK-3308 and squashes the following commits:
990077a [Yin Huai] Handle top level JSON arrays.
Author: Cheng Hao <hao.cheng@intel.com>
Closes#2392 from chenghao-intel/trim and squashes the following commits:
e52024f [Cheng Hao] trim the string message
Author: Michael Armbrust <michael@databricks.com>
Closes#2164 from marmbrus/shufflePartitions and squashes the following commits:
0da1e8c [Michael Armbrust] test hax
ef2d985 [Michael Armbrust] more test hacks.
2dabae3 [Michael Armbrust] more test fixes
0bdbf21 [Michael Armbrust] Make parquet tests less order dependent
b42eeab [Michael Armbrust] increase test parallelism
80453d5 [Michael Armbrust] Decrease partitions when testing
This is a major refactoring of the in-memory columnar storage implementation, aims to eliminate boxing costs from critical paths (building/accessing column buffers) as much as possible. The basic idea is to refactor all major interfaces into a row-based form and use them together with `SpecificMutableRow`. The difficult part is how to adapt all compression schemes, esp. `RunLengthEncoding` and `DictionaryEncoding`, to this design. Since in-memory compression is disabled by default for now, and this PR should be strictly better than before no matter in-memory compression is enabled or not, maybe I'll finish that part in another PR.
**UPDATE** This PR also took the chance to optimize `HiveTableScan` by
1. leveraging `SpecificMutableRow` to avoid boxing cost, and
1. building specific `Writable` unwrapper functions a head of time to avoid per row pattern matching and branching costs.
TODO
- [x] Benchmark
- [ ] ~~Eliminate boxing costs in `RunLengthEncoding`~~ (left to future PRs)
- [ ] ~~Eliminate boxing costs in `DictionaryEncoding` (seems not easy to do without specializing `DictionaryEncoding` for every supported column type)~~ (left to future PRs)
## Micro benchmark
The benchmark uses a 10 million line CSV table consists of bytes, shorts, integers, longs, floats and doubles, measures the time to build the in-memory version of this table, and the time to scan the whole in-memory table.
Benchmark code can be found [here](https://gist.github.com/liancheng/fe70a148de82e77bd2c8#file-hivetablescanbenchmark-scala). Script used to generate the input table can be found [here](https://gist.github.com/liancheng/fe70a148de82e77bd2c8#file-tablegen-scala).
Speedup:
- Hive table scanning + column buffer building: **18.74%**
The original benchmark uses 1K as in-memory batch size, when increased to 10K, it can be 28.32% faster.
- In-memory table scanning: **7.95%**
Before:
| Building | Scanning
------- | -------- | --------
1 | 16472 | 525
2 | 16168 | 530
3 | 16386 | 529
4 | 16184 | 538
5 | 16209 | 521
Average | 16283.8 | 528.6
After:
| Building | Scanning
------- | -------- | --------
1 | 13124 | 458
2 | 13260 | 529
3 | 12981 | 463
4 | 13214 | 483
5 | 13583 | 500
Average | 13232.4 | 486.6
Author: Cheng Lian <lian.cs.zju@gmail.com>
Closes#2327 from liancheng/prevent-boxing/unboxing and squashes the following commits:
4419fe4 [Cheng Lian] Addressing comments
e5d2cf2 [Cheng Lian] Bug fix: should call setNullAt when field value is null to avoid NPE
8b8552b [Cheng Lian] Only checks for partition batch pruning flag once
489f97b [Cheng Lian] Bug fix: TableReader.fillObject uses wrong ordinals
97bbc4e [Cheng Lian] Optimizes hive.TableReader by by providing specific Writable unwrappers a head of time
3dc1f94 [Cheng Lian] Minor changes to eliminate row object creation
5b39cb9 [Cheng Lian] Lowers log level of compression scheme details
f2a7890 [Cheng Lian] Use SpecificMutableRow in InMemoryColumnarTableScan to avoid boxing
9cf30b0 [Cheng Lian] Added row based ColumnType.append/extract
456c366 [Cheng Lian] Made compression decoder row based
edac3cd [Cheng Lian] Makes ColumnAccessor.extractSingle row based
8216936 [Cheng Lian] Removes boxing cost in IntDelta and LongDelta by providing specialized implementations
b70d519 [Cheng Lian] Made some in-memory columnar storage interfaces row-based
This PR aims to correctly handle JSON arrays in the type of `ArrayType(...(ArrayType(StructType)))`.
JIRA: https://issues.apache.org/jira/browse/SPARK-3390.
Author: Yin Huai <huai@cse.ohio-state.edu>
Closes#2364 from yhuai/SPARK-3390 and squashes the following commits:
46db418 [Yin Huai] Handle JSON arrays in the type of ArrayType(...(ArrayType(StructType))).
Author: Cheng Hao <hao.cheng@intel.com>
Closes#1846 from chenghao-intel/ctas and squashes the following commits:
56a0578 [Cheng Hao] remove the unused imports
9a57abc [Cheng Hao] Avoid table creation in logical plan analyzing
LogicalPlan contains a ‘resolved’ attribute indicating that all of its execution requirements have been resolved. This attribute is not checked before query execution. The analyzer contains a step to check that all Expressions are resolved, but this is not equivalent to checking all LogicalPlans. In particular, the Union plan’s implementation of ‘resolved’ verifies that the types of its children’s columns are compatible. Because the analyzer does not check that a Union plan is resolved, it is possible to execute a Union plan that outputs different types in the same column. See SPARK-2781 for an example.
This patch adds two checks to the analyzer’s CheckResolution rule. First, each logical plan is checked to see if it is not resolved despite its children being resolved. This allows the ‘problem’ unresolved plan to be included in the TreeNodeException for reporting. Then as a backstop the root plan is checked to see if it is resolved, which recursively checks that the entire plan tree is resolved. Note that the resolved attribute is implemented recursively, and this patch also explicitly checks the resolved attribute on each logical plan in the tree. I assume the query plan trees will not be large enough for this redundant checking to meaningfully impact performance.
Because this patch starts validating that LogicalPlans are resolved before execution, I had to fix some cases where unresolved plans were passing through the analyzer as part of the implementation of the hive query system. In particular, HiveContext applies the CreateTables and PreInsertionCasts, and ExtractPythonUdfs rules manually after the analyzer runs. I moved these rules to the analyzer stage (for hive queries only), in the process completing a code TODO indicating the rules should be moved to the analyzer.
It’s worth noting that moving the CreateTables rule means introducing an analyzer rule with a significant side effect - in this case the side effect is creating a hive table. The rule will only attempt to create a table once even if its batch is executed multiple times, because it converts the InsertIntoCreatedTable plan it matches against into an InsertIntoTable. Additionally, these hive rules must be added to the Resolution batch rather than as a separate batch because hive rules rules may be needed to resolve non-root nodes, leaving the root to be resolved on a subsequent batch iteration. For example, the hive compatibility test auto_smb_mapjoin_14, and others, make use of a query plan where the root is a Union and its children are each a hive InsertIntoTable.
Mixing the custom hive rules with standard analyzer rules initially resulted in an additional failure because of policy differences between spark sql and hive when casting a boolean to a string. Hive casts booleans to strings as “true” / “false” while spark sql casts booleans to strings as “1” / “0” (causing the cast1.q test to fail). This behavior is a result of the BooleanCasts rule in HiveTypeCoercion.scala, and from looking at the implementation of BooleanCasts I think converting to to “1”/“0” is potentially a programming mistake. (If the BooleanCasts rule is disabled, casting produces “true”/“false” instead.) I believe “true” / “false” should be the behavior for spark sql - I changed the behavior so bools are converted to “true”/“false” to be consistent with hive, and none of the existing spark tests failed.
Finally, in some initial testing with hive it appears that an implicit type coercion of boolean to string results in a lowercase string, e.g. CONCAT( TRUE, “” ) -> “true” while an explicit cast produces an all caps string, e.g. CAST( TRUE AS STRING ) -> “TRUE”. The change I’ve made just converts to lowercase strings in all cases. I believe it is at least more correct than the existing spark sql implementation where all Cast expressions become “1” / “0”.
Author: Aaron Staple <aaron.staple@gmail.com>
Closes#1706 from staple/SPARK-2781 and squashes the following commits:
32683c4 [Aaron Staple] Fix compilation failure due to merge.
7c77fda [Aaron Staple] Move ExtractPythonUdfs to Analyzer's extendedRules in HiveContext.
d49bfb3 [Aaron Staple] Address review comments.
915b690 [Aaron Staple] Fix merge issue causing compilation failure.
701dcd2 [Aaron Staple] [SPARK-2781][SQL] Check resolution of LogicalPlans in Analyzer.
