* Various changes and fixes to UNNEST.
Native changes:
1) UnnestDataSource: Replace "column" and "outputName" with "virtualColumn".
This enables pushing expressions into the datasource. This in turn
allows us to do the next thing...
2) UnnestStorageAdapter: Logically apply query-level filters and virtual
columns after the unnest operation. (Physically, filters are pulled up,
when possible.) This is beneficial because it allows filters and
virtual columns to reference the unnested column, and because it is
consistent with how the join datasource works.
3) Various documentation updates, including declaring "unnest" as an
experimental feature for now.
SQL changes:
1) Rename DruidUnnestRel (& Rule) to DruidUnnestRel (& Rule). The rel
is simplified: it only handles the UNNEST part of a correlated join.
Constant UNNESTs are handled with regular inline rels.
2) Rework DruidCorrelateUnnestRule to focus on pulling Projects from
the left side up above the Correlate. New test testUnnestTwice verifies
that this works even when two UNNESTs are stacked on the same table.
3) Include ProjectCorrelateTransposeRule from Calcite to encourage
pushing mappings down below the left-hand side of the Correlate.
4) Add a new CorrelateFilterLTransposeRule and CorrelateFilterRTransposeRule
to handle pulling Filters up above the Correlate. New tests
testUnnestWithFiltersOutside and testUnnestTwiceWithFilters verify
this behavior.
5) Require a context feature flag for SQL UNNEST, since it's undocumented.
As part of this, also cleaned up how we handle feature flags in SQL.
They're now hooked into EngineFeatures, which is useful because not
all engines support all features.
With SuperSorter using the PartitionedOutputChannels for sorting, it might OOM on inputs of reasonable size because the channel consists of both the writable frame channel and the frame allocator, both of which are not required once the output channel has been written to.
This change adds a readOnly to the output channel which contains only the readable channel, due to which unnecessary memory references to the writable channel and the memory allocator are lost once the output channel has been written to, preventing the OOM.
* Window planning: use collation traits, improve subquery logic.
SQL changes:
1) Attach RelCollation (sorting) trait to any PartialDruidQuery
that ends in AGGREGATE or AGGREGATE_PROJECT. This allows planning to
take advantage of the fact that Druid sorts by dimensions when
doing aggregations.
2) Windowing: inspect RelCollation trait from input, and insert naiveSort
if, and only if, necessary.
3) Windowing: add support for Project after Window, when the Project
is a simple mapping. Helps eliminate subqueries.
4) DruidRules: update logic for considering subqueries to reflect that
subqueries are not required to be GroupBys, and that we have a bunch
of new Stages now. With all of this evolution that has happened, the
old logic didn't quite make sense.
Native changes:
1) Use merge sort (stable) rather than quicksort when sorting
RowsAndColumns. Makes it easier to write test cases for plans that
involve re-sorting the data.
* Changes from review.
* Mark the bad test as failing.
* Additional update.
* Fix failingTest.
* Fix tests.
* Mark a var final.
* Improve memory efficiency of WrappedRoaringBitmap.
Two changes:
1) Use an int[] for sizes 4 or below.
2) Remove the boolean compressRunOnSerialization. Doesn't save much
space, but it does save a little, and it isn't adding a ton of value
to have it be configurable. It was originally configurable in case
anything broke when enabling it, but it's been a while and nothing
has broken.
* Slight adjustment.
* Adjust for inspection.
* Updates.
* Update snaps.
* Update test.
* Adjust test.
* Fix snaps.
* use custom case operator conversion instead of direct operator conversion, to produce native nvl expression for SQL NVL and 2 argument COALESCE, and add optimization for certain case filters from coalesce and nvl statements
* Sort-merge join and hash shuffles for MSQ.
The main changes are in the processing, multi-stage-query, and sql modules.
processing module:
1) Rename SortColumn to KeyColumn, replace boolean descending with KeyOrder.
This makes it nicer to model hash keys, which use KeyOrder.NONE.
2) Add nullability checkers to the FieldReader interface, and an
"isPartiallyNullKey" method to FrameComparisonWidget. The join
processor uses this to detect null keys.
3) Add WritableFrameChannel.isClosed and OutputChannel.isReadableChannelReady
so callers can tell which OutputChannels are ready for reading and which
aren't.
4) Specialize FrameProcessors.makeCursor to return FrameCursor, a random-access
implementation. The join processor uses this to rewind when it needs to
replay a set of rows with a particular key.
5) Add MemoryAllocatorFactory, which is embedded inside FrameWriterFactory
instead of a particular MemoryAllocator. This allows FrameWriterFactory
to be shared in more scenarios.
multi-stage-query module:
1) ShuffleSpec: Add hash-based shuffles. New enum ShuffleKind helps callers
figure out what kind of shuffle is happening. The change from SortColumn
to KeyColumn allows ClusterBy to be used for both hash-based and sort-based
shuffling.
2) WorkerImpl: Add ability to handle hash-based shuffles. Refactor the logic
to be more readable by moving the work-order-running code to the inner
class RunWorkOrder, and the shuffle-pipeline-building code to the inner
class ShufflePipelineBuilder.
3) Add SortMergeJoinFrameProcessor and factory.
4) WorkerMemoryParameters: Adjust logic to reserve space for output frames
for hash partitioning. (We need one frame per partition.)
sql module:
1) Add sqlJoinAlgorithm context parameter; can be "broadcast" or
"sortMerge". With native, it must always be "broadcast", or it's a
validation error. MSQ supports both. Default is "broadcast" in
both engines.
2) Validate that MSQs do not use broadcast join with RIGHT or FULL join,
as results are not correct for broadcast join with those types. Allow
this in native for two reasons: legacy (the docs caution against it,
but it's always been allowed), and the fact that it actually *does*
generate correct results in native when the join is processed on the
Broker. It is much less likely that MSQ will plan in such a way that
generates correct results.
3) Remove subquery penalty in DruidJoinQueryRel when using sort-merge
join, because subqueries are always required, so there's no reason
to penalize them.
4) Move previously-disabled join reordering and manipulation rules to
FANCY_JOIN_RULES, and enable them when using sort-merge join. Helps
get to better plans where projections and filters are pushed down.
* Work around compiler problem.
* Updates from static analysis.
* Fix @param tag.
* Fix declared exception.
* Fix spelling.
* Minor adjustments.
* wip
* Merge fixups
* fixes
* Fix CalciteSelectQueryMSQTest
* Empty keys are sortable.
* Address comments from code review. Rename mux -> mix.
* Restore inspection config.
* Restore original doc.
* Reorder imports.
* Adjustments
* Fix.
* Fix imports.
* Adjustments from review.
* Update header.
* Adjust docs.
This function is notorious for causing memory exhaustion and excessive
CPU usage; so much so that it was valuable to work around it in the
SQL planner in #13206. Hopefully, a warning comment will encourage
developers to stay away and come up with solutions that do not involve
computing all possible buckets.
You can now do the following operations with TupleSketches in Post Aggregation Step
Get the Sketch Output as Base64 String
Provide a constant Tuple Sketch in post-aggregation step that can be used in Set Operations
Get the Estimated Value(Sum) of Summary/Metrics Objects associated with Tuple Sketch
The FiniteFirehoseFactory and InputRowParser classes were deprecated in 0.17.0 (#8823) in favor of InputSource & InputFormat. This PR removes the FiniteFirehoseFactory and all its implementations along with classes solely used by them like Fetcher (Used by PrefetchableTextFilesFirehoseFactory). Refactors classes including tests using FiniteFirehoseFactory to use InputSource instead.
Removing InputRowParser may not be as trivial as many classes that aren't deprecated depends on it (with no alternatives), like EventReceiverFirehoseFactory. Hence FirehoseFactory, EventReceiverFirehoseFactory, and Firehose are marked deprecated.
* move numeric null value coercion out of expression processing engine
* add ExprEval.valueOrDefault() to allow consumers to automatically coerce to default values
* rename Expr.buildVectorized as Expr.asVectorProcessor more consistent naming with Function and ApplyFunction; javadocs for some stuff
* Speed up composite key joins on IndexedTable.
Prior to this patch, IndexedTable indexes are sorted IntList. This works
great when we have a single-column join key: we simply retrieve the list
and we know what rows match. However, when we have a composite key, we
need to merge the sorted lists. This is inefficient when one is very dense
and others are very sparse.
This patch switches from sorted IntList to IntSortedSet, and changes
to the following intersection algorithm:
1) Initialize the intersection set to the smallest matching set from the
various parts of the composite key.
2) For each element in that smallest set, check other sets for that element.
If any do *not* include it, then remove the element from the intersection
set.
This way, complexity scales with the size of the smallest set, not the
largest one.
* RangeIntSet stuff.
* merge druid-core, extendedset, and druid-hll into druid-processing to simplify everything
* fix poms and license stuff
* mockito is evil
* allow reset of JvmUtils RuntimeInfo if tests used static injection to override
* fix array_agg to work with complex types and bugs with expression aggregator complex array handling
* more consistent handling of array expressions, numeric arrays more consistently honor druid.generic.useDefaultValueForNull, fix array_ordinal sql output type
* Allow users to add additional metadata to ingestion metrics
When submitting an ingestion spec, users may pass a map of metadata
in the ingestion spec config that will be added to ingestion metrics.
This will make it possible for operators to tag metrics with other
metadata that doesn't necessarily line up with the existing tags
like taskId.
Druid clusters that ingest these metrics can take advantage of the
nested data columns feature to process this additional metadata.
* rename to tags
* docs
* tests
* fix test
* make code cov happy
* checkstyle
changes:
* modified druid schema column type compution to special case COMPLEX<json> handling to choose COMPLEX<json> if any column in any segment is COMPLEX<json>
* NestedFieldVirtualColumn can now work correctly on any type of column, returning either a column selector if a root path, or nil selector if not
* fixed a random bug with NilVectorSelector when using a vector size larger than the default and druid.generic.useDefaultValueForNull=false would have the nulls vector set to all false instead of true
* fixed an overly aggressive check in ExprEval.ofType when handling complex types which would try to treat any string as base64 without gracefully falling back if it was not in fact base64 encoded, along with special handling for complex<json>
* added ExpressionVectorSelectors.castValueSelectorToObject and ExpressionVectorSelectors.castObjectSelectorToNumeric as convience methods to cast vector selectors using cast expressions without the trouble of constructing an expression. the polymorphic nature of the non-vectorized engine (and significantly larger overhead of non-vectorized expression processing) made adding similar methods for non-vectorized selectors less attractive and so have not been added at this time
* fix inconsistency between nested column indexer and serializer in handling values (coerce non primitive and non arrays of primitives using asString)
* ExprEval best effort mode now handles byte[] as string
* added test for ExprEval.bestEffortOf, and add missing conversion cases that tests uncovered
* more tests more better
* Fallback virtual column
This virtual columns enables falling back to another column if
the original column doesn't exist. This is useful when doing
column migrations and you have some old data with column X,
new data with column Y and you want to use Y if it exists, X
otherwise so that you can run a consistent query against all of
the data.
* Adjust Operators to be Pausable
This enables "merge" style operations that
combine multiple streams.
This change includes a naive implementation
of one such merge operator just to provide
concrete evidence that the refactoring is
effective.
* adds the SQL component of the native unnest functionality in Druid to unnest SQL queries on a table dimension, virtual column or a constant array and convert them into native Druid queries
* unnest in SQL is implemented as a combination of Correlate (the comma join part) and Uncollect (the unnest part)
* discover nested columns when using nested column indexer for schemaless
* move useNestedColumnIndexerForSchemaDiscovery from AppendableIndexSpec to DimensionsSpec
* Semantic Implementations for ArrayListRAC
This adds implementations of semantic interfaces
to optimize (eliminate object creation) the
window processing on top of an ArrayListSegment.
Tests are also added to cover the interplay
between the semantic interfaces that are expected
for this use case
* Kinesis: More robust default fetch settings.
1) Default recordsPerFetch and recordBufferSize based on available memory
rather than using hardcoded numbers. For this, we need an estimate
of record size. Use 10 KB for regular records and 1 MB for aggregated
records. With 1 GB heaps, 2 processors per task, and nonaggregated
records, recordBufferSize comes out to the same as the old
default (10000), and recordsPerFetch comes out slightly lower (1250
instead of 4000).
2) Default maxRecordsPerPoll based on whether records are aggregated
or not (100 if not aggregated, 1 if aggregated). Prior default was 100.
3) Default fetchThreads based on processors divided by task count on
Indexers, rather than overall processor count.
4) Additionally clean up the serialized JSON a bit by adding various
JsonInclude annotations.
* Updates for tests.
* Additional important verify.
* single typed "root" only nested columns now mimic "regular" columns of those types
* incremental index can now use nested column indexer instead of string indexer for discovered columns
* Addition of NaiveSortMaker and Default implementation
Add the NaiveSortMaker which makes a sorter
object and a default implementation of the
interface.
This also allows us to plan multiple different window
definitions on the same query.
* Validate response headers and fix exception logging
A class of QueryException were throwing away their
causes making it really hard to determine what's
going wrong when something goes wrong in the SQL
planner specifically. Fix that and adjust tests
to do more validation of response headers as well.
We allow 404s and 307s to be returned even without
authorization validated, but others get converted to 403
* Unify the handling of HTTP between SQL and Native
The SqlResource and QueryResource have been
using independent logic for things like error
handling and response context stuff. This
became abundantly clear and painful during a
change I was making for Window Functions, so
I unified them into using the same code for
walking the response and serializing it.
Things are still not perfectly unified (it would
be the absolute best if the SqlResource just
took SQL, planned it and then delegated the
query run entirely to the QueryResource), but
this refactor doesn't take that fully on.
The new code leverages async query processing
from our jetty container, the different
interaction model with the Resource means that
a lot of tests had to be adjusted to align with
the async query model. The semantics of the
tests remain the same with one exception: the
SqlResource used to not log requests that failed
authorization checks, now it does.
This PR expands `StringDimensionIndexer` to handle conversion of `byte[]` to base64 encoded strings, rather than the current behavior of calling java `toString`.
This issue was uncovered by a regression of sorts introduced by #13519, which updated the protobuf extension to directly convert stuff to java types, resulting in `bytes` typed values being converted as `byte[]` instead of a base64 string which the previous JSON based conversion created. While outputting `byte[]` is more consistent with other input formats, and preferable when the bytes can be consumed directly (such as complex types serde), when fed to a `StringDimensionIndexer`, it resulted in an ugly java `toString` because `processRowValsToUnsortedEncodedKeyComponent` is fed the output of `row.getRaw(..)`. Converting `byte[]` to a base64 string within `StringDimensionIndexer` is consistent with the behavior of calling `row.getDimension(..)` which does do this coercion (and why many tests on binary types appeared to be doing the expected thing).
I added some protobuf `bytes` tests, but they don't really hit the new `StringDimensionIndexer` behavior because they operate on the `InputRow` directly, and call `getDimension` to validate stuff. The parser based version still uses the old conversion mechanisms, so when not using a flattener incorrectly calls `toString` on the `ByteString`. I have encoded this behavior in the test for now, if we either update the parser to use the new flattener or just .. remove parsers we can remove this test stuff.
* bump nested column format version
changes:
* nested field files are now named by their position in field paths list, rather than directly by the path itself. this fixes issues with valid json properties with commas and newlines breaking the csv file meta.smoosh
* update StructuredDataProcessor to deal in NestedPathPart to be consistent with other abstract path handling rather than building JQ syntax strings directly
* add v3 format segment and test
This commit fixes a bug with nested column "value set" indexes caused by not properly
validating that the globalId looked up for value is present in the global dictionary prior to
looking it up in the local dictionary, which when "adjusting" the global ids for value type
can cause incorrect selection of value indexes.
To use an example of a variant typed nested column with 3 values `["1", null, -2]`.
The string dictionary is `[null, "1"]`, the long dictionary is `[-2]` and our local dictionary is `[0, 1, 2]`.
The code for variant typed indexes checks if the value is present in all global dictionaries
and returns indexes for all matches. So in this case, we first lookup "1" in the string dictionary,
find it at global id 1, all is good. Now, we check the long dictionary for `1`, which due to
`-(insertionpoint + 1)` gives us `-(1 + 2) = -2`. Since the global id space is actually stacked
dictionaries, global ids for long and double values must be "adjusted" by the size of string
dictionary, and size of string + size of long for doubles.
Prior to this patch we were not checking that the globalId is 0 or larger, we then immediately
looked up the `localDictionary.indexOf(-2 + adjustLong) = localDictionary.indexOf(-2 + 2) = localDictionary.indexOf(0)` ... which is an actual value contained in the dictionary! The fix is
to skip the longs completely since there were no global matches.
On to doubles, `-(insertionPoint + 1)` gives us `-(0 + 1) = -1`. The double adjust value is '3'
since 2 strings and 1 long, so `localDictionary.indexOf(-1 + 3)` = `localDictionary.indexOf(2)`
which is also a real value in our local dictionary that is definitely not '1'.
So in this one case, looking for '1' incorrectly ended up matching every row.
* Support Framing for Window Aggregations
This adds support for framing over ROWS
for window aggregations.
Still not implemented as yet:
1. RANGE frames
2. Multiple different frames in the same query
3. Frames on last/first functions
This commit adds a new class `InputStats` to track the total bytes processed by a task.
The field `processedBytes` is published in task reports along with other row stats.
Major changes:
- Add class `InputStats` to track processed bytes
- Add method `InputSourceReader.read(InputStats)` to read input rows while counting bytes.
> Since we need to count the bytes, we could not just have a wrapper around `InputSourceReader` or `InputEntityReader` (the way `CountableInputSourceReader` does) because the `InputSourceReader` only deals with `InputRow`s and the byte information is already lost.
- Classic batch: Use the new `InputSourceReader.read(inputStats)` in `AbstractBatchIndexTask`
- Streaming: Increment `processedBytes` in `StreamChunkParser`. This does not use the new `InputSourceReader.read(inputStats)` method.
- Extend `InputStats` with `RowIngestionMeters` so that bytes can be exposed in task reports
Other changes:
- Update tests to verify the value of `processedBytes`
- Rename `MutableRowIngestionMeters` to `SimpleRowIngestionMeters` and remove duplicate class
- Replace `CacheTestSegmentCacheManager` with `NoopSegmentCacheManager`
- Refactor `KafkaIndexTaskTest` and `KinesisIndexTaskTest`
Refactor DataSource to have a getAnalysis method()
This removes various parts of the code where while loops and instanceof
checks were being used to walk through the structure of DataSource objects
in order to build a DataSourceAnalysis. Instead we just ask the DataSource
for its analysis and allow the stack to rebuild whatever structure existed.
* Processors for Window Processing
This is an initial take on how to use Processors
for Window Processing. A Processor is an interface
that transforms RowsAndColumns objects.
RowsAndColumns objects are essentially combinations
of rows and columns.
The intention is that these Processors are the start
of a set of operators that more closely resemble what
DB engineers would be accustomed to seeing.
* Wire up windowed processors with a query type that
can run them end-to-end. This code can be used to
actually run a query, so yay!
* Wire up windowed processors with a query type that
can run them end-to-end. This code can be used to
actually run a query, so yay!
* Some SQL tests for window functions. Added wikipedia
data to the indexes available to the
SQL queries and tests validating the windowing
functionality as it exists now.
Co-authored-by: Gian Merlino <gianmerlino@gmail.com>
* Moving all unnest cursor code atop refactored code for unnest
* Updating unnest cursor
* Removing dedup and fixing up some null checks
* AllowList changes
* Fixing some NPEs
* Using bitset for allowlist
* Updating the initialization only when cursor is in non-done state
* Updating code to skip rows not in allow list
* Adding a flag for cases when first element is not in allowed list
* Updating for a null in allowList
* Splitting unnest cursor into 2 subclasses
* Intercepting some apis with columnName for new unnested column
* Adding test cases and renaming some stuff
* checkstyle fixes
* Moving to an interface for Unnest
* handling null rows in a dimension
* Updating cursors after comments part-1
* Addressing comments and adding some more tests
* Reverting a change to ScanQueryRunner and improving a comment
* removing an unused function
* Updating cursors after comments part 2
* One last fix for review comments
* Making some functions private, deleting some comments, adding a test for unnest of unnest with allowList
* Adding an exception for a case
* Closure for unnest data source
* Adding some javadocs
* One minor change in makeDimSelector of columnarCursor
* Updating an error message
* Update processing/src/main/java/org/apache/druid/segment/DimensionUnnestCursor.java
Co-authored-by: Abhishek Agarwal <1477457+abhishekagarwal87@users.noreply.github.com>
* Unnesting on virtual columns was missing an object array, adding that to support virtual columns unnesting
* Updating exceptions to use UOE
* Renamed files, added column capability test on adapter, return statement and made unnest datasource not cacheable for the time being
* Handling for null values in dim selector
* Fixing a NPE for null row
* Updating capabilities
* Updating capabilities
Co-authored-by: Abhishek Agarwal <1477457+abhishekagarwal87@users.noreply.github.com>
SQL test framework extensions
* Capture planner artifacts: logical plan, etc.
* Planner test builder validates the logical plan
* Validation for the SQL resut schema (we already have
validation for the Druid row signature)
* Better Guice integration: properties, reuse Guice modules
* Avoid need for hand-coded expr, macro tables
* Retire some of the test-specific query component creation
* Fix query log hook race condition
* fixes BlockLayoutColumnarLongs close method to nullify internal buffer.
* fixes other BlockLayoutColumnar supplier close methods to nullify internal buffers.
* fix spotbugs
* we can read where we want to
we can leave your bounds behind
'cause if the memory is not there
we really don't care
and we'll crash this process of mine
We added compression to the latest/first pair storage, but
the code change was forcing new things to be persisted
with the new format, meaning that any segment created with
the new code cannot be read by the old code. Instead, we
need to default to creating the old format and then remove that default in a future version.
* Add string comparison methods to StringUtils, fix dictionary comparisons.
There are various places in Druid code where we assume that String.compareTo
is consistent with Unicode code-point ordering. Sadly this is not the case.
To help deal with this, this patch introduces the following helpers:
1) compareUnicode: Compares two Strings in Unicode code-point order.
2) compareUtf8: Compares two UTF-8 byte arrays in Unicode code-point order.
Equivalent to comparison as unsigned bytes.
3) compareUtf8UsingJavaStringOrdering: Compares two UTF-8 byte arrays, or
ByteBuffers, in a manner consistent with String.compareTo.
There is no helper for comparing two Strings in a manner consistent
with String.compareTo, because for that we can use compareTo directly.
The patch also fixes an inconsistency between the String and UTF-8
dictionary GenericIndexed flavors of string-typed columns: they were
formerly using incompatible comparators.
* Adjust test.
* FrontCodedIndexed updates.
* Add test.
* Fix comments.
Changes:
- Add a metric for partition-wise kafka/kinesis lag for streaming ingestion.
- Emit lag metrics for streaming ingestion when supervisor is not suspended and state is in {RUNNING, IDLE, UNHEALTHY_TASKS, UNHEALTHY_SUPERVISOR}
- Document metrics
* Compaction: Fetch segments one at a time on main task; skip when possible.
Compact tasks include the ability to fetch existing segments and determine
reasonable defaults for granularitySpec, dimensionsSpec, and metricsSpec.
This is a useful feature that makes compact tasks work well even when the
user running the compaction does not have a clear idea of what they want
the compacted segments to be like.
However, this comes at a cost: it takes time, and disk space, to do all
of these fetches. This patch improves the situation in two ways:
1) When segments do need to be fetched, download them one at a time and
delete them when we're done. This still takes time, but minimizes the
required disk space.
2) Don't fetch segments on the main compact task when they aren't needed.
If the user provides a full granularitySpec, dimensionsSpec, and
metricsSpec, we can skip it.
* Adjustments.
* Changes from code review.
* Fix logic for determining rollup.
* Use lookup memory footprint in MSQ memory computations.
Two main changes:
1) Add estimateHeapFootprint to LookupExtractor.
2) Use this in MSQ's IndexerWorkerContext when determining the total
amount of available memory. It's taken off the top.
This prevents MSQ tasks from running out of memory when there are lookups
defined in the cluster.
* Updates from code review.
* First set of changes for framework
* Second set of changes to move segment map function to data source
* Minot change to server manager
* Removing the createSegmentMapFunction from JoinableFactoryWrapper and moving to JoinDataSource
* Checkstyle fixes
* Patching Eric's fix for injection
* Checkstyle and fixing some CI issues
* Fixing code inspections and some failed tests and one injector for test in avatica
* Another set of changes for CI...almost there
* Equals and hashcode part update
* Fixing injector from Eric + refactoring for broadcastJoinHelper
* Updating second injector. Might revert later if better way found
* Fixing guice issue in JoinableFactory
* Addressing review comments part 1
* Temp changes refactoring
* Revert "Temp changes refactoring"
This reverts commit 9da42a9ef0.
* temp
* Temp discussions
* Refactoring temp
* Refatoring the query rewrite to refer to a datasource
* Refactoring getCacheKey by moving it inside data source
* Nullable annotation check in injector
* Addressing some comments, removing 2 analysis.isJoin() checks and correcting the benchmark files
* Minor changes for refactoring
* Addressing reviews part 1
* Refactoring part 2 with new test cases for broadcast join
* Set for nullables
* removing instance of checks
* Storing nullables in guice to avoid checking on reruns
* Fixing a test case and removing an irrelevant line
* Addressing the atomic reference review comments
* add FrontCodedIndexed for delta string encoding
* now for actual segments
* fix indexOf
* fixes and thread safety
* add bucket size 4, which seems generally better
* fixes
* fixes maybe
* update indexes to latest interfaces
* utf8 support
* adjust
* oops
* oops
* refactor, better, faster
* more test
* fixes
* revert
* adjustments
* fix prefixing
* more chill
* sql nested benchmark too
* refactor
* more comments and javadocs
* better get
* remove base class
* fix
* hot rod
* adjust comments
* faster still
* minor adjustments
* spatial index support
* spotbugs
* add isSorted to Indexed to strengthen indexOf contract if set, improve javadocs, add docs
* fix docs
* push into constructor
* use base buffer instead of copy
* oops
* SQL: Use timestamp_floor when granularity is not safe.
PR #12944 added a check at the execution layer to avoid materializing
excessive amounts of time-granular buckets. This patch modifies the SQL
planner to avoid generating queries that would throw such errors, by
switching certain plans to use the timestamp_floor function instead of
granularities. This applies both to the Timeseries query type, and the
GroupBy timestampResultFieldGranularity feature.
The patch also goes one step further: we switch to timestamp_floor
not just in the ETERNITY + non-ALL case, but also if the estimated
number of time-granular buckets exceeds 100,000.
Finally, the patch modifies the timestampResultFieldGranularity
field to consistently be a String rather than a Granularity. This
ensures that it can be round-trip serialized and deserialized, which is
useful when trying to execute the results of "EXPLAIN PLAN FOR" with
GroupBy queries that use the timestampResultFieldGranularity feature.
* Fix test, address PR comments.
* Fix ControllerImpl.
* Fix test.
* Fix unused import.
We introduce two new configuration keys that refine the query context security model controlled by druid.auth.authorizeQueryContextParams. When that value is set to true then two other configuration options become available:
druid.auth.unsecuredContextKeys: The set of query context keys that do not require a security check. Use this for the "white-list" of key to allow. All other keys go through the existing context key security checks.
druid.auth.securedContextKeys: The set of query context keys that do require a security check. Use this when you want to allow all but a specific set of keys: only these keys go through the existing context key security checks.
Both are set using JSON list format:
druid.auth.securedContextKeys=["secretKey1", "secretKey2"]
You generally set one or the other values. If both are set, unsecuredContextKeys acts as exceptions to securedContextKeys.
In addition, Druid defines two query context keys which always bypass checks because Druid uses them internally:
sqlQueryId
sqlStringifyArrays
This is in preparation for eventually retiring the flag `useMaxMemoryEstimates`,
after which the footprint of a value in the dimension dictionary will always be
estimated using the `estimateSizeOfValue()` method.
* fix json_value sql planning with decimal type, fix vectorized expression math null value handling in default mode
changes:
* json_value 'returning' decimal will now plan to native double typed query instead of ending up with default string typing, allowing decimal vector math expressions to work with this type
* vector math expressions now zero out 'null' values even in 'default' mode (druid.generic.useDefaultValueForNull=false) to prevent downstream things that do not check the null vector from producing incorrect results
* more better
* test and why not vectorize
* more test, more fix
* Expose HTTP Response headers from SqlResource
This change makes the SqlResource expose HTTP response
headers in the same way that the QueryResource exposes them.
Fundamentally, the change is to pipe the QueryResponse
object all the way through to the Resource so that it can
populate response headers. There is also some code
cleanup around DI, as there was a superfluous FactoryFactory
class muddying things up.
changes:
* long and double value columns are now written directly, at the same time as writing out the 'intermediary' dictionaryid column with unsorted ids
* remove reverse value lookup from GlobalDictionaryIdLookup since it is no longer needed
* more consistent expression error messages
* review stuff
* add NamedFunction for Function, ApplyFunction, and ExprMacro to share common stuff
* fixes
* add expression transform name to transformer failure, better parse_json error messaging
-Add classes for writing cell values in LZ4 block compressed format.
Payloads are indexed by element number for efficient random lookup
-update SerializablePairLongStringComplexMetricSerde to use block
compression
-SerializablePairLongStringComplexMetricSerde also uses delta encoding
of the Long by doing 2-pass encoding: buffers first to find min/max
numbers and delta-encodes as integers if possible
Entry points for doing block-compressed storage of byte[] payloads
are the CellWriter and CellReader class. See
SerializablePairLongStringComplexMetricSerde for how these are used
along with how to do full column-based storage (delta encoding here)
which includes 2-pass encoding to compute a column header
* FrameFile: Java 17 compatibility.
DataSketches Memory.map is not Java 17 compatible, and from discussions
with the team, is challenging to make compatible with 17 while also
retaining compatibility with 8 and 11. So, in this patch, we switch away
from Memory.map and instead use the builtin JDK mmap functionality. Since
it only supports maps up to Integer.MAX_VALUE, we also implement windowing
in FrameFile, such that we can still handle large files.
Other changes:
1) Add two new "map" functions to FileUtils, which we use in this patch.
2) Add a footer checksum to the FrameFile format. Individual frames
already have checksums, but the footer was missing one.
* Changes for static analysis.
* wip
* Fixes.
* Fix accounting of bytesAdded in ReadableByteChunksFrameChannel.
Could cause WorkerInputChannelFactory to get into an infinite loop when
reading the footer of a frame file.
* Additional tests.
During ingestion, if a row containing multiple values for a numeric dimension is encountered,
the whole ingestion task fails. Ideally, this should just be registered as a parse exception.
Changes:
- Remove `instanceof List` check from `LongDimensionIndexer`, `FloatDimensionIndexer` and `DoubleDimensionIndexer`.
Any invalid type, including list, throws a parse exception in `DimensionHandlerUtils.convertObjectToXXX`
methods. `ParseException` is already handled in `OnHeapIncrementalIndex` and does not fail the entire task.
* json_value adjustments
changes:
* native json_value expression now has optional 3rd argument to specify type, which will cast all values to the specified type
* rework how JSON_VALUE is wired up in SQL. Now we are using a custom convertlet to translate JSON_VALUE(... RETURNING type) into dedicated JSON_VALUE_BIGINT, JSON_VALUE_DOUBLE, JSON_VALUE_VARCHAR, JSON_VALUE_ANY instead of using the calcite StandardConvertletTable that wraps JSON_VALUE_ANY in a CAST, so that we preserve the typing of JSON_VALUE to pass down to the native expression as the 3rd argument
* fix json_value_any to be usable by humans too, coverage
* fix bug
* checkstyle
* checkstyle
* review stuff
* validate that options to json_value are the supported options rather than ignore them
* remove more legacy undocumented functions
* KLL sketch
* added documentation
* direct static refs
* direct static refs
* fixed test
* addressed review points
* added KLL sketch related terms
* return a copy from get
* Copy unions when returning them from "get".
* Remove redundant "final".
Co-authored-by: AlexanderSaydakov <AlexanderSaydakov@users.noreply.github.com>
Co-authored-by: Gian Merlino <gianmerlino@gmail.com>
The method wasn't following its contract, leading to pollution of the
overall planner context, when really we just want to create a new
context for a specific query.
* Error handling improvements for frame channels.
Two changes:
1) Send errors down in-memory channels (BlockingQueueFrameChannel) on
failure. This ensures that in situations where a chain of processors
has been set up on a single machine, all processors see the root
cause error. In particular, this means the final processor in the
chain reports the root cause error, which ensures that someone with
a handle to the final processor will get the proper error.
2) Update FrameFileHttpResponseHandler to expect that the final fetch,
rather than being simply empty, is also empty with a special header.
This ensures that the handler is able to tell the difference between
an empty fetch due to being at EOF, and an empty fetch due to a
truncated HTTP response (after the 200 OK and headers are sent down,
but before any content appears).
* Fix tests, imports.
* Checkstyle!
* Refactor SqlLifecycle into statement classes
Create direct & prepared statements
Remove redundant exceptions from tests
Tidy up Calcite query tests
Make PlannerConfig more testable
* Build fixes
* Added builder to SqlQueryPlus
* Moved Calcites system properties to saffron.properties
* Build fix
* Resolve merge conflict
* Fix IntelliJ inspection issue
* Revisions from reviews
Backed out a revision to Calcite tests that didn't work out as planned
* Build fix
* Fixed spelling errors
* Fixed failed test
Prepare now enforces security; before it did not.
* Rebase and fix IntelliJ inspections issue
* Clean up exception handling
* Fix handling of JDBC auth errors
* Build fix
* More tweaks to security messages
* Introduce defaultOnDiskStorage config for groupBy
* add debug log to groupby query config
* Apply config change suggestion from review
* Remove accidental new lines
* update default value of new default disk storage config
* update debug log to have more descriptive text
* Make maxOnDiskStorage and defaultOnDiskStorage HumanRedadableBytes
* improve test coverage
* Provide default implementation to new default method on advice of reviewer
In the current druid code base, we have the interface DataSegmentPusher which allows us to push segments to the appropriate deep storage without the extension being worried about the semantics of how to push too deep storage.
While working on #12262, whose some part of the code will go as an extension, I realized that we do not have an interface that allows us to do basic "write, get, delete, deleteAll" operations on the appropriate deep storage without let's say pulling the s3-storage-extension dependency in the custom extension.
Hence, the idea of StorageConnector was born where the storage connector sits inside the druid core so all extensions have access to it.
Each deep storage implementation, for eg s3, GCS, will implement this interface.
Now with some Jackson magic, we bind the implementation of the correct deep storage implementation on runtime using a type variable.
* Adjust "in" filter null behavior to match "selector".
Now, both of them match numeric nulls if constructed with a "null" value.
This is consistent as far as native execution goes, but doesn't match
the behavior of SQL = and IN. So, to address that, this patch also
updates the docs to clarify that the native filters do match nulls.
This patch also updates the SQL docs to describe how Boolean logic is
handled in addition to how NULL values are handled.
Fixes#12856.
* Fix test.
* Frame processing and channels.
Follow-up to #12745. This patch adds three new concepts:
1) Frame channels are interfaces for doing nonblocking reads and writes
of frames.
2) Frame processors are interfaces for doing nonblocking processing of
frames received from input channels and sent to output channels.
3) Cluster-by keys, which can be used for sorting or partitioning.
The patch also adds SuperSorter, a user of these concepts, both to
illustrate how they are used, and also because it is going to be useful
in future work.
Central classes:
- ReadableFrameChannel. Implementations include
BlockingQueueFrameChannel (in-memory channel that implements both interfaces),
ReadableFileFrameChannel (file-based channel),
ReadableByteChunksFrameChannel (byte-stream-based channel), and others.
- WritableFrameChannel. Implementations include BlockingQueueFrameChannel
and WritableStreamFrameChannel (byte-stream-based channel).
- ClusterBy, a sorting or partitioning key.
- FrameProcessor, nonblocking processor of frames. Implementations include
FrameChannelBatcher, FrameChannelMerger, and FrameChannelMuxer.
- FrameProcessorExecutor, an executor service that runs FrameProcessors.
- SuperSorter, a class that uses frame channels and processors to
do parallel external merge sort of any amount of data (as long as there
is enough disk space).
* Additional tests, fixes.
* Changes from review.
* Better implementation for ReadableInputStreamFrameChannel.
* Rename getFrameFileReference -> newFrameFileReference.
* Add InterruptedException to runIncrementally; add more tests.
* Cancellation adjustments.
* Review adjustments.
* Refactor BlockingQueueFrameChannel, rename doneReading and doneWriting to close.
* Additional changes from review.
* Additional changes.
* Fix test.
* Adjustments.
* Adjustments.
* Refactor Guice initialization
Builders for various module collections
Revise the extensions loader
Injector builders for server startup
Move Hadoop init to indexer
Clean up server node role filtering
Calcite test injector builder
* Revisions from review comments
* Build fixes
* Revisions from review comments
* Improved Java 17 support and Java runtime docs.
1) Add a "Java runtime" doc page with information about supported
Java versions, garbage collection, and strong encapsulation..
2) Update asm and equalsverifier to versions that support Java 17.
3) Add additional "--add-opens" lines to surefire configuration, so
tests can pass successfully under Java 17.
4) Switch openjdk15 tests to openjdk17.
5) Update FrameFile to specifically mention Java runtime incompatibility
as the cause of not being able to use Memory.map.
6) Update SegmentLoadDropHandler to log an error for Errors too, not
just Exceptions. This is important because an IllegalAccessError is
encountered when the correct "--add-opens" line is not provided,
which would otherwise be silently ignored.
7) Update example configs to use druid.indexer.runner.javaOptsArray
instead of druid.indexer.runner.javaOpts. (The latter is deprecated.)
* Adjustments.
* Use run-java in more places.
* Add run-java.
* Update .gitignore.
* Exclude hadoop-client-api.
Brought in when building on Java 17.
* Swap one more usage of java.
* Fix the run-java script.
* Fix flag.
* Include link to Temurin.
* Spelling.
* Update examples/bin/run-java
Co-authored-by: Xavier Léauté <xl+github@xvrl.net>
Co-authored-by: Xavier Léauté <xl+github@xvrl.net>
add NumericRangeIndex interface and BoundFilter support
changes:
* NumericRangeIndex interface, like LexicographicalRangeIndex but for numbers
* BoundFilter now uses NumericRangeIndex if comparator is numeric and there is no extractionFn
* NestedFieldLiteralColumnIndexSupplier.java now supports supplying NumericRangeIndex for single typed numeric nested literal columns
* better faster stronger and (ever so slightly) more understandable
* more tests, fix bug
* fix style
* Automatic sizing for GroupBy dictionary sizes.
Merging and selector dictionary sizes currently both default to 100MB.
This is not optimal, because it can lead to OOM on small servers and
insufficient resource utilization on larger servers. It also invites
end users to try to tune it when queries run out of dictionary space,
which can make things worse if the end user sets it to too high.
So, this patch:
- Adds automatic tuning for selector and merge dictionaries. Selectors
use up to 15% of the heap and merge buffers use up to 30% of the heap
(aggregate across all queries).
- Updates out-of-memory error messages to emphasize enabling disk
spilling vs. increasing memory parameters. With the memory parameters
automatically sized, it is more likely that an end user will get
benefit from enabling disk spilling.
- Removes the query context parameters that allow lowering of configured
dictionary sizes. These complicate the calculation, and I don't see a
reasonable use case for them.
* Adjust tests.
* Review adjustments.
* Additional comment.
* Remove unused import.
When we return DISK_FULL to a processing thread, it skips the rest of
the segment and the query is canceled. However, it's possible that the
next segment starts processing before cancellation can kick in. We want
that one, if it occurs, to see DISK_FULL too.
* Preserve column order in DruidSchema, SegmentMetadataQuery.
Instead of putting columns in alphabetical order. This is helpful
because it makes query order better match ingestion order. It also
allows tools, like the reindexing flow in the web console, to more
easily do follow-on ingestions using a column order that matches the
pre-existing column order.
We prefer the order from the latest segments. The logic takes all
columns from the latest segments in the order they appear, then adds
on columns from older segments after those.
* Additional test adjustments.
* Adjust imports.
* Frame format for data transfer and short-term storage.
As we move towards query execution plans that involve more transfer
of data between servers, it's important to have a data format that
provides for doing this more efficiently than the options available to
us today.
This patch adds:
- Columnar frames, which support fast querying.
- Row-based frames, which support fast sorting via memory comparison
and fast whole-row copies via memory copying.
- Frame files, a container format that can be stored on disk or
transferred between servers.
The idea is we should use row-based frames when data is expected to
be sorted, and columnar frames when data is expected to be queried.
The code in this patch is not used in production yet. Therefore, the
patch involves minimal changes outside of the org.apache.druid.frame
package. The main ones are adjustments to SqlBenchmark to add benchmarks
for queries on frames, and the addition of a "forEach" method to Sequence.
* Fixes based on tests, static analysis.
* Additional fixes.
* Skip DS mapping tests on JDK 14+
* Better JDK checking in tests.
* Fix imports.
* Additional comment.
* Adjustments from code review.
* Update test case.
* jvm gc to mxbeans
* add zgc and shenandoah #12476
* remove tryCreateGcCounter
* separate the space collector
* blend GcGenerationCollector into GcCollector
* add jdk surefire argLine
* Poison StupidPool and fix resource leaks
There are various resource leaks from test setup as well as some
corners in query processing. We poison the StupidPool to start failing
tests when the leaks come and fix any issues uncovered from that so
that we can start from a clean baseline.
Unfortunately, because of how poisoning works,
we can only fail future checkouts from the same pool,
which means that there is a natural race between a
leak happening -> GC occurs -> leak detected -> pool poisoned.
This race means that, depending on interleaving of tests,
if the very last time that an object is checked out
from the pool leaks, then it won't get caught.
At some point in the future, something will catch it,
however and from that point on it will be deterministic.
* Remove various things left over from iterations
* Clean up FilterAnalysis and add javadoc on StupidPool
* Revert changes to .idea/misc.xml that accidentally got pushed
* Style and test branches
* Stylistic woes
True, false, and null have different meanings: true/false mean "legacy"
and "not legacy"; null means use the default set by ScanQueryConfig.
So, we need to respect this in the JsonIgnore setup.
* Fix self-referential shape inspection in BaseExpressionColumnValueSelector.
The new test would throw StackOverflowError on the old code.
* Restore prior test.
* Remove null and empty fields from native queries
* Test fixes
* Attempted IT fix.
* Revisions from review comments
* Build fixes resulting from changes suggested by reviews
* IT fix for changed segment size
* Clean up query contexts
Uses constants in place of literal strings for context keys.
Moves some QueryContext methods to QueryContexts for reuse.
* Revisions from review comments
Often users are submitting queries, and ingestion specs that work only if the relevant extension is not loaded. However, the error is too technical for the users and doesn't suggest them to check for missing extensions. This PR modifies the error message so users can at least check their settings before assuming that the error is because of a bug.
fixes an issue caused by a test modification in #12408 that was closing buffers allocated by the compression strategy instead of allowing the closer to do it
RowBasedColumnSelectorFactory inherited strange behavior from
Rows.objectToStrings for nulls that appear in lists: instead of being
left as a null, it is replaced with the string "null". Some callers may
need compatibility with this strange behavior, but it should be opt-in.
Query-time call sites are changed to opt-out of this behavior, since it
is not consistent with query-time expectations. The IncrementalIndex
ingestion-time call site retains the old behavior, as this is traditionally
when Rows.objectToStrings would be used.
* Add RowIdSupplier to ColumnSelectorFactory.
This enables virtual columns to cache their outputs in case they are
called multiple times on the same underlying row. This is common for
numeric selectors, where the common pattern is to call isNull() and
then follow with getLong(), getFloat(), or getDouble(). Here, output
caching reduces the number of expression evals by half.
* Fix tests.
* Adding zstandard compression library
* 1. Took @clintropolis's advice to have ZStandard decompressor use the byte array when the buffers are not direct.
2. Cleaned up checkstyle issues.
* Fixing zstandard version to latest stable version in pom's and updating license files
* Removing zstd from benchmarks and adding to processing (poms)
* fix the intellij inspection issue
* Removing the prefix v for the version in the license check for ztsd
* Fixing license checks
Co-authored-by: Rahul Gidwani <r_gidwani@apple.com>
Adds a default implementation of getQueryContext, which was added to the Query interface in #12396. Query is marked with @ExtensionPoint, and lately we have been trying to be less volatile on these interfaces by providing default implementations to be more chill for extension writers.
The way this default implementation is done in this PR is a bit strange due to the way that getQueryContext is used (mutated with system default and system generated keys); the default implementation has a specific object that it returns, and I added another temporary default method isLegacyContext that checks if the getQueryContext returns that object or not. If not, callers fall back to using getContext and withOverriddenContext to set these default and system values.
I am open to other ideas as well, but this way should work at least without exploding, and added some tests to ensure that it is wired up correctly for QueryLifecycle, including the context authorization stuff.
The added test shows the strange behavior if query context authorization is enabled, mainly that the system default and system generated query context keys also need to be granted as permissions for things to function correctly. This is not great, so I mentioned it in the javadocs as well. Not sure if it needs to be called out anywhere else.
Description
Fixes a bug when running q's like
SELECT cntarray,
Count(*)
FROM (SELECT dim1,
dim2,
Array_agg(cnt) AS cntarray
FROM (SELECT dim1,
dim2,
dim3,
Count(*) AS cnt
FROM foo
GROUP BY 1,
2,
3)
GROUP BY 1,
2)
GROUP BY 1
This generates an error:
org.apache.druid.java.util.common.ISE: Unable to convert type [Ljava.lang.Object; to org.apache.druid.segment.data.ComparableList
at org.apache.druid.segment.DimensionHandlerUtils.convertToList(DimensionHandlerUtils.java:405) ~[druid-xx]
Because it's an array of numbers it looks like it does the convertToList call, which looks like:
@Nullable
public static ComparableList convertToList(Object obj)
{
if (obj == null) {
return null;
}
if (obj instanceof List) {
return new ComparableList((List) obj);
}
if (obj instanceof ComparableList) {
return (ComparableList) obj;
}
throw new ISE("Unable to convert type %s to %s", obj.getClass().getName(), ComparableList.class.getName());
}
I.e. it doesn't know about arrays. Added the array handling as part of this PR.
* Emit state of replace and append for native batch tasks
* Emit count of one depending on batch ingestion mode (APPEND, OVERWRITE, REPLACE)
* Add metric to compaction job
* Avoid null ptr exc when null emitter
* Coverage
* Emit tombstone & segment counts
* Tasks need a type
* Spelling
* Integrate BatchIngestionMode in batch ingestion tasks functionality
* Typos
* Remove batch ingestion type from metric since it is already in a dimension. Move IngestionMode to AbstractTask to facilitate having mode as a dimension. Add metrics to streaming. Add missing coverage.
* Avoid inner class referenced by sub-class inspection. Refactor computation of IngestionMode to make it more robust to null IOConfig and fix test.
* Spelling
* Avoid polluting the Task interface
* Rename computeCompaction methods to avoid ambiguous java compiler error if they are passed null. Other minor cleanup.
* ConcurrentGrouper: Add option to always slice up merge buffers thread-locally.
Normally, the ConcurrentGrouper shares merge buffers across processing
threads until spilling starts, and then switches to a thread-local model.
This minimizes memory use and reduces likelihood of spilling, which is
good, but it creates thread contention. The new mergeThreadLocal option
causes a query to start in thread-local mode immediately, and allows us
to experiment with the relative performance of the two modes.
* Fix grammar in docs.
* Fix race in ConcurrentGrouper.
* Fix issue with timeouts.
* Remove unused import.
* Add "tradeoff" to dictionary.
* Direct UTF-8 access for "in" filters.
Directly related:
1) InDimFilter: Store stored Strings (in ValuesSet) plus sorted UTF-8
ByteBuffers (in valuesUtf8). Use valuesUtf8 whenever possible. If
necessary, the input set is copied into a ValuesSet. Much logic is
simplified, because we always know what type the values set will be.
I think that there won't even be an efficiency loss in most cases.
InDimFilter is most frequently created by deserialization, and this
patch updates the JsonCreator constructor to deserialize
directly into a ValuesSet.
2) Add Utf8ValueSetIndex, which InDimFilter uses to avoid UTF-8 decodes
during index lookups.
3) Add unsigned comparator to ByteBufferUtils and use it in
GenericIndexed.BYTE_BUFFER_STRATEGY. This is important because UTF-8
bytes can be compared as bytes if, and only if, the comparison
is unsigned.
4) Add specialization to GenericIndexed.singleThreaded().indexOf that
avoids needless ByteBuffer allocations.
5) Clarify that objects returned by ColumnIndexSupplier.as are not
thread-safe. DictionaryEncodedStringIndexSupplier now calls
singleThreaded() on all relevant GenericIndexed objects, saving
a ByteBuffer allocation per access.
Also:
1) Fix performance regression in LikeFilter: since #12315, it applied
the suffix matcher to all values in range even for type MATCH_ALL.
2) Add ObjectStrategy.canCompare() method. This fixes LikeFilterBenchmark,
which was broken due to calls to strategy.compare in
GenericIndexed.fromIterable.
* Add like-filter implementation tests.
* Add in-filter implementation tests.
* Add tests, fix issues.
* Fix style.
* Adjustments from review.
* Ensure ByteBuffers allocated in tests get freed.
Many tests had problems where a direct ByteBuffer would be allocated
and then not freed. This is bad because it causes flaky tests.
To fix this:
1) Add ByteBufferUtils.allocateDirect(size), which returns a ResourceHolder.
This makes it easy to free the direct buffer. Currently, it's only used
in tests, because production code seems OK.
2) Update all usages of ByteBuffer.allocateDirect (off-heap) in tests either
to ByteBuffer.allocate (on-heap, which are garbaged collected), or to
ByteBufferUtils.allocateDirect (wherever it seemed like there was a good
reason for the buffer to be off-heap). Make sure to close all direct
holders when done.
* Changes based on CI results.
* A different approach.
* Roll back BitmapOperationTest stuff.
* Try additional surefire memory.
* Revert "Roll back BitmapOperationTest stuff."
This reverts commit 49f846d9e3.
* Add TestBufferPool.
* Revert Xmx change in tests.
* Better behaved NestedQueryPushDownTest. Exit tests on OOME.
* Fix TestBufferPool.
* Remove T1C from ARM tests.
* Somewhat safer.
* Fix tests.
* Fix style stuff.
* Additional debugging.
* Reset null / expr configs better.
* ExpressionLambdaAggregatorFactory thread-safety.
* Alter forkNode to try to get better info when a JVM crashes.
* Fix buffer retention in ExpressionLambdaAggregatorFactory.
* Remove unused import.
In the majority of cases, this improves performance.
There's only one case I'm aware of where this may be a net negative: for time_floor(__time, <period>) where there are many repeated __time values. In nonvectorized processing, SingleLongInputCachingExpressionColumnValueSelector implements an optimization to avoid computing the time_floor function on every row. There is no such optimization in vectorized processing.
IMO, we shouldn't mention this in the docs. Rationale: It's too fiddly of a thing: it's not guaranteed that nonvectorized processing will be faster due to the optimization, because it would have to overcome the inherent speed advantage of vectorization. So it'd always require testing to determine the best setting for a specific dataset. It would be bad if users disabled vectorization thinking it would speed up their queries, and it actually slowed them down. And even if users do their own testing, at some point in the future we'll implement the optimization for vectorized processing too, and it's likely that users that explicitly disabled vectorization will continue to have it disabled. I'd like to avoid this outcome by encouraging all users to enable vectorization at all times. Really advanced users would be following development activity anyway, and can read this issue
setting thread names takes a measurable amount of time in the case where segment scans are very quick. In high-QPS testing we found a slight performance boost from turning off processing thread renaming. This option makes that possible.
* Add ipaddress library as dependency.
* IPv4 functions to use the inet.ipaddr package.
* Remove unused imports.
* Add new function.
* Minor rename.
* Add more unit tests.
* IPv4 address expr utils unit tests and address options.
* Adjust the IPv4Util functions.
* Move the UTs a bit around.
* Javadoc comments.
* Add license info for IPAddress.
* Fix groupId, artifact and version in license.yaml.
* Remove redundant subnet in messages - fixes UT.
* Remove unused commons-net dependency for /processing project.
* Make class and methods public so it can be accessed.
* Add initial version of benchmark
* Add subnetutils package for benchmarks.
* Auto generate ip addresses.
* Add more v4 address representations in setup to avoid bias.
* Use ThreadLocalRandom to avoid forbidden API usage.
* Adjust IPv4AddressBenchmark to adhere to codestyle rules.
* Update ipaddress library to latest 5.3.4
* Add ipaddress package dependency to benchmarks project.
Following up on #12315, which pushed most of the logic of building ImmutableBitmap into BitmapIndex in order to hide the details of how column indexes are implemented from the Filter implementations, this PR totally refashions how Filter consume indexes. The end result, while a rather dramatic reshuffling of the existing code, should be extraordinarily flexible, eventually allowing us to model any type of index we can imagine, and providing the machinery to build the filters that use them, while also allowing for other column implementations to implement the built-in index types to provide adapters to make use indexing in the current set filters that Druid provides.
* Add feature flag for sql planning of TimeBoundary queries
* fixup! Add feature flag for sql planning of TimeBoundary queries
* Add documentation for enableTimeBoundaryPlanning
* fixup! Add documentation for enableTimeBoundaryPlanning
* Vectorized version of string last aggregator
* Updating string last and adding testcases
* Updating code and adding testcases for serializable pairs
* Addressing review comments
* GroupBy: Reduce allocations by reusing entry and key holders.
Two main changes:
1) Reuse Entry objects returned by various implementations of
Grouper.iterator.
2) Reuse key objects contained within those Entry objects.
This is allowed by the contract, which states that entries must be
processed and immediately discarded. However, not all call sites
respected this, so this patch also updates those call sites.
One particularly sneaky way that the old code retained entries too long
is due to Guava's MergingIterator and CombiningIterator. Internally,
these both advance to the next value prior to returning the current
value. So, this patch addresses that in two ways:
1) For merging, we have our own implementation MergeIterator already,
although it had the same problem. So, this patch updates our
implementation to return the current item prior to advancing to the
next item. It also adds a forbidden-api entry to ensure that this
safer implementation is used instead of Guava's.
2) For combining, we address the problem in a different way: by copying
the key when creating the new, combined entry.
* Attempt to fix test.
* Remove unused import.
* Reduce allocations due to Jackson serialization.
This patch attacks two sources of allocations during Jackson
serialization:
1) ObjectMapper.writeValue and JsonGenerator.writeObject create a new
DefaultSerializerProvider instance for each call. It has lots of
fields and creates pressure on the garbage collector. So, this patch
adds helper functions in JacksonUtils that enable reuse of
SerializerProvider objects and updates various call sites to make
use of this.
2) GroupByQueryToolChest copies the ObjectMapper for every query to
install a special module that supports backwards compatibility with
map-based rows. This isn't needed if resultAsArray is set and
all servers are running Druid 0.16.0 or later. This release was a
while ago. So, this patch disables backwards compatibility by default,
which eliminates the need to copy the heavyweight ObjectMapper. The
patch also introduces a configuration option that allows admins to
explicitly enable backwards compatibility.
* Add test.
* Update additional call sites and add to forbidden APIs.
* JvmMonitor: Handle more generation and collector scenarios.
ZGC on Java 11 only has a generation 1 (there is no 0). This causes
a NullPointerException when trying to extract the spacesCount for
generation 0. In addition, ZGC on Java 15 has a collector number 2
but no spaces in generation 2, which breaks the assumption that
collectors always have same-numbered spaces.
This patch adjusts things to be more robust, enabling the JvmMonitor
to work properly for ZGC on both Java 11 and 15.
* Test adjustments.
* Improve surefire arglines.
* Need a placeholder
* Vectorizing Latest aggregator Part 1
* Updating benchmark tests
* Changing appropriate logic for vectors for null handling
* Introducing an abstract class and moving the commonalities there
* Adding vectorization for StringLast aggregator (initial version)
* Updated bufferized version of numeric aggregators
* Adding some javadocs
* Making sure this PR vectorizes numeric latest agg only
* Adding another benchmarking test
* Fixing intellij inspections
* Adding tests for double
* Adding test cases for long and float
* Updating testcases
* Checkstyle oops..
* One tiny change in test case
* Fixing spotbug and rhs not being used
This PR enables ARM builds on Travis. I've ported over the changes from @martin-g on reducing heap requirements for some of the tests to ensure they run well on Travis arm instances.
* Support array based results in timeBoundary query
* Fix bug with query interval in timeBoundary
* Convert min(__time) and max(__time) SQL queries to timeBoundary
* Add tests for timeBoundary backed SQL queries
* Fix query plans for existing tests
* fixup! Convert min(__time) and max(__time) SQL queries to timeBoundary
* fixup! Add tests for timeBoundary backed SQL queries
* fixup! Fix bug with query interval in timeBoundary
The query context is a way that the user gives a hint to the Druid query engine, so that they enforce a certain behavior or at least let the query engine prefer a certain plan during query planning. Today, there are 3 types of query context params as below.
Default context params. They are set via druid.query.default.context in runtime properties. Any user context params can be default params.
User context params. They are set in the user query request. See https://druid.apache.org/docs/latest/querying/query-context.html for parameters.
System context params. They are set by the Druid query engine during query processing. These params override other context params.
Today, any context params are allowed to users. This can cause
1) a bad UX if the context param is not matured yet or
2) even query failure or system fault in the worst case if a sensitive param is abused, ex) maxSubqueryRows.
This PR adds an ability to limit context params per user role. That means, a query will fail if you have a context param set in the query that is not allowed to you. To do that, this PR adds a new built-in resource type, QUERY_CONTEXT. The resource to authorize has a name of the context param (such as maxSubqueryRows) and the type of QUERY_CONTEXT. To allow a certain context param for a user, the user should be granted WRITE permission on the context param resource. Here is an example of the permission.
{
"resourceAction" : {
"resource" : {
"name" : "maxSubqueryRows",
"type" : "QUERY_CONTEXT"
},
"action" : "WRITE"
},
"resourceNamePattern" : "maxSubqueryRows"
}
Each role can have multiple permissions for context params. Each permission should be set for different context params.
When a query is issued with a query context X, the query will fail if the user who issued the query does not have WRITE permission on the query context X. In this case,
HTTP endpoints will return 403 response code.
JDBC will throw ForbiddenException.
Note: there is a context param called brokerService that is used only by the router. This param is used to pin your query to run it in a specific broker. Because the authorization is done not in the router, but in the broker, if you have brokerService set in your query without a proper permission, your query will fail in the broker after routing is done. Technically, this is not right because the authorization is checked after the context param takes effect. However, this should not cause any user-facing issue and thus should be OK. The query will still fail if the user doesn’t have permission for brokerService.
The context param authorization can be enabled using druid.auth.authorizeQueryContextParams. This is disabled by default to avoid any hassle when someone upgrades his cluster blindly without reading release notes.
* add impl
* add impl
* fix checkstyle
* add impl
* add unit test
* fix stuff
* fix stuff
* fix stuff
* add unit test
* add more unit tests
* add more unit tests
* add IT
* add IT
* add IT
* add IT
* add ITs
* address comments
* fix test
* fix test
* fix test
* address comments
* address comments
* address comments
* fix conflict
* fix checkstyle
* address comments
* fix test
* fix checkstyle
* fix test
* fix test
* fix IT
* Counting nulls in String cardinality with a config
* Adding tests for the new config
* Wrapping the vectorize part to allow backward compatibility
* Adding different tests, cleaning the code and putting the check at the proper position, handling hasRow() and hasValue() changes
* Updating testcase and code
* Adding null handling test to improve coverage
* Checkstyle fix
* Adding 1 more change in docs
* Making docs clearer
* Store null columns in the segments
* fix test
* remove NullNumericColumn and unused dependency
* fix compile failure
* use guava instead of apache commons
* split new tests
* unused imports
* address comments
Added Calcites InQueryThreshold as a query context parameter. Setting this parameter appropriately reduces the time taken for queries with large number of values in their IN conditions.
* Adding null handling for double mean aggregator
* Updating code to handle nulls in DoubleMean aggregator
* oops last one should have checkstyle issues. fixed
* Updating some code and test cases
* Checking on object is null in case of numeric aggregator
* Adding one more test to improve coverage
* Changing one test as asked in the review
* Changing one test as asked in the review for nulls
* Fix error message for groupByEnableMultiValueUnnesting.
It referred to the incorrect context parameter.
Also, create a dedicated exception class, to allow easier detection of this
specific error.
* Fix other test.
* More better error messages.
* Test getDimensionName method.
Currently, the CNF conversion of a filter is unbounded, which means that it can create as many filters as possible thereby also leading to OOMs in historical heap. We should throw an error or disable CNF conversion if the filter count starts getting out of hand. There are ways to do CNF conversion with linear increase in filters as well but that has been left out of the scope of this change since those algorithms add new variables in the predicate - which can be contentious.
* Tombstone support for replace functionality
* A used segment interval is the interval of a current used segment that overlaps any of the input intervals for the spec
* Update compaction test to match replace behavior
* Adapt ITAutoCompactionTest to work with tombstones rather than dropping segments. Add support for tombstones in the broker.
* Style plus simple queriableindex test
* Add segment cache loader tombstone test
* Add more tests
* Add a method to the LogicalSegment to test whether it has any data
* Test filter with some empty logical segments
* Refactor more compaction/dropexisting tests
* Code coverage
* Support for all empty segments
* Skip tombstones when looking-up broker's timeline. Discard changes made to tool chest to avoid empty segments since they will no longer have empty segments after lookup because we are skipping over them.
* Fix null ptr when segment does not have a queriable index
* Add support for empty replace interval (all input data has been filtered out)
* Fixed coverage & style
* Find tombstone versions from lock versions
* Test failures & style
* Interner was making this fail since the two segments were consider equal due to their id's being equal
* Cleanup tombstone version code
* Force timeChunkLock whenever replace (i.e. dropExisting=true) is being used
* Reject replace spec when input intervals are empty
* Documentation
* Style and unit test
* Restore test code deleted by mistake
* Allocate forces TIME_CHUNK locking and uses lock versions. TombstoneShardSpec added.
* Unused imports. Dead code. Test coverage.
* Coverage.
* Prevent killer from throwing an exception for tombstones. This is the killer used in the peon for killing segments.
* Fix OmniKiller + more test coverage.
* Tombstones are now marked using a shard spec
* Drop a segment factory.json in the segment cache for tombstones
* Style
* Style + coverage
* style
* Add TombstoneLoadSpec.class to mapper in test
* Update core/src/main/java/org/apache/druid/segment/loading/TombstoneLoadSpec.java
Typo
Co-authored-by: Jonathan Wei <jon-wei@users.noreply.github.com>
* Update docs/configuration/index.md
Missing
Co-authored-by: Jonathan Wei <jon-wei@users.noreply.github.com>
* Typo
* Integrated replace with an existing test since the replace part was redundant and more importantly, the test file was very close or exceeding the 10 min default "no output" CI Travis threshold.
* Range does not work with multi-dim
Co-authored-by: Jonathan Wei <jon-wei@users.noreply.github.com>
* add topn heap optimization when string is dictionary encoded, but not uniquely
* use array instead
* is same
* fix javadoc
* fix
* Update StringTopNColumnAggregatesProcessor.java
* GroupBy: Cap dictionary-building selector memory usage.
New context parameter "maxSelectorDictionarySize" controls when the
per-segment processing code should return early and trigger a trip
to the merge buffer.
Includes:
- Vectorized and nonvectorized implementations.
- Adjustments to GroupByQueryRunnerTest to exercise this code in
the v2SmallDictionary suite. (Both the selector dictionary and
the merging dictionary will be small in that suite.)
- Tests for the new config parameter.
* Fix issues from tests.
* Add "pre-existing" to dictionary.
* Simplify GroupByColumnSelectorStrategy interface by removing one of the writeToKeyBuffer methods.
* Adjustments from review comments.
There aren't any changes in this patch that improve Java 11
compatibility; these changes have already been done separately. This
patch merely updates documentation and explicit Java version checks.
The log message adjustments in DruidProcessingConfig are there to make
things a little nicer when running in Java 11, where we can't measure
direct memory _directly_, and so we may auto-size processing buffers
incorrectly.
Row stats are reported for single phase tasks in the `/liveReports` and `/rowStats` APIs
and are also a part of the overall task report. This commit adds changes to report
row stats for multiphase tasks too.
Changes:
- Add `TaskReport` in `GeneratedPartitionsReport` generated during hash and range partitioning
- Collect the reports for `index_generate` phase in `ParallelIndexSupervisorTask`
* upgrade Airline to Airline 2
https://github.com/airlift/airline is no longer maintained, updating to
https://github.com/rvesse/airline (Airline 2) to use an actively
maintained version, while minimizing breaking changes.
Note, this is a backwards incompatible change, and extensions relying on
the CliCommandCreator extension point will also need to be updated.
* fix dependency checks where jakarta.inject is now resolved first instead
of javax.inject, due to Airline 2 using jakarta
* perf: indexing: Introduce a bulk getValuesInto function to read values in bulk
If large number of values are required from DimensionDictionary
during indexing, fetch them all in a single lock/unlock instead of
lock/unlock each individual item.
* refactor: rename key to keys in function args
* fix: check explicitly that argument length on arrays match
* refactor: getValuesInto renamed to getValues, now creates and returns a new T[] rather than filling
As part of #12078 one of the followup's was to have a specific config which does not allow accidental unnesting of multi value columns if such columns become part of the grouping key.
Added a config groupByEnableMultiValueUnnesting which can be set in the query context.
The default value of groupByEnableMultiValueUnnesting is true, therefore it does not change the current engine behavior.
If groupByEnableMultiValueUnnesting is set to false, the query will fail if it encounters a multi-value column in the grouping key.
* Moving in filter check to broker
* Adding more unit tests, making error message meaningful
* Spelling and doc changes
* Updating default to -1 and making this feature hide by default. The number of IN filters can grow upto a max limit of 100
* Removing upper limit of 100, updated docs
* Making documentation more meaningful
* Moving check outside to PlannerConfig, updating test cases and adding back max limit
* Updated with some additional code comments
* Missed removing one line during the checkin
* Addressing doc changes and one forbidden API correction
* Final doc change
* Adding a speling exception, correcting a testcase
* Reading entire filter tree to address combinations of ANDs and ORs
* Specifying in docs that, this case works only for ORs
* Revert "Reading entire filter tree to address combinations of ANDs and ORs"
This reverts commit 81ca8f8496.
* Covering a class cast exception and updating docs
* Counting changed
Co-authored-by: Jihoon Son <jihoonson@apache.org>
* rework sql planner expression and virtual column handling
* simplify a bit
* add back and deprecate old methods, more tests, fix multi-value string coercion bug and associated tests
* spotbugs
* fix bugs with multi-value string array expression handling
* javadocs and adjust test
* better
* fix tests
* array_concat_agg and array_agg support for array inputs
changes:
* added array_concat_agg to aggregate arrays into a single array
* added array_agg support for array inputs to make nested array
* added 'shouldAggregateNullInputs' and 'shouldCombineAggregateNullInputs' to fix a correctness issue with STRING_AGG and ARRAY_AGG when merging results, with dual purpose of being an optimization for aggregating
* fix test
* tie capabilities type to legacy mode flag about coercing arrays to strings
* oops
* better javadoc
* Harmonize implementations of "visit" for Exprs from ExprMacros.
Many of them had bugs where they would not visit all of the original
arguments. I don't think this has user-visible consequences right now,
but it's possible it would in a future world where "visit" is used
for more stuff than it is today.
So, this patch all updates all implementations to a more consistent
style that emphasizes reapplying the macro to the shuttled args.
* Test fixes, test coverage, PR review comments.
Fixes#12022
### Description
The current implementations of memory estimation in `OnHeapIncrementalIndex` and `StringDimensionIndexer` tend to over-estimate which leads to more persistence cycles than necessary.
This PR replaces the max estimation mechanism with getting the incremental memory used by the aggregator or indexer at each invocation of `aggregate` or `encode` respectively.
### Changes
- Add new flag `useMaxMemoryEstimates` in the task context. This overrides the same flag in DefaultTaskConfig i.e. `druid.indexer.task.default.context` map
- Add method `AggregatorFactory.factorizeWithSize()` that returns an `AggregatorAndSize` which contains
the aggregator instance and the estimated initial size of the aggregator
- Add method `Aggregator.aggregateWithSize()` which returns the incremental memory used by this aggregation step
- Update the method `DimensionIndexer.processRowValsToKeyComponent()` to return the encoded key component as well as its effective size in bytes
- Update `OnHeapIncrementalIndex` to use the new estimations only if `useMaxMemoryEstimates = false`
* VirtualizedColumnSelectorFactory: Allow virtual columns to reference each other.
This matches the behavior of QueryableIndex and IncrementalIndex based cursors.
* Fixes to getColumnCapabilities.
* init multiValue column group by
* Changing sorting to Lexicographic as default
* Adding initial tests
* 1.Fixing test cases adding
2.Optimized inmem structs
* Linking SQL layer to native layer
* Adding multiDimension support to group by column strategy
* 1. Removing array coercion in Calcite layer
2. Removing ResultRowDeserializer
* 1. Supporting all primitive array types
2. Removing dimension spec as part of columnSelector
* 1. Supporting all primitive array types
2. Removing dimension spec as part of columnSelector
* 1. Checkstyle things
2. Removing flag
* Minor naming things
* CheckStyle Things
* Fixing test case
* Fixing hashing
* 1. Adding the MV function
2. Added few test cases
* 1. Adding MV function test cases
* Adding Selector strategy function test cases
* Fixing ClientQuerySegmentWalkerTest
* Adding GroupByQueryRunnerTest test cases
* Fixing test cases
* Adding few more test cases
* Fixing Exception asset statement and intellij inspection
* Adding null compatibility tests
* Review comments
* Fixing few failing tests
* Fixing few failing tests
* Do no convert to topN Q incase of group by on array
* Fixing checkstyle
* Fixing differences between jdk's class cast exception message
* 1. Fixing ordering if the grouping key is an array
* Fixing DefaultLimitSpec
* Fixing CalciteArraysQueryTest
* Dummy commit for LGTM
* changes:
* only coerce multi-value string null values when `ExpressionPlan.Trait.NEEDS_APPLIED` is set
* correct return type inference for ARRAY_APPEND,ARRAY_PREPEND,ARRAY_SLICE,ARRAY_CONCAT
* fix bug with ExprEval.ofType when actual type of object from binding doesn't match its claimed type
* Review comments
* Fixing test cases
* Fixing spot bugs
* Fixing strict compile
Co-authored-by: Clint Wylie <cwylie@apache.org>
* Pass VirtualColumnRegistry in PlannerContext for join expression planning
* Allow for including VCs from join fact table expression
* Optmize MV_FILTER functions to use a VC when in join fact table expression
* fixup! Allow for including VCs from join fact table expression
* Address review comments
This change mimics what was done in PR #11917 to
fix the incompatibilities produced by #11713. #11917
fixed it with AggregatorFactory by creating default
methods to allow for extensions built against old
jars to still work. This does the same for PostAggregator
* fix delegated smoosh writer and some new facilities for segment writeout medium
changes:
* fixed issue with delegated `SmooshedWriter` when writing files that look like paths, causing `NoSuchFileException` exceptions when attempting to open a channel to the file
* `FileSmoosher.addWithSmooshedWriter` when _not_ delegating now checks that it is still open when closing, making it a no-op if already closed (allowing column serializers to add additional files and avoid delegated mode if they are finished writing out their own content and ned to add additional files)
* add `makeChildWriteOutMedium` to `SegmentWriteOutMedium` interface, which allows users of a shared medium to clean up `WriteOutBytes` if they fully control the lifecycle. there are no callers of this yet, adding for future functionality
* `OnHeapByteBufferWriteOutBytes` now can be marked as not open so it `OnHeapMemorySegmentWriteOutMedium` can now behave identically to other medium implementations
* fix to address nit - use AtomicLong
* Thread pool for broker
* Updating two tests to improve coverage for new method added
* Updating druidProcessingConfigTest to cover coverage
* Adding missed spelling errors caused in doc
* Adding test to cover lines of new function added
changes:
* IncrementalIndex is now a ColumnInspector
* fixes performance regression from using map of ColumnCapabilities from IncrementalIndex as a RowSignature
DruidSchema consists of a concurrent HashMap of DataSource -> Segement -> AvailableSegmentMetadata. AvailableSegmentMetadata contains RowSignature of the segment, and for each segment, a new object is getting created. RowSignature is an immutable class, and hence it can be interned, and this can lead to huge savings of memory being used in broker, since a lot of the segments of a table would potentially have same RowSignature.
* fix bug where queries fail immediately when timeout is 0 instead of using default timeout
* fix to use serverside max
* more better
* less flaky test
* oops
* Refactor ResponseContext
Fixes a number of issues in preparation for request trailers
and the query profile.
* Converts keys from an enum to classes for smaller code
* Wraps stored values in functions for easier capture for other uses
* Reworks the "header squeezer" to handle types other than arrays.
* Uses metadata for visibility, and ability to compress,
to replace ad-hoc code.
* Cleans up JSON serialization for the response context.
* Other miscellaneous cleanup.
* Handle unknown keys in deserialization
Also, make "Visibility" into a boolean.
* Revised comment
* Renamd variable
changes:
* adds new config, druid.expressions.useStrictBooleans which make longs the official boolean type of all expressions
* vectorize logical operators and boolean functions, some only if useStrictBooleans is true
* Code cleanup from query profile project
* Fix spelling errors
* Fix Javadoc formatting
* Abstract out repeated test code
* Reuse constants in place of some string literals
* Fix up some parameterized types
* Reduce warnings reported by Eclipse
* Reverted change due to lack of tests
* Use intermediate-persist IndexSpec during multiphase merge.
The main change is the addition of an intermediate-persist IndexSpec
to the main "merge" method in IndexMerger. There are also a few minor
adjustments to the IndexMerger interface to encourage more harmonious
usage of its methods in the future.
* Additional changes inspired by the test coverage checker.
- Remove unused-in-production IndexMerger methods "append" and "convert".
- Add additional unit tests to UnifiedIndexerAppenderatorsManager.
* Additional adjustments.
* Even more additional adjustments.
* Test fixes.
Add a "guessAggregatorHeapFootprint" method to AggregatorFactory that
mitigates #6743 by enabling heap footprint estimates based on a specific
number of rows. The idea is that at ingestion time, the number of rows
that go into an aggregator will be 1 (if rollup is off) or will likely
be a small number (if rollup is on).
It's a heuristic, because of course nothing guarantees that the rollup
ratio is a small number. But it's a common case, and I expect this logic
to go wrong much less often than the current logic. Also, when it does
go wrong, users can fix it by lowering maxRowsInMemory or
maxBytesInMemory. The current situation is unintuitive: when the
estimation goes wrong, users get an OOME, but actually they need to
*raise* these limits to fix it.
Add the ability to pass time column in first/last aggregator (and latest/earliest SQL functions). It is to support cases where the time to query upon is stored as a part of a column different than __time. Also, some other logical time column can be specified.
Simplifies logic for callers that only want to get a list of all the
column names, or column names and types. Updated callers SegmentAnalyzer,
HashJoinSegmentStorageAdapter, and DruidSegmentReader.
* SQL INSERT planner support.
The main changes are:
1) DruidPlanner is able to validate and authorize INSERT queries. They
require WRITE permission on the target datasource.
2) QueryMaker is now an interface, and there is a QueryMakerFactory that
creates instances of it. There is only one production implementation
of each (NativeQueryMaker and NativeQueryMakerFactory), which
together behave the same way as the former QueryMaker class. But this
opens the door to executing queries in ways other than the Druid
query stack, and is used by unit tests (CalciteInsertDmlTest) to
test the INSERT planning functionality.
3) Adds an EXTERN table macro that allows references external data using
InputSource and InputFormat from Druid's batch ingestion API. This is
not exposed in production yet, but is used by unit tests.
4) Adds a QueryFeature concept that enables the planner to change its
behavior slightly depending on the capabilities of the execution
system.
5) Adds an "AuthorizableOperator" concept that enables SqlOperators
to require additional permissions. This is used by the EXTERN table
macro.
Related odds and ends:
- Add equals, hashCode, toString methods to InlineInputSource. Aids in
the "from external" tests in CalciteInsertDmlTest.
- Add JSON-serializability to RowSignature.
- Move the SQL string inside PlannerContext so it is "baked into" the
planner when the planner is created. Cleans up the code a bit, since
in practice, the same query is passed in every time to the
same planner anyway.
* Fix up calls to CalciteTests.createMockQueryLifecycleFactory.
* Fix checkstyle issues.
* Adjustments for CI.
* Adjust DruidAvaticaHandlerTest for stricter test authorizations.
Important because an earlier call to getCachedColumn may have been
done with a different class, leading to a ClassCastException on the
second call. In the prior code, this could happen if a complex column
had makeDimensionSelector called on it after makeColumnValueSelector had
already been called.
Usually, "execute" is called by methods defined in the superclass
AbstractExecutorService, and the passed-in Runnable has been wrapped
by newTaskFor inside a PrioritizedListenableFutureTask. But this method
can also be called directly, and if so, the same wrapping is necessary
for the delegate to get a Runnable that can be entered into a priority
queue with the others.
* add back and deprecate aggregator factory methods so i can say i told you so when i delete these later
* rename to make less ambiguous, fix fill method
* adjust
* Scan: Add "orderBy" parameter.
This patch adds an API for requesting non-time orderings, although it
does not actually add the ability to execute such queries.
The changes are done in such a way that no matter how Scan query objects
are constructed, they will have a correct "getOrderBy". This will enable
us to switch the execution to exclusively use "getOrderBy" later on when
it's implemented.
Scan queries are serialized such that they only include "order" (time
order) if the ordering is time-based, and they only include "orderBy" if
the ordering is non-time-based. This maximizes compatibility with
the existing API while also providing a clean look for formatted queries.
Because this patch does not include execution logic, if someone actually
tries to run a query with non-time ordering, then they will get an error
like "Cannot execute query with orderBy [quality ASC]".
* SQL module fixes.
* Add spotbugs-exclude.
* Remove unused method.
Add method ShardSpec.getType() to get name of shard spec type
List all names of shard spec types in the interface ShardSpec itself
for easy reference and maintenance
Add dimension partitioningType to metric segment/added/bytes
PR #11882 introduced a type comparison using ==, but while it was in flight,
another PR #11713 changed the type enum to a class. So the comparison should
properly be done with "equals".
There are 3 types of query IDs - id, subQueryId, sqlQueryId. Currently, whenever a query generates subqueries, the subquery's subQueryId is populated randomly. Also, subquery's Id is not set to the parent query Id. Therefore there is no way of linking the subqueries to the parent query, and one loses the ability to look at end to end view of the query.
This PR aims to implement following couple of things:
Populate the subqueries with it's parent's id (and sqlQueryId if present)
Populate the subqueryId such that it forms a hierarchical relationship amongs themselves. For example, if there is a query which launches a subquery, which in turn launches a couple of subqueries, then the ids and subQueryIds should have following structure.
* revert ColumnAnalysis type, add typeSignature and use it for DruidSchema
* review stuffs
* maybe null
* better maybe null
* Update docs/querying/segmentmetadataquery.md
* Update docs/querying/segmentmetadataquery.md
Co-authored-by: Charles Smith <techdocsmith@gmail.com>
* fix null right
* sad
* oops
* Update batch_hadoop_queries.json
Co-authored-by: Charles Smith <techdocsmith@gmail.com>
* RowBasedSegment: Use Sequence instead of Iterable.
The main reason this is good is that Sequences can include baggage that
must be closed after iteration is finished. This enables creating
RowBasedSegments on top of closeable sequences of rows.
To preserve the optimization that allows reversing a List without
copying it, this patch also makes SimpleSequence its own class and allows
extracting the Iterable that was used to create it.
* Fix tests.
* Add Finalization option to RowSignature.addAggregators.
This make type signatures more useful when the caller knows whether it will
be reading aggregation results in their finalized or intermediate types.
* Fix call site.
* add missing json type for ListFilteredVirtualColumn, and tests to try to avoid this happening again
* fixes
* ugly, but maybe this
* oops
* too many mappers
* Remove StorageAdapter.getColumnTypeName.
It was only used by SegmentAnalyzer, and isn't necessary anymore due to
the recent improvements to ColumnCapabilities.
Also: tidy ColumnDescriptor.read slightly by removing an instanceof
check, and moving the relevant logic into ComplexColumnPartSerde.
* Fix spellings.
This could happen for right or full outer joins in certain cases. Tests
weren't catching this because existing Cursor implementations generally
ignore extraneous calls to "advance". So, to help catch this in tests,
extra state validations are also added to RowWalker, which is used by
RowBasedSegment.
* RowBasedCursor: Add column-value-reuse optimization.
Most of the logic is in RowBasedColumnSelectorFactory, although in this
patch its only user is RowBasedCursor. This improves performance of
features that use RowBasedSegment, like lookup and inline datasources.
It's especially helpful for inline datasources that contain lengthy
arrays, due to the fact that the transformed array can be reused.
* Changes from code review.
* Fixes for ColumnCapabilitiesImplTest.
* complex typed expressions
* add built-in hll collector expressions to get coverage on druid-processing, more types, more better
* rampage!!!
* more javadoc
* adjustments
* oops
* lol
* remove unused dependency
* contradiction?
* more test
* Remove OffheapIncrementalIndex and clarify aggregator thread-safety needs.
This patch does the following:
- Removes OffheapIncrementalIndex.
- Clarifies that Aggregators are required to be thread safe.
- Clarifies that BufferAggregators and VectorAggregators are not
required to be thread safe.
- Removes thread safety code from some DataSketches aggregators that
had it. (Not all of them did, and that's OK, because it wasn't necessary
anyway.)
- Makes enabling "useOffheap" with groupBy v1 an error.
Rationale for removing the offheap incremental index:
- It is only used in one rare scenario: groupBy v1 (which is non-default)
in "useOffheap" mode (also non-default). So you have to go pretty deep
into the wilderness to get this code to activate in production. It is
never used during ingestion.
- Its existence complicates developer efforts to reason about how
aggregators get used, because the way it uses buffer aggregators is so
different from how every other query engine uses them.
- It doesn't have meaningful testing.
By the way, I do believe that the given way the offheap incremental index
works, it actually didn't require buffer aggregators to be thread-safe.
It synchronizes on "aggregate" and doesn't call "get" until it has
stopped calling "aggregate". Nevertheless, this is a bother to think about,
and for the above reasons I think it makes sense to remove the code anyway.
* Remove things that are now unused.
* Revert removal of getFloat, getLong, getDouble from BufferAggregator.
* OAK-related warnings, suppressions.
* Unused item suppressions.
* Remove CloseQuietly and migrate its usages to other methods.
These other methods include:
1) New method CloseableUtils.closeAndWrapExceptions, which wraps IOExceptions
in RuntimeExceptions for callers that just want to avoid dealing with
checked exceptions. Most usages were migrated to this method, because it
looks like they were mainly attempts to avoid declaring a throws clause,
and perhaps were unintentionally suppressing IOExceptions.
2) New method CloseableUtils.closeInCatch, designed to properly close something
in a catch block without losing exceptions. Some usages from catch blocks
were migrated here, when it seemed that they were intended to avoid checked
exception handling, and did not really intend to also suppress IOExceptions.
3) New method CloseableUtils.closeAndSuppressExceptions, which sends all
exceptions to a "chomper" that consumes them. Nothing is thrown or returned.
The behavior is slightly different: with this method, _all_ exceptions are
suppressed, not just IOExceptions. Calls that seemed like they had good
reason to suppress exceptions were migrated here.
4) Some calls were migrated to try-with-resources, in cases where it appeared
that CloseQuietly was being used to avoid throwing an exception in a finally
block.
🎵 You don't have to go home, but you can't stay here... 🎵
* Remove unused import.
* Fix up various issues.
* Adjustments to tests.
* Fix null handling.
* Additional test.
* Adjustments from review.
* Fixup style stuff.
* Fix NPE caused by holder starting out null.
* Fix spelling.
* Chomp Throwables too.
* extract generic dictionary encoded column indexing and merging stuffs to pave the path towards supporting other types of dictionary encoded columns
* spotbugs and inspections fixes
* friendlier
* javadoc
* better name
* adjust
* add ColumnInspector argument to PostAggregator.getType to allow post-aggs to compute their output type based on input types
* add test for test for coverage
* simplify
* Remove unused imports.
Co-authored-by: Gian Merlino <gian@imply.io>
* latest datasketches-java and datasketches-memory
* updated versions of datasketches-java and datasketches-memory
Co-authored-by: AlexanderSaydakov <AlexanderSaydakov@users.noreply.github.com>
* better type system
* needle in a haystack
* ColumnCapabilities is a TypeSignature instead of having one, INFORMATION_SCHEMA support
* fixup merge
* more test
* fixup
* intern
* fix
* oops
* oops again
* ...
* more test coverage
* fix error message
* adjust interning, more javadocs
* oops
* more docs more better
* add MV_FILTER_ONLY SQL function, and list filter virtual column
* MV_FILTER_NONE and more tests
* formatting
* o yeah, forgot can do easy thing
* style
* hmm why was that there
* test filtering on virtual column
* style
* meh
* do it right
* good bot
* fix goldilocks bug with HashVectorGrouper improperly initializing memory that causes failure when there exists room to only grow one time
* fix unintended change
* cleanup
This PR adds a new property druid.router.sql.enable which allows the
Router to handle SQL queries when set to true.
This change does not affect Avatica JDBC requests and they are still routed
by hashing the Connection ID.
To allow parsing of the request object as a SqlQuery (contained in module druid-sql),
some classes have been moved from druid-server to druid-services with
the same package name.
This change allows the selection of a specific broker service (or broker tier) by the Router.
The newly added ManualTieredBrokerSelectorStrategy works as follows:
Check for the parameter brokerService in the query context. If this is a valid broker service, use it.
Check if the field defaultManualBrokerService has been set in the strategy. If this is a valid broker service, use it.
Move on to the next strategy
* Add a new metric query/segments/count that is not emitted by default
* docs
* test the default implementation of the metric
* fix spelling error in docs
* document the fact that query retries will result in additional metric emissions
* update using recommended text from @jihoonson
This PR refactors the code related to segment loading specifically SegmentLoader and SegmentLoaderLocalCacheManager. SegmentLoader is marked UnstableAPI which means, it can be extended outside core druid in custom extensions. Here is a summary of changes
SegmentLoader returns an instance of ReferenceCountingSegment instead of Segment. Earlier, SegmentManager was wrapping Segment objects inside ReferenceCountingSegment. That is now moved to SegmentLoader. With this, a custom implementation can track the references of segments. It also allows them to create custom ReferenceCountingSegment implementations. For this reason, the constructor visibility in ReferenceCountingSegment is changed from private to protected.
SegmentCacheManager has two additional methods called - reserve(DataSegment) and release(DataSegment). These methods let the caller reserve or release space without calling SegmentLoader#getSegment. We already had similar methods in StorageLocation and now they are available in SegmentCacheManager too which wraps multiple locations.
Refactoring to simplify the code in SegmentCacheManager wherever possible. There is no change in the functionality.
* improve groupBy query granularity translation when issued from sql layer
* fix style
* use virtual column to determine timestampResult granularity
* dont' apply postaggregators on compute nodes
* relocate constants
* fix order by correctness issue
* fix ut
* use more easier understanding code in DefaultLimitSpec
* address comment
* rollback use virtual column to determine timestampResult granularity
* fix style
* fix style
* address the comment
* add more detail document to explain the tradeoff
* address the comment
* address the comment
* add single input string expression dimension vector selector and better expression planning
* better
* fixes
* oops
* rework how vector processor factories choose string processors, fix to be less aggressive about vectorizing
* oops
* javadocs, renaming
* more javadocs
* benchmarks
* use string expression vector processor with vector size 1 instead of expr.eval
* better logging
* javadocs, surprising number of the the
* more
* simplify
This PR refactors the code for QueryRunnerFactory#mergeRunners to accept a new interface called QueryProcessingPool instead of ExecutorService for concurrent execution of query runners. This interface will let custom extensions inject their own implementation for deciding which query-runner to prioritize first. The default implementation is the same as today that takes the priority of query into account. QueryProcessingPool can also be used as a regular executor service. It has a dedicated method for accepting query execution work so implementations can differentiate between regular async tasks and query execution tasks. This dedicated method also passes the QueryRunner object as part of the task information. This hook will let custom extensions carry any state from QuerySegmentWalker to QueryProcessingPool#mergeRunners which is not possible currently.
* upgrade error-prone to 2.7.1 and support checks with Java 11+
- upgrade error-prone to 2.7.1
- support running error-prone with Java 11 and above using -Xplugin
instead of custom compiler
- add compiler arguments to ignore warnings/errors in Java 15/16
- introduce strictCompile property to enable strict profiles since we
now need multiple strict profiles for Java 8
- properly exclude all generated source files from error-prone
- fix druid-processing overriding annotation processors from parent pom
- fix druid-core disabling most non-default checks
- align plugin and annotation errorprone versions
- fix / suppress additional issues found by error-prone:
* fix bug in SeekableStreamSupervisor initializing ArrayList size with
the taskGroupdId
* fix missing @Override annotations
- remove outdated compiler plugin in benchmarks
- remove deleted ParameterPackage error-prone rule
- re-enable checks on benchmark module as well
* fix IntelliJ inspections
* disable LongFloatConversion due to bug in error-prone with JDK 8
* add comment about InsecureCrypto
* enrich expression cache key information to support expressions which depend on external state such as lookups
* cache rules everything around me
* low carb
* rename
* Fix is null selector returning incorrect value for Long data type
* Fix style errors
* Refactor getObject method to also cache null column values
* Make lastInput variable nullable
* Refactor unit test
* Use new boolean lastInputIsNull instead of Long for lastInput to avoid boxing
* Refactor to remove Long for input variable
* Make a separate null caching variable
* Cleaner null caching implementation
* fix count and average SQL aggregators on constant virtual columns
* style
* even better, why are we tracking virtual columns in aggregations at all if we have a virtual column registry
* oops missed a few
* remove unused
* this will fix it
* SQL timeseries no longer skip empty buckets with all granularity
* add comment, fix tests
* the ol switcheroo
* revert unintended change
* docs and more tests
* style
* make checkstyle happy
* docs fixes and more tests
* add docs, tests for array_agg
* fixes
* oops
* doc stuffs
* fix compile, match doc style
* Fix vectorized cardinality bug on certain string columns.
Fixes a bug introduced in #11182, related to the fact that in some cases,
ColumnProcessors.makeVectorProcessor will call "makeObjectProcessor"
instead of "makeSingleValueDimensionProcessor" or
"makeMultiValueDimensionProcessor". CardinalityVectorProcessorFactory
improperly ignored calls to "makeObjectProcessor".
In addition to fixing the bug, I added this detail to the javadocs for
VectorColumnProcessorFactory, to prevent others from running into the
same thing in the future. They do not currently call out this case.
* Improve test coverage.
* Additional fixes.
* ARRAY_AGG sql aggregator function
* add javadoc
* spelling
* review stuff, return null instead of empty when nil input
* review stuff
* Update sql.md
* use type inference for finalize, refactor some things
* Vectorize the cardinality aggregator.
Does not include a byRow implementation, so if byRow is true then
the aggregator still goes through the non-vectorized path.
Testing strategy:
- New tests that exercise both styles of "aggregate" for supported types.
- Some existing tests have also become active (note the deleted
"cannotVectorize" lines).
* Adjust whitespace.
* Vectorize the DataSketches quantiles aggregator.
Also removes synchronization for the BufferAggregator and VectorAggregator
implementations, since it is not necessary (similar to #11115).
Extends DoublesSketchAggregatorTest and DoublesSketchSqlAggregatorTest
to run all test cases in vectorized mode.
* Style fix.
* Add a way to retrieve UTF-8 bytes directly via DimensionDictionarySelector.
The idea is that certain operations (like count distinct on strings) will
be faster if they are able to run directly on UTF-8 bytes instead of on
Java Strings decoded by "lookupName".
* Add license header.
* Updates suggested by robots.
Lexicographic ordering of UTF-8 byte sequences and in-memory UTF-16
strings are equivalent. So, we can skip the (expensive) conversion and
get an equivalent result. Thank you, Unicode!
The prior code did not include separation between values, and encoded
null ambiguously. This patch fixes both of those issues by encoding
strings as length + value instead of just value.
I think cache key computation was OK prior to #9800. Prior to that
patch, the cache key was computed using CacheKeyBuilder.appendStrings,
which encodes strings as UTF-8 and inserts a separator byte (0xff)
between them that cannot appear in a UTF-8 stream.
* InDimFilter: Fix NPE involving certain Set types.
Normally, InDimFilters that come from JSON have HashSets for "values".
However, programmatically-generated filters (like the ones from #11068)
may use other set types. Some set types, like TreeSets with natural
ordering, will throw NPE on "contains(null)", which causes the
InDimFilter's ValueMatcher to throw NPE if it encounters a null value.
This patch adds code to detect if the values set can support
contains(null), and if not, wrap that in a null-checking lambda.
Also included:
- Remove unneeded NullHandling.needsEmptyToNull method.
- Update IndexedTableJoinable to generate a TreeSet that does not
require lambda-wrapping. (This particular TreeSet is how I noticed
the bug in the first place.)
* Test fixes.
* Improve test coverage
* add experimental expression aggregator
* add test
* fix lgtm
* fix test
* adjust test
* use not null constant
* array_set_concat docs
* add equals and hashcode and tostring
* fix it
* spelling
* do multi-value magic for expression agg, more javadocs, tests
* formatting
* fix inspection
* more better
* nullable
* Enable rewriting certain inner joins as filters.
The main logic for doing the rewrite is in JoinableFactoryWrapper's
segmentMapFn method. The requirements are:
- It must be an inner equi-join.
- The right-hand columns referenced by the condition must not contain any
duplicate values. (If they did, the inner join would not be guaranteed
to return at most one row for each left-hand-side row.)
- No columns from the right-hand side can be used by anything other than
the join condition itself.
HashJoinSegmentStorageAdapter is also modified to pass through to
the base adapter (even allowing vectorization!) in the case where 100%
of join clauses could be rewritten as filters.
In support of this goal:
- Add Query getRequiredColumns() method to help us figure out whether
the right-hand side of a join datasource is being used or not.
- Add JoinConditionAnalysis getRequiredColumns() method to help us
figure out if the right-hand side of a join is being used by later
join clauses acting on the same base.
- Add Joinable getNonNullColumnValuesIfAllUnique method to enable
retrieving the set of values that will form the "in" filter.
- Add LookupExtractor canGetKeySet() and keySet() methods to support
LookupJoinable in its efforts to implement the new Joinable method.
- Add "enableRewriteJoinToFilter" feature flag to
JoinFilterRewriteConfig. The default is disabled.
* Test improvements.
* Test fixes.
* Avoid slow size() call.
* Remove invalid test.
* Fix style.
* Fix mistaken default.
* Small fixes.
* Fix logic error.
PR #10936 renamed BitmapBenchmark, the parent of a couple of bitmap tests, to
BitmapOperationTest. This patch renames it to BitmapOperationTestBase so JUnit
doesn't pick it up as a test case. When JUnit picks it up, it becomes a flaky
test, since its behavior and correctness depends on whether it runs before
or after its subclasses.
Regression introduced in #11004 due to overzealous optimization. Even though
we replaced stateful usage of ByteBuffer with stateless usage of Memory, we
still need to create a new object on "duplicate" due to semantics of setBuffer.
* fix nested groupby got empty result when using virtual column
* move to query.getVirtualColumns().wrap instead of new VirtualizedColumnSelectorFactory
* move test to GroupByQueryRunnerTest
* Update processing/src/test/java/org/apache/druid/query/groupby/GroupByQueryRunnerTest.java
Co-authored-by: huagnhui.bigrey <huanghui.bigrey@bytedance.com>
Co-authored-by: Jihoon Son <jihoonson@apache.org>
* Vectorize LongDeserializers.
Also, add many more tests.
* more faster
* more more faster
* more cleanup
* fixes
* forbidden
* benchmark style
* idk why
* adjust
* add preconditions for value >= 0 for writers
* add 64 bit exception
Co-authored-by: Gian Merlino <gian@imply.io>
* DruidInputSource: Fix issues in column projection, timestamp handling.
DruidInputSource, DruidSegmentReader changes:
1) Remove "dimensions" and "metrics". They are not necessary, because we
can compute which columns we need to read based on what is going to
be used by the timestamp, transform, dimensions, and metrics.
2) Start using ColumnsFilter (see below) to decide which columns we need
to read.
3) Actually respect the "timestampSpec". Previously, it was ignored, and
the timestamp of the returned InputRows was set to the `__time` column
of the input datasource.
(1) and (2) together fix a bug in which the DruidInputSource would not
properly read columns that are used as inputs to a transformSpec.
(3) fixes a bug where the timestampSpec would be ignored if you attempted
to set the column to something other than `__time`.
(1) and (3) are breaking changes.
Web console changes:
1) Remove "Dimensions" and "Metrics" from the Druid input source.
2) Set timestampSpec to `{"column": "__time", "format": "millis"}` for
compatibility with the new behavior.
Other changes:
1) Add ColumnsFilter, a new class that allows input readers to determine
which columns they need to read. Currently, it's only used by the
DruidInputSource, but it could be used by other columnar input sources
in the future.
2) Add a ColumnsFilter to InputRowSchema.
3) Remove the metric names from InputRowSchema (they were unused).
4) Add InputRowSchemas.fromDataSchema method that computes the proper
ColumnsFilter for given timestamp, dimensions, transform, and metrics.
5) Add "getRequiredColumns" method to TransformSpec to support the above.
* Various fixups.
* Uncomment incorrectly commented lines.
* Move TransformSpecTest to the proper module.
* Add druid.indexer.task.ignoreTimestampSpecForDruidInputSource setting.
* Fix.
* Fix build.
* Checkstyle.
* Misc fixes.
* Fix test.
* Move config.
* Fix imports.
* Fixup.
* Fix ShuffleResourceTest.
* Add import.
* Smarter exclusions.
* Fixes based on tests.
Also, add TIME_COLUMN constant in the web console.
* Adjustments for tests.
* Reorder test data.
* Update docs.
* Update docs to say Druid 0.22.0 instead of 0.21.0.
* Fix test.
* Fix ITAutoCompactionTest.
* Changes from review & from merging.
* expression filter support for vectorized query engines
* remove unused codes
* more tests
* refactor, more tests
* suppress
* more
* more
* more
* oops, i was wrong
* comment
* remove decorate, object dimension selector, more javadocs
* style
* fix SQL issue for group by queries with time filter that gets optimized to false
* short circuit always false in CombineAndSimplifyBounds
* adjust
* javadocs
* add preconditions for and/or filters to ensure they have children
* add comments, remove preconditions
* where filter left first draft
* Revert changes in calcite test
* Refactor a bit
* Fixing the Tests
* Changes
* Adding tests
* Add tests for correlated queries
* Add comment
* Fix typos
* Fix runtime error when IndexedTableJoinMatcher matches long selector to unique string index.
The issue arises when matching against a long selector on the left-hand side to a string
typed Index on the right-hand side, and when that Index also returns true from areKeysUnique.
In this case, IndexedTableJoinMatcher would generate a ConditionMatcher that implements
matchSingleRow by calling findUniqueLong on the Index. This is inappropriate because the Index
is actually string typed. The fix is to check the type of the Index before deciding how to
implement the ConditionMatcher.
The patch adds "testMatchSingleRowToUniqueStringIndex" to IndexedTableJoinMatcherTest, which
explores this case.
* Update tests.
* Granularity: Introduce primitive-typed bucketStart, increment methods.
Saves creation of unnecessary DateTime objects in timestamp_floor and
timestamp_ceil expressions.
* Fix style.
* Amp up the test coverage.
They all use Long.compare, but they don't need to. Changing to
regular comparisons simplifies the code and also removes branches.
(Internally, Long.compare has two branches.)
* Vectorized theta sketch aggregator.
Also a refactoring of BufferAggregator and VectorAggregator such that
they share a common interface, BaseBufferAggregator. This allows
implementing both in the same file with an abstract + dual subclass
structure.
* Rework implementation to use composition instead of inheritance.
* Rework things to enable working properly for both complex types and
regular types.
Involved finally moving makeVectorProcessor from DimensionHandlerUtils
into ColumnProcessors and harmonizing the two things.
* Add missing method.
* Style and name changes.
* Fix issues from inspections.
* Fix style issue.
* Retain order of AND, OR filter children.
If we retain the order, it enables short-circuiting. People can put a
more selective filter earlier in the list and lower the chance that
later filters will need to be evaluated.
Short-circuiting was working before #9608, which switched to unordered
sets to solve a different problem. This patch tries to solve that
problem a different way.
This patch moves filter simplification logic from "optimize" to
"toFilter", because that allows the code to be shared with Filters.and
and Filters.or. The simplification has become more complicated and so
it's useful to share it.
This patch also removes code from CalciteCnfHelper that is no longer
necessary because Filters.and and Filters.or are now doing the work.
* Fixes for inspections.
* Fix tests.
* Back to a Set.
* ready to test
* tested on dev cluster
* tested
* code review
* add UTs
* add UTs
* ut passed
* ut passed
* opti imports
* done
* done
* fix checkstyle
* modify uts
* modify logs
* changing the package of SegmentLazyLoadFailCallback.java to org.apache.druid.segment
* merge from master
* modify import orders
* merge from master
* merge from master
* modify logs
* modify docs
* modify logs to rerun ci
* modify logs to rerun ci
* modify logs to rerun ci
* modify logs to rerun ci
* modify logs to rerun ci
* modify logs to rerun ci
* modify logs to rerun ci
* modify logs to rerun ci
Co-authored-by: yuezhang <yuezhang@freewheel.tv>
* OrFilter: Properly handle child matchers that return the original mask.
This happens when a child matcher is literally true (for example,
BooleanVectorValueMatcher). In this case, OrFilter would throw this
exception from its call to removeAll while processing the next filter:
java.lang.IllegalStateException: 'other' must be a different instance from 'this'
Also update the javadocs for VectorValueMatcher to call out that the
returned object may be the same as the input mask.
* Fix style.
No existing code relies on being able to call these methods in this way.
The new tests exhaustively test all vectors up to size 7, and also test
behavior the run-on-self behavior that has been adjusted by this patch.
* Tidy up query error codes
* fix tests
* Restore query exception type in JsonParserIterator
* address review comments; add a comment explaining the ugly switch
* fix test
* cache expression selector results by associating vector expression bindings to underlying vector offset
* better coverage, fix floats
* style
* stupid bot
* stupid me
* more test
* intellij threw me under the bus when it generated those junit methods
* narrow interface instead of passing around offset
* javascript filter result convert to java boolean
* use type convert replace script convert, and add more unit test
Co-authored-by: qinzhen <qinzhen@kuaishou.com>
* Remove redundant IncrementalIndex.Builder
* Parametrize incremental index tests and benchmarks
- Reveal and fix a bug in OffheapIncrementalIndex
* Fix forbiddenapis error: Forbidden method invocation: java.lang.String#format(java.lang.String,java.lang.Object[]) [Uses default locale]
* Fix Intellij errors: declared exception is never thrown
* Add documentation and validate before closing objects on tearDown.
* Add documentation to OffheapIncrementalIndexTestSpec
* Doc corrections and minor changes.
* Add logging for generated rows.
* Refactor new tests/benchmarks.
* Improve IncrementalIndexCreator documentation
* Add required tests for DataGenerator
* Revert "rollupOpportunity" to be a string
* Multiphase merge for IndexMergerV9
* JSON fix
* Cleanup temp files
* Docs
* Address logging and add IT
* Fix spelling and test unloader datasource name
* First draft of grouping_id function
* Add more tests and documentation
* Add calcite tests
* Fix travis failures
* bit of a change
* Add documentation
* Fix typos
* typo fix
* Add context dimension to DefaultQueryMetrics
* remove redundant addition of context dimension from DruidMetrics now that QueryMetrics adds it by default
* update SearchQueryMetrics to reflect the same pattern as other default dimensions in QueryMetrics
* add PublicApi annotation for context in QueryMetrics Interface
* Add new coordinator metrics for duty runtimes
* fix spelling for a constant variable value
* add comment clarifying why the global runtime metric is emitted where it is
* Remove duty alias in lieu of using the class name for metrics
* fix docs
* CoordinatorStats tests + add duty stats to accumulate() logic
* support multi-line text
* add test cases
* split json text into lines case by case
* improve exception handle
* fix CI
* use IntermediateRowParsingReader as base of JsonReader
* update doc
* ignore the non-immutable field in test case
* add more test cases
* mark `lineSplittable` as final
* fix testcases
* fix doc
* add a test case for SqlReader
* return all raw columns when exception occurs
* fix CI
* fix test cases
* resolve review comments
* handle ParseException returned by index.add
* apply Iterables.getOnlyElement
* fix CI
* fix test cases
* improve code in more graceful way
* fix test cases
* fix test cases
* add a test case to check multiple json string in one text block
* fix inspection check
* support for vectorizing expressions with non-existent inputs, more consistent type handling for non-vectorized expressions
* inspector
* changes
* more test
* clean
* Introduce a Configurable Index Type
* Change to @UnstableApi
* Add AppendableIndexSpecTest
* Update doc
* Add spelling exception
* Add tests coverage
* Revert some of the changes to reduce diff
* Minor fixes
* Update getMaxBytesInMemoryOrDefault() comment
* Fix typo, remove redundant interface
* Remove off-heap spec (postponed to a later PR)
* Add javadocs to AppendableIndexSpec
* Describe testCreateTask()
* Add tests for AppendableIndexSpec within TuningConfig
* Modify hashCode() to conform with equals()
* Add comment where building incremental-index
* Add "EqualsVerifier" tests
* Revert some of the API back to AppenderatorConfig
* Don't use multi-line comments
* Remove knob documentation (deferred)
* Proposed changes for making joins cacheable
* Add unit tests
* Fix tests
* simplify logic
* Pull empty byte array logic out of CachingQueryRunner
* remove useless null check
* Minor refactor
* Fix tests
* Fix segment caching on Broker
* Move join cache key computation in Broker
Move join cache key computation in Broker from ResultLevelCachingQueryRunner to CachingClusteredClient
* Fix compilation
* Review comments
* Add more tests
* Fix inspection errors
* Pushed condition analysis to JoinableFactory
* review comments
* Disable join caching for broker and add prefix key to BroadcastSegmentIndexedTable
* Remove commented lines
* Fix populateCache
* Disable caching for selective datasources
Refactored the code so that we can decide at the data source level, whether to enable cache for broker or data nodes
* RowBasedIndexedTable: Add specialized index types for long keys.
Two new index types are added:
1) Use an int-array-based index in cases where the difference between
the min and max values isn't too large, and keys are unique.
2) Use a Long2ObjectOpenHashMap (instead of the prior Java HashMap) in
all other cases.
In addition:
1) RowBasedIndexBuilder, a new class, is responsible for picking which
index implementation to use.
2) The IndexedTable.Index interface is extended to support using
unboxed primitives in the unique-long-keys case, and callers are
updated to use the new functionality.
Other key types continue to use indexes backed by Java HashMaps.
* Fixup logic.
* Add tests.
* vectorize remaining math expressions
* fixes
* remove cannotVectorize() where no longer true
* disable vectorized groupby for numeric columns with nulls
* fixes
* Store hash partition function in dataSegment and allow segment pruning only when hash partition function is provided
* query context
* fix tests; add more test
* javadoc
* docs and more tests
* remove default and hadoop tests
* consistent name and fix javadoc
* spelling and field name
* default function for partitionsSpec
* other comments
* address comments
* fix tests and spelling
* test
* doc
* Include Sequence-building time in CPU time metric.
Meaningful work can be done while building Sequences, and we should
count this work. On the Broker, this includes subquery processing
work done by the mergeResults call of the GroupByQueryQueryToolChest.
* Add test.
* push down ValueType to ExprType conversion, tidy up
* determine expr output type for given input types
* revert unintended name change
* add nullable
* tidy up
* fixup
* more better
* fix signatures
* naming things is hard
* fix inspection
* javadoc
* make default implementation of Expr.getOutputType that returns null
* rename method
* more test
* add output for contains expr macro, split operation and function auto conversion
* Add IndexMergerRollupTest
This changelist adds a test to merge indexes with StringFirst/StringLast aggregator.
* Fix StringFirstAggregateCombiner/StringLastAggregateCombiner
The segment-level type for stringFirst/stringLast is SerializablePairLongString,
not String. This changelist fixes it.
* Fix EarliestLatestAnySqlAggregator to handle COMPLEX type
This changelist allows EarliestLatestAnySqlAggregator to accept COMPLEX
type as an operand. For its return type, we set it to VARCHAR, since
COMPLEX column is only generated by stringFirst/stringLast during ingestion
rollup.
* Return value with smaller timestamp in StringFirstAggregatorFactory.combine function
* Add integration tests for stringFirst/stringLast during ingestion
* Use one EarliestLatestReturnTypeInference instance
Co-authored-by: Joy Kent <joy@automonic.ai>
* Fix handling of 'join' on top of 'union' datasources.
The problem is that unions are typically rewritten into a series of
individual queries on the underlying tables, but this isn't done when
the union is wrapped in a join.
The main changes are in UnionQueryRunner:
1) Replace an instanceof UnionQueryRunner check with DataSourceAnalysis.
2) Replace a "query.withDataSource" call with a new function, "Queries.withBaseDataSource".
Together, these enable UnionQueryRunner to "see through" a join.
* Tests.
* Adjust heap sizes for integration tests.
* Different approach, more tests.
* Tweak.
* Styling.
* Add support for all partitioing schemes for auto compaction
* annotate last compaction state for multi phase parallel indexing
* fix build and tests
* test
* better home
* better type tracking: add typed postaggs, finalized types for agg factories
* more javadoc
* adjustments
* transition to getTypeName to be used exclusively for complex types
* remove unused fn
* adjust
* more better
* rename getTypeName to getComplexTypeName
* setup expression post agg for type inference existing
* more javadocs
* fixup
* oops
* more test
* more test
* more comments/javadoc
* nulls
* explicitly handle only numeric and complex aggregators for incremental index
* checkstyle
* more tests
* adjust
* more tests to showcase difference in behavior
* timeseries longsum array
* Make NUMERIC_HASHING_THRESHOLD configurable
Change the default numeric hashing threshold to 1 and make it configurable.
Benchmarks attached to this PR show that binary searches are not more faster
than doing a set contains check. The attached flamegraph shows the amount of
time a query spent in the binary search. Given the benchmarks, we can expect
to see roughly a 2x speed up in this part of the query which works out to
~ a 10% faster query in this instance.
* Remove NUMERIC_HASHING_THRESHOLD
* Remove stale docs
Previously, this was disallowed, because expressions treated multi-values
as nulls. But now, if there's a single multi-value column that can be
mapped over, it's okay to use the index. Expression selectors already do
this.
* Optimize large InDimFilters
For large InDimFilters, in default mode, the filter does a linear check of the
set to see if it contains either an empty or null. If it does, the empties are
converted to nulls by passing through the entire list again.
Instead of this, in default mode, we attempt to remove an empty string from the
values that are passed to the InDimFilter. If an empty string was removed, we
add null to the set
* code review
* Revert "code review"
This reverts commit 61fe33ebf7.
* code review - less brittle
* Segment backed broadcast join IndexedTable
* fix comments
* fix tests
* sharing is caring
* fix test
* i hope this doesnt fix it
* filter by schema to maybe fix test
* changes
* close join stuffs so it does not leak, allow table to directly make selector factory
* oops
* update comment
* review stuffs
* better check
* Add "offset" parameter to the Scan query.
It works by doing the query as normal and then throwing away the first
"offset" number of rows on the broker.
* Fix constructor call.
* Fix up JSONs.
* Fix call to ScanQuery.
* Doc update.
* Fix javadocs.
* Spotbugs, LGTM suppressions.
* Javadocs.
* Fix suppression.
* Stabilize Scan query result order, add tests.
* Update LGTM comment.
* Fixup.
* Test different batch sizes too.
* Nicer tests.
* Fix comment.
1) lookupId could return IDs beyond maxId if called with a recently added value.
2) getRow could return an ID for null beyond maxId, if null was recently
encountered in a dimension that initially didn't appear at all. (In this case,
the dictionary ID for null can be > 0).
Also add a comment explaining how this stuff is supposed to work.
* fix bug with realtime expressions on sparse string columns
* fix test
* add comment back
* push capabilities for dimensions to dimension indexers since they know things
* style
* style
* fixes
* getting a bit carried away
* missed one
* fix it
* benchmark build fix
* review stuffs
* javadoc and comments
* add comment
* more strict check
* fix missed usaged of impl instead of interface
* LongMaxVectorAggregator support and test case.
* DoubleMinVectorAggregator and test cases.
* DoubleMaxVectorAggregator and unit test.
* FloatMinVectorAggregator and FloatMaxVectorAggregator.
* Documentation update to include the other vector aggregators.
* Bug fix.
* checkstyle formatting fixes.
* CalciteQueryTest cases update.
* Separate test classes for FloatMaxAggregation and FloatMniAggregation.
* remove the cannotVectorize for float max/min aggregator in test.
* Tests in GroupByQueryRunner, GroupByTimeseriesQueryRunner and TimeseriesQueryRunner.
* Combine InDimFilter, InFilter.
There are two motivations:
1. Ensure that when HashJoinSegmentStorageAdapter compares its Filter
to the original one, and it is an "in" type, the comparison is by
reference and does not need to check deep equality. This is useful
when the "in" filter is very large.
2. Simplify things. (There isn't a great reason for the DimFilter and
Filter logic to be separate, and combining them reduces some
duplication.)
* Fix test.
* Add "offset" parameter to GroupBy query.
It works by doing the query as normal and then throwing away the first
"offset" number of rows on the broker.
* Stabilize GroupBy sorts.
* Fix inspections.
* Fix suppression.
* Fixups.
* Move TopNSequence to druid-core.
* Addl comments.
* NumberedElement equals verification.
* Changes from review.
* Fix minor formatting in docs.
* Add Nullhandling initialization for test to run from IDE.
* Vectorize longMin aggregator.
- A new vectorized class for the vectorized long min aggregator.
- Changes to AggregatorFactory to support vectorize functionality.
- Few changes to schema evolution test to add LongMinAggregatorFactory.
* Add longSum to the supported vectorized aggregator implementations.
* Add MIN() long min to calcite query test that can vectorize.
* Add simple long aggregations test.
* Fixup formatting per checkstyle guide.
* fixup and add more tests for long min aggregator.
* Override test for groupBy since timestamps are handled differently.
* Null compatibility check in test.
* Review comment: Add a test case to LongMinAggregationTest.
* support unit suffix on byte-related properties
* add doc
* change default value of byte-related properites in example files
* fix coding style
* fix doc
* fix CI
* suppress spelling errors
* improve code according to comments
* rename Bytes to HumanReadableBytes
* add getBytesInInt to get value safely
* improve doc
* fix problem reported by CI
* fix problem reported by CI
* resolve code review comments
* improve error message
* improve code & doc according to comments
* fix CI problem
* improve doc
* suppress spelling check errors
* Fix timeseries query constructor when postAggregator has an expression reading timestamp result column
* fix npe
* Fix postAgg referencing timestampResultField and add a test for it
* fix test
* doc
* revert doc
* Ensure that join filter pre-analysis operates on optimized filters, add DimFilter.toOptimizedFilter
* Remove aggressive equality check that was used for testing
* Use Suppliers.memoize
* Checkstyle
* Fix RetryQueryRunner to actually do the job
* more javadoc
* fix test and checkstyle
* don't combine for testing
* address comments
* fix unit tests
* always initialize response context in cachingClusteredClient
* fix subquery
* address comments
* fix test
* query id for builders
* make queryId optional in the builders and ClusterQueryResult
* fix test
* suppress tests and unused methods
* exclude groupBy builder
* fix jacoco exclusion
* add tests for builders
* address comments
* don't truncate
* Join filter pre-analysis simplifications and sanity checks.
- At pre-analysis time, only compute pre-analysis for the innermost
root query, since this is the one that will run on the join that involves
the base datasource. Previously, pre-analyses were computed for multiple
levels of the query, some of which were unnecessary.
- Remove JoinFilterPreAnalysisGroup and join query level gathering code,
since they existed to support precomputation of multiple pre-analyses.
- Embed JoinFilterPreAnalysisKey into JoinFilterPreAnalysis and use it to
sanity check at processing time that the correct pre-analysis was done.
Tangentially related changes:
- Remove prioritizeAndLaneQuery functionality from LocalQuerySegmentWalker.
The computed priority and lanes were not being used.
- Add "getBaseQuery" method to DataSourceAnalysis to support identification
of the proper subquery for filter pre-analysis.
* Fix compilation errors.
* Adjust tests.
* Filter http requests by http method
Add a config that allows a user which http methods to allow against their
Druid server.
Druid will only accept http requests with the method: GET, PUT, POST, DELETE
and OPTIONS.
If a Druid admin wants to allow other methods, they can do so by using the
ServerConfig#allowedHttpMethods config.
If a Druid user would like to disallow OPTIONS, this can be done by changing
the AuthConfig#allowUnauthenticatedHttpOptions config
* Exclude OPTIONS from always supported HTTP methods
Add HEAD as an allowed method for web console e2e tests
* fix docs
* fix security IT
* Actually fix the web console e2e tests
* Ignore icode coverage for nitialization classes
* code review
* IntelliJ inspection and checkstyle rule for "Collection.EMPTY_* field accesses replaceable with Collections.empty*()"
* Reverted checkstyle rule
* Added tests to pass CI
* Codestyle
* ROUND and having comparators correctly handle doubles
Double.NaN, Double.POSITIVE_INFINITY and Double.NEGATIVE_INFINITY are not real
numbers. Because of this, they can not be converted to BigDecimal and instead
throw a NumberFormatException.
This change adds support for calculations that produce these numbers either
for use in the `ROUND` function or the HavingSpecMetricComparator by not
attempting to convert the number to a BigDecimal.
The bug in ROUND was first introduced in #7224 where we added the ability to
round to any decimal place. This PR changes the behavior back to using
`Math.round` if we recognize a number that can not be converted to a
BigDecimal.
* Add tests and fix spellcheck
* update error message in ExpressionsTest
* Address comments
* fix up round for infinity
* round non numeric doubles returns a double
* fix spotbugs
* Update docs/misc/math-expr.md
* Update docs/querying/sql.md
* Remove LegacyDataSource.
Its purpose was to enable deserialization of strings into TableDataSources.
But we can do this more straightforwardly with Jackson annotations.
* Slight test improvement.
The parameters generator uses CompressionStrategy.noNoneValues() instead
of CompressionStrategyTest.compressionStrategies() which wrapped each
strategy in a single element array. This improves readability of the
test.
* make joinables closeable
* tests and adjustments
* refactor to make join stuffs impelement ReferenceCountedObject instead of Closable, more tests
* fixes
* javadocs and stuff
* fix bugs
* more test
* fix lgtm alert
* simplify
* fixup javadoc
* review stuffs
* safeguard against exceptions
* i hate this checkstyle rule
* make IndexedTable extend Closeable
* remove incorrect and unnecessary overrides from BooleanVectorValueMatcher
* add test case
* add unit tests for ... part of VectorValueMatcherColumnProcessorFactory
* Update VectorValueMatcherColumnProcessorFactoryTest.java
* move benchmark data generator into druid-processing, add a GeneratorInputSource to fill up a cluster with data
* newlines
* make test coverage not fail maybe
* remove useless test
* Update pom.xml
* Update GeneratorInputSourceTest.java
* less passive aggressive test names
* fix groupBy with literal in subquery grouping
* fix groupBy with literal in subquery grouping
* fix groupBy with literal in subquery grouping
* address comments
* update javadocs
* Fix join
* Fix Subquery could not be converted to groupBy query
* Fix Subquery could not be converted to groupBy query
* Fix Subquery could not be converted to groupBy query
* Fix Subquery could not be converted to groupBy query
* Fix Subquery could not be converted to groupBy query
* Fix Subquery could not be converted to groupBy query
* Fix Subquery could not be converted to groupBy query
* Fix Subquery could not be converted to groupBy query
* add tests
* address comments
* fix failing tests
* Add REGEXP_LIKE, fix empty-pattern bug in REGEXP_EXTRACT.
- Add REGEXP_LIKE function that returns a boolean, and is useful in
WHERE clauses.
- Fix REGEXP_EXTRACT return type (should be nullable; causes incorrect
filter elision).
- Fix REGEXP_EXTRACT behavior for empty patterns: should always match
(previously, they threw errors).
- Improve error behavior when REGEXP_EXTRACT and REGEXP_LIKE are passed
non-literal patterns.
- Improve documentation of REGEXP_EXTRACT.
* Changes based on PR review.
* Fix arg check.
* Important fixes!
* Add speller.
* wip
* Additional tests.
* Fix up tests.
* Add validation error tests.
* Additional tests.
* Remove useless call.
* - GroupByQueryEngineV2: Fix leak of intermediate processing buffer when
exceptions are thrown before result sequence is created.
- PooledTopNAlgorithm: Fix leak of intermediate processing buffer when
exceptions are thrown before the PooledTopNParams object is created.
- BlockingPool: Remove unused "take" methods.
* Add tests to verify that buffers have been returned.
* Fix various Yielder leaks.
- CombiningSequence leaked the input yielder from "toYielder" if it ran
into an exception while accumulating the last value from the input
yielder.
- MergeSequence leaked input yielders from "toYielder" if it ran into
an exception while building the initial priority queue.
- ScanQueryRunnerFactory leaked the input yielder in its
"priorityQueueSortAndLimit" strategy if it ran into an exception
while scanning and sorting.
- YieldingSequenceBase.accumulate chomped IOExceptions thrown in
"accumulate" during yielder closing.
* Add tests.
* Fix braces.
* Refactor JoinFilterAnalyzer
This patch attempts to make it easier to follow the join filter analysis code
with the hope of making it easier to add rewrite optimizations in the future.
To keep the patch small and easy to review, this is the first of at least 2
patches that are planned.
This patch adds a builder to the Pre-Analysis, so that it is easier to
instantiate the preAnalysis. It also moves some of the filter normalization
code out to Fitlers with associated tests.
* fix tests
* Refactor JoinFilterAnalyzer - part 2
This change introduces the following components:
* RhsRewriteCandidates - a wrapper for a list of candidates and associated
functions to operate on the set of candidates.
* JoinableClauses - a wrapper for the list of JoinableClause that represent
a join condition and the associated functions to operate on the clauses.
* Equiconditions - a wrapper representing the equiconditions that are used
in the join condition.
And associated test changes.
This refactoring surfaced 2 bugs:
- Missing equals and hashcode implementation for RhsRewriteCandidate, thus
allowing potential duplicates in the rhs rewrite candidates
- Missing Filter#supportsRequiredColumnRewrite check in
analyzeJoinFilterClause, which could result in UnsupportedOperationException
being thrown by the filter
* fix compile error
* remove unused class
* Refactor JoinFilterAnalyzer - Correlations
Move the correlation related code out into it's own class so it's easier
to maintain.
Another patch should follow this one so that the query path uses the
correlation object instead of it's underlying maps.
* Optimize join queries where filter matches nothing
Fixes#9787
This PR changes the Joinable interface to return an Optional set of correlated
values for a column.
This allows the JoinFilterAnalyzer to differentiate between the case where the
column has no matching values and when the column could not find matching
values.
This PR chose not to distinguish between cases where correlated values could
not be computed because of a config that has this behavior disabled or because
of user error - like a column that could not be found. The reasoning was that
the latter is likely an error and the non filter pushdown path will surface the
error if it is.
* Refactor JoinFilterAnalyzer
This patch attempts to make it easier to follow the join filter analysis code
with the hope of making it easier to add rewrite optimizations in the future.
To keep the patch small and easy to review, this is the first of at least 2
patches that are planned.
This patch adds a builder to the Pre-Analysis, so that it is easier to
instantiate the preAnalysis. It also moves some of the filter normalization
code out to Fitlers with associated tests.
* fix tests
* Refactor JoinFilterAnalyzer - part 2
This change introduces the following components:
* RhsRewriteCandidates - a wrapper for a list of candidates and associated
functions to operate on the set of candidates.
* JoinableClauses - a wrapper for the list of JoinableClause that represent
a join condition and the associated functions to operate on the clauses.
* Equiconditions - a wrapper representing the equiconditions that are used
in the join condition.
And associated test changes.
This refactoring surfaced 2 bugs:
- Missing equals and hashcode implementation for RhsRewriteCandidate, thus
allowing potential duplicates in the rhs rewrite candidates
- Missing Filter#supportsRequiredColumnRewrite check in
analyzeJoinFilterClause, which could result in UnsupportedOperationException
being thrown by the filter
* fix compile error
* remove unused class
* Refactor JoinFilterAnalyzer
This patch attempts to make it easier to follow the join filter analysis code
with the hope of making it easier to add rewrite optimizations in the future.
To keep the patch small and easy to review, this is the first of at least 2
patches that are planned.
This patch adds a builder to the Pre-Analysis, so that it is easier to
instantiate the preAnalysis. It also moves some of the filter normalization
code out to Fitlers with associated tests.
* fix tests
* add flag to flattenSpec to keep null columns
* remove changes to inputFormat interface
* add comment
* change comment message
* update web console e2e test
* move keepNullColmns to JSONParseSpec
* fix merge conflicts
* fix tests
* set keepNullColumns to false by default
* fix lgtm
* change Boolean to boolean, add keepNullColumns to hash, add tests for keepKeepNullColumns false + true with no nuulul columns
* Add equals verifier tests
* Fix potential NPEs in joins
intelliJ reported issues with potential NPEs. This was first hit in testing
with a filter being pushed down to the left hand table when joining against
an indexed table.
* More null check cleanup
* Optimize filter value rewrite for IndexedTable
* Add unit tests for LookupJoinable
* Add tests for IndexedTableJoinable
* Add non null assert for dimension selector
* Supress null warning in LookupJoinMatcher
* remove some null checks on hot path
* optimize FileWriteOutBytes to avoid high sys cpu
* optimize FileWriteOutBytes to avoid high sys cpu -- remove IOException
* optimize FileWriteOutBytes to avoid high sys cpu -- remove IOException in writeOutBytes.size
* Revert "optimize FileWriteOutBytes to avoid high sys cpu -- remove IOException in writeOutBytes.size"
This reverts commit 965f7421
* Revert "optimize FileWriteOutBytes to avoid high sys cpu -- remove IOException"
This reverts commit 149e08c0
* optimize FileWriteOutBytes to avoid high sys cpu -- avoid IOEception never thrown check
* Fix size counting to handle IOE in FileWriteOutBytes + tests
* remove unused throws IOException in WriteOutBytes.size()
* Remove redundant throws IOExcpetion clauses
* Parameterize IndexMergeBenchmark
Co-authored-by: huanghui.bigrey <huanghui.bigrey@bytedance.com>
Co-authored-by: Suneet Saldanha <suneet.saldanha@imply.io>
* fixes for inline subqueries when multi-value dimension is present
* fix test
* allow missing capabilities for vectorized group by queries to be treated as single dims since it means that column doesnt exist
* add comment
* fix issue with group by limit pushdown for extractionFn, expressions, joins, etc
* remove unused
* fix test
* revert unintended change
* more tests
* consider capabilities for StringGroupByColumnSelectorStrategy
* fix test
* fix and more test
* revert because im scared
* Fix off-by-one in IndexedTableJoinMatcher.getCardinality.
It would report a cardinality that is one lower than the actual cardinality.
The missing value is the phantom null that can be generated by outer joins.
* Fix tests.
ApproximateHistogram - seems unlikely
SegmentAnalyzer - unclear if this is an actual issue
GenericIndexedWriter - unclear if this is an actual issue
IncrementalIndexRow and OnheapIncrementalIndex are non-issues becaus it's very
unlikely for the number of dims to be large enough to hit the overflow
condition
* IntelliJ inspections cleanup
* Standard Charset object can be used
* Redundant Collection.addAll() call
* String literal concatenation missing whitespace
* Statement with empty body
* Redundant Collection operation
* StringBuilder can be replaced with String
* Type parameter hides visible type
* fix warnings in test code
* more test fixes
* remove string concatenation inspection error
* fix extra curly brace
* cleanup AzureTestUtils
* fix charsets for RangerAdminClient
* review comments
This change fixes a potential integer overflow in BufferArrayGrouper that
was flagged by LGTM. It also adds a check that the vectorized arrays are
initialized before aggregateVector is called.
The changes in HashTableUtils should not have any effect since the numbers
being multiplied are small, but the change will remove the warnings from
being flagged in LGTM.
* fix MAX_INTERMEDIATE_SIZE for DoubleMeanHolder
* byte[] type handling in deserialize and finalizeComputation for DoubleMeanAggregatorFactory
* DoubleMeanAggregatorFactory tests: Max Intermediate Size, Deserialize, finalizeComputation
* moved byte[] check to first position
Co-authored-by: Stanislav <S.Poryadnyi@abcconsulting.ru>
* fix nullhandling exceptions related to test ordering
Tests might get executed in different order depending on the maven
version and the test environment. This may lead to "NullHandling module
not initialized" errors for some tests where we do not initialize
null-handling explicitly.
* use InitializedNullHandlingTest
* SQL support for joins on subqueries.
Changes to SQL module:
- DruidJoinRule: Allow joins on subqueries (left/right are no longer
required to be scans or mappings).
- DruidJoinRel: Add cost estimation code for joins on subqueries.
- DruidSemiJoinRule, DruidSemiJoinRel: Removed, since DruidJoinRule can
handle this case now.
- DruidRel: Remove Nullable annotation from toDruidQuery, because
it is no longer needed (it was used by DruidSemiJoinRel).
- Update Rules constants to reflect new rules available in our current
version of Calcite. Some of these are useful for optimizing joins on
subqueries.
- Rework cost estimation to be in terms of cost per row, and place all
relevant constants in CostEstimates.
Other changes:
- RowBasedColumnSelectorFactory: Don't set hasMultipleValues. The lack
of isComplete is enough to let callers know that columns might have
multiple values, and explicitly setting it to true causes
ExpressionSelectors to think it definitely has multiple values, and
treat the inputs as arrays. This behavior interfered with some of the
new tests that involved queries on lookups.
- QueryContexts: Add maxSubqueryRows parameter, and use it in druid-sql
tests.
* Fixes for tests.
* Adjustments.
* Broker: Add ability to inline subqueries.
The main changes:
- ClientQuerySegmentWalker: Add ability to inline queries.
- Query: Add "getSubQueryId" and "withSubQueryId" methods.
- QueryMetrics: Add "subQueryId" dimension.
- ServerConfig: Add new "maxSubqueryRows" parameter, which is used by
ClientQuerySegmentWalker to limit how many rows can be inlined per
query.
- IndexedTableJoinMatcher: Allow creating keys on top of unknown types,
by assuming they are strings. This is useful because not all types are
known for fields in query results.
- InlineDataSource: Store RowSignature rather than component parts. Add
more zealous "equals" and "hashCode" methods to ease testing.
- Moved QuerySegmentWalker test code from CalciteTests and
SpecificSegmentsQueryWalker in druid-sql to QueryStackTests in
druid-server. Use this to spin up a new ClientQuerySegmentWalkerTest.
* Adjustments from CI.
* Fix integration test.
* Move RowSignature from druid-sql to druid-processing and make use of it.
1) Moved (most of) RowSignature from sql to processing. Left behind the SQL-specific
stuff in a RowSignatures utility class. It also picked up some new convenience
methods along the way.
2) There were a lot of places in the code where Map<String, ValueType> was used to
associate columns with type info. These are now all replaced with RowSignature.
3) QueryToolChest's resultArrayFields method is replaced with resultArraySignature,
and it now provides type info.
* Fix up extensions.
* Various fixes
* Link up row-based datasources to serving layer.
- Add SegmentWrangler interface that allows linking of DataSources to Segments.
- Add LocalQuerySegmentWalker that uses SegmentWranglers to compute queries on
data that is available locally.
- Modify ClientQuerySegmentWalker to use LocalQuerySegmentWalker when the base
datasource is concrete and not a table.
- Add SegmentWranglerModule to the Broker so it has them available and can
properly instantiate . LocalQuerySegmentWalkers.
- Set InlineDataSource and LookupDataSource to concrete, since they can be
directly queried now.
* Fix tests.
* Ability to directly query row-based datasources.
Includes:
- Foundational classes RowBasedSegment, RowBasedStorageAdapter,
RowBasedCursor provide a queryable interface on top of a
RowBasedColumnSelectorFactory.
- Add LookupSegment: A RowBasedSegment that is built on lookup data.
- Improve capability reporting in RowBasedColumnSelectorFactory.
* Fix import.
* Remove unthrown IOException.
* Harmonization and bug-fixing for selector and filter behavior on unknown types.
- Migrate ValueMatcherColumnSelectorStrategy to newer ColumnProcessorFactory
system, and set defaultType COMPLEX so unknown types can be dynamically matched.
- Remove ValueGetters in favor of ColumnComparisonFilter doing its own thing.
- Switch various methods to use convertObjectToX when casting to numbers, rather
than ad-hoc and inconsistent logic.
- Fix bug in RowBasedExpressionColumnValueSelector: isBindingArray should return
true even for 0- or 1- element arrays.
- Adjust various javadocs.
* Add throwParseExceptions option to Rows.objectToNumber, switch back to that.
* Update tests.
* Adjust moment sketch tests.
* Add OnHeapMemorySegmentWriteOutMediumFactory
Add a factory for OnHeapMemorySegmentWriteOutMedium to support direct writing via Spark.
* Register OnHeapMemorySegmentWriteOutMediumFactory.
Register OnHeapMemorySegmentWriteOutMediumFactory with SegmentWriteOutMediumFactory.
* Remove unnecessary throws
The base `makeSegmentWriteOutMedium` throws an IOException, but the particular implementation of OnHeapMemorySegmentWriteOutMediumFactory does not throw a checked exception.
* Update SegmentWriteOutMedium docs to include onHeapMemory
Update the SegmentWriteOutMedium section of the indexing docs to include a description of the new OnHeapSegmentMediumWriteOut option.
* BufferArrayGrouper: Fix potential overflow in requiredBufferCapacity.
If cardinality was high, the computation could overflow an int. There
were tests for this, but the tests were wrong.
* Nicer.
* Add SQL GROUPING SETS support.
Built on top of the subtotalsSpec feature in the groupBy query. This also involves
two changes to subtotalsSpec:
- Alter behavior so limitSpec is applied after subtotalsSpec, rather than applied to
each grouping set. This is more in line with SQL standard behavior. I think it is okay
to make this change, since the old behavior was not documented, so users should
hopefully not be depending on it.
- Fix a bug where virtual columns were included in the subtotal queries, but they
should not have been.
Also fixes two bugs in query equality checking:
- BaseQuery: Use getDuration() instead of "duration" in equals and hashCode, since the
latter is lazily initialized and might be null in one query but not the other.
- GroupByQuery: Include subtotalsSpec in equals and hashCode.
* Fix bugs.
* Fix tests.
* PR updates.
* Grouping class hygiene.
* add Expr.stringify which produces parseable expression strings, parser support for null values in arrays, and parser support for empty numeric arrays
* oops, macros are expressions too
* style
* spotbugs
* qualified type arrays
* review stuffs
* simplify grammar
* more permissive array parsing
* reuse expr joiner
* fix it
* Run IntelliJ inspections on Travis
Running IntelliJ inspections currently takes about 90 minutes, but they
can be run in about 30 minutes on Travis.
* Restore assert statements
* Fix timestamp extract fn to match postgres
Update the timestamp extract function so that it matches the PostgreSQL docs.
Examples from the PostgreSQL docs were added as tests for DECADE, CENTURY
and MILLENIUM extraction.
There were bugs in CENTURY and MILLENIUM that were spotted because of intelliJ
inspections - 'Integer division in floating point context'
* Update CalciteQueryTest
* remove useless round
* mark integer division as an error
When the time extraction Top N algorithm is looking for aggregators, it makes
2 calls to hashCode on the key. Use Map#computeIfAbsent instead so that the
hashCode is calculated only once
* Codestyle - use java style array declaration
Replaced C-style array declarations with java style declarations and marked
the intelliJ inspection as an error
* cleanup test code
* Forbid easily misused HashSet and HashMap constructors
* Add two LinkedHashMap constructors to forbidden-apis and create utility method as replacement for them
* Fix visibility of constant in CollectionUtils.java
* Make an exception for an instance of LinkedHashMap#<init>(int) because proper sizing is used
* revert changes to sql module tests that should be in separate PR
* Finish reverting changes to sql module tests that were flagged in checkstyle during CI
* Add netty dependency resulting from SupressForbidden
* Add HashVectorGrouper based on MemoryOpenHashTable.
Additional supporting changes:
1) Modifies VectorGrouper interface to use Memory instead of ByteBuffers.
2) Modifies BufferArrayGrouper to match the new VectorGrouper interface.
3) Removes "implements VectorGrouper" from BufferHashGrouper.
* Fix comment.
* Fix another comment.
* Remove unused stuff.
* Include hoisted bounds checks.
* Checks against too-large keySpaces.
* Add MemoryOpenHashTable, a table similar to ByteBufferHashTable.
With some key differences to improve speed and design simplicity:
1) Uses Memory rather than ByteBuffer for its backing storage.
2) Uses faster hashing and comparison routines (see HashTableUtils).
3) Capacity is always a power of two, allowing simpler design and more
efficient implementation of findBucket.
4) Does not implement growability; instead, leaves that to its callers.
The idea is this removes the need for subclasses, while still giving
callers flexibility in how to handle table-full scenarios.
* Fix LGTM warnings.
* Adjust dependencies.
* Remove easymock from druid-benchmarks.
* Adjustments from review.
* Fix datasketches unit tests.
* Fix checkstyle.
* Speed up joins on indexed tables with string keys
When joining on index tables with string keys, caching the computation
of row id to row numbers improves performance on the
JoinAndLookupBenchmark.joinIndexTableStringKey* benchmarks by about 10%
if the column cache is enabled an by about 100% if the column cache is
disabled.
* Faster cache impl and handle unknown cardinality
* Remove unused dependency
* Hoist cardinality check outside of hot loop
* Fix dummy DimensionSelector for tests
* Guicify druid sql module
Break up the SQLModule in to smaller modules and provide a binding that
modules can use to register schemas with druid sql.
* fix some tests
* address code review
* tests compile
* Working tests
* Add all the tests
* fix up licenses and dependencies
* add calcite dependency to druid-benchmarks
* tests pass
* rename the schemas
* SQL join support for lookups.
1) Add LookupSchema to SQL, so lookups show up in the catalog.
2) Add join-related rels and rules to SQL, allowing joins to be planned into
native Druid queries.
* Add two missing LookupSchema calls in tests.
* Fix tests.
* Fix typo.
* Add LookupJoinableFactory.
Enables joins where the right-hand side is a lookup. Includes an
integration test.
Also, includes changes to LookupExtractorFactoryContainerProvider:
1) Add "getAllLookupNames", which will be needed to eventually connect
lookups to Druid's SQL catalog.
2) Convert "get" from nullable to Optional return.
3) Swap out most usages of LookupReferencesManager in favor of the
simpler LookupExtractorFactoryContainerProvider interface.
* Fixes for tests.
* Fix another test.
* Java 11 message fix.
* Fixups.
* Fixup benchmark class.
* Add getRightColumns to JoinConditionAnalysis
This change other implementations of JoinableFactory to ask the analysis
for the right key columns instead of having to calculate it themselves.
* Address some review comments
* more code review stuff
Add microbenchmark for joins. Enabling the column cache improves
performance by ~70% for the benchmarks for joins with string keys.
Adjusting LookupJoinMatcher.matchCondition() to have fewer branches,
improves performance by ~10% for the benchmarks for joins with lookups.
* intelliJ inspections cleanup
- remove redundant escapes
- performance warnings
- access static member via instance reference
- static method declared final
- inner class may be static
Most of these changes are aesthetic, however, they will allow inspections to
be enabled as part of CI checks going forward
The valuable changes in this delta are:
- using StringBuilder instead of string addition in a loop
indexing-hadoop/.../Utils.java
processing/.../ByteBufferMinMaxOffsetHeap.java
- Use class variables instead of static variables for parameterized test
processing/src/.../ScanQueryLimitRowIteratorTest.java
* Add intelliJ inspection warnings as errors to druid profile
* one more static inner class
* Make JoinableFactory an extension point
This change makes it so that extensions can register a JoinableFactory that
should be used for a DataSource.
Extensions can provide the factories via DruidBinders#joinableFactoryBinder
Known DataSources - like InlineDataSource are provided in the
JoinableFactoryModule. This module installs a FactoryWarehouse that is
used to decide which factory should be used to generate the Joinable for
the provided DataSource.
The ExtensionPoint is marked as Beta since it is not yet clear if this
needs to remain available to other extensions or if the best way to
register a factory is by using the datasource class.
* Add module test
* remove useless bindings in test
* remove ExtensionPoint annotation
* Make LifecycleLock not final to help with testing
* Add JoinableFactory interface and use it in the query stack.
Also includes InlineJoinableFactory, which enables joining against
inline datasources. This is the first patch where a basic join query
actually works. It includes integration tests.
* Fix test issues.
* Adjustments from code review.
Builds on #9235, using the datasource analysis functionality to replace various ad-hoc
approaches. The most interesting changes are in ClientQuerySegmentWalker (brokers),
ServerManager (historicals), and SinkQuerySegmentWalker (indexing tasks).
Other changes related to improving how we analyze queries:
1) Changes TimelineServerView to return an Optional timeline, which I thought made
the analysis changes cleaner to implement.
2) Added QueryToolChest#canPerformSubquery, which is now used by query entry points to
determine whether it is safe to pass a subquery dataSource to the query toolchest.
Fixes an issue introduced in #5471 where subqueries under non-groupBy-typed queries
were silently ignored, since neither the query entry point nor the toolchest did
anything special with them.
3) Removes the QueryPlus.withQuerySegmentSpec method, which was mostly being used in
error-prone ways (ignoring any potential subqueries, and not verifying that the
underlying data source is actually a table). Replaces with a new function,
Queries.withSpecificSegments, that includes sanity checks.
* Add join-related DataSource types, and analysis functionality.
Builds on #9111 and implements the datasource analysis mentioned in #8728. Still can't
handle join datasources, but we're a step closer.
Join-related DataSource types:
1) Add "join", "lookup", and "inline" datasources.
2) Add "getChildren" and "withChildren" methods to DataSource, which will be used
in the future for query rewriting (e.g. inlining of subqueries).
DataSource analysis functionality:
1) Add DataSourceAnalysis class, which breaks down datasources into three components:
outer queries, a base datasource (left-most of the highest level left-leaning join
tree), and other joined-in leaf datasources (the right-hand branches of the
left-leaning join tree).
2) Add "isConcrete", "isGlobal", and "isCacheable" methods to DataSource in order to
support analysis.
Other notes:
1) Renamed DataSource#getNames to DataSource#getTableNames, which I think is clearer.
Also, made it a Set, so implementations don't need to worry about duplicates.
2) The addition of "isCacheable" should work around #8713, since UnionDataSource now
returns false for cacheability.
* Remove javadoc comment.
* Updates reflecting code review.
* Add comments.
* Add more comments.
* Optimize JoinCondition matching
The LookupJoinMatcher needs to check if a condition is always true or false
multiple times. This can be pre-computed to speed up the match checking
This change reduces the time it takes to perform a for joining on a long key
from ~ 36 ms/op to 23 ms/ op
* Rename variables
* fix typo
* null handling for numeric first/last aggregators, refactor to not extend nullable numeric agg since they are complex typed aggs
* initially null or not based on config
* review stuff, make string first/last consistent with null handling of numeric columns, more tests
* docs
* handle nil selectors, revert to primitive first/last types so groupby v1 works...
Gian Merlino
imply-cheddar
Laksh Singla
Clint Wylie
Abhishek Agarwal
Anshu Makkar
Tejaswini Bandlamudi
hqx871
Kashif Faraz
Paul Rogers
Suneet Saldanha
Jason Koch
Tijo Thomas
somu-imply
Rohan Garg
Adarsh Sanjeev
Karan Kumar
AmatyaAvadhanula
Frank Chen
sr
Alexander Saydakov
Lucas Capistrant
Jonathan Wei
Maytas Monsereenusorn
Jianhuan Liu
Dr. Sizzles
Agustin Gonzalez
machine424
Abhishek Radhakrishnan
Gian Merlino
Will Xu
Jihoon Son
Xavier Léauté
mchades
Laksh Singla
Liran Funaro
Jian Wang
kaijianding
dependabot[bot]
Maria Sitkovets
Harini Rajendran
BIGrey
zhangyue19921010
秦臻
Aleksey Plekhanov
Liran Funaro
Himanshu
Atul Mohan
Dylan Wylie
Joy Kent
Nishant Bangarwa
Samarth Jain
chenyuzhi459
Maytas Monsereenusorn
Suneet Saldanha
Stefan Birkner
mcbrewster
BIGrey
Jenson
Stanislav Poryadnyi
Julian Jaffe
Chi Cao Minh
Maytas Monsereenusorn