* Lazy-ify ValueMatcher BitSet optimization for string dimensions.
The idea is that if the prior evaluated filters are decently selective,
such that they mean we won't see all possible values of the later
filters, then the eager version of the optimization is too wasteful.
This involves checking an extra bitset, but the overhead is small even
if the lazy-ification is useless.
* Remove import.
* Minor transformation
* fixes#5814
changes:
* pass `StorageAdapter` to topn algorithms to get things like if column is 'sorted' or if query interval is smaller than segment granularity, instead of using `io.druid.segment.Capabilities`
* remove `io.druid.segment.Capabilities` since it had one purpose, supplying `dimensionValuesSorted` which is now provided directly by `StorageAdapter`.
* added test for topn optimization path checking
* add Capabilities back since StorageAdapter is marked PublicApi
* oops
* add javadoc, fix build i think
* correctly revert api changes
* fix intellij fail
* fix typo :(
* Revert "Consider waiting and pending compaction tasks as well as running tasks in DruidCoordinatorSegmentCompactor (#5704)"
This reverts commit c7a59394e0.
* Revert "Fix metrics for inserting segments (#5749)"
This reverts commit c9d645103b.
* Revert "Typo fix in historical doc (#5753)"
This reverts commit aa23fe6386.
* Revert "Use a bimap for reverse lookups on injective maps (#5681)"
This reverts commit e1277d306c.
* The check for maxBytesInMemory should be >= 0 instead of > 0
* if the default value is 0, the actual check could be skipped
* fix the message for persistReasons
* Address PR comments
* if maxBytes set -1, make is Long.MAX_VAL, so we do not need to check if it's 0 or -1
* set the maxBytesTuningconfig in AppenderatorImpl constructor to avoid duplicate code
* fix the failing test cases
* Address PR comments
* This commit introduces a new tuning config called 'maxBytesInMemory' for ingestion tasks
Currently a config called 'maxRowsInMemory' is present which affects how much memory gets
used for indexing.If this value is not optimal for your JVM heap size, it could lead
to OutOfMemoryError sometimes. A lower value will lead to frequent persists which might
be bad for query performance and a higher value will limit number of persists but require
more jvm heap space and could lead to OOM.
'maxBytesInMemory' is an attempt to solve this problem. It limits the total number of bytes
kept in memory before persisting.
* The default value is 1/3(Runtime.maxMemory())
* To maintain the current behaviour set 'maxBytesInMemory' to -1
* If both 'maxRowsInMemory' and 'maxBytesInMemory' are present, both of them
will be respected i.e. the first one to go above threshold will trigger persist
* Fix check style and remove a comment
* Add overlord unsecured paths to coordinator when using combined service (#5579)
* Add overlord unsecured paths to coordinator when using combined service
* PR comment
* More error reporting and stats for ingestion tasks (#5418)
* Add more indexing task status and error reporting
* PR comments, add support in AppenderatorDriverRealtimeIndexTask
* Use TaskReport instead of metrics/context
* Fix tests
* Use TaskReport uploads
* Refactor fire department metrics retrieval
* Refactor input row serde in hadoop task
* Refactor hadoop task loader names
* Truncate error message in TaskStatus, add errorMsg to task report
* PR comments
* Allow getDomain to return disjointed intervals (#5570)
* Allow getDomain to return disjointed intervals
* Indentation issues
* Adding feature thetaSketchConstant to do some set operation in PostAgg (#5551)
* Adding feature thetaSketchConstant to do some set operation in PostAggregator
* Updated review comments for PR #5551 - Adding thetaSketchConstant
* Fixed CI build issue
* Updated review comments 2 for PR #5551 - Adding thetaSketchConstant
* Fix taskDuration docs for KafkaIndexingService (#5572)
* With incremental handoff the changed line is no longer true.
* Add doc for automatic pendingSegments (#5565)
* Add missing doc for automatic pendingSegments
* address comments
* Fix indexTask to respect forceExtendableShardSpecs (#5509)
* Fix indexTask to respect forceExtendableShardSpecs
* add comments
* Deprecate spark2 profile in pom.xml (#5581)
Deprecated due to https://github.com/druid-io/druid/pull/5382
* CompressionUtils: Add support for decompressing xz, bz2, zip. (#5586)
Also switch various firehoses to the new method.
Fixes#5585.
* This commit introduces a new tuning config called 'maxBytesInMemory' for ingestion tasks
Currently a config called 'maxRowsInMemory' is present which affects how much memory gets
used for indexing.If this value is not optimal for your JVM heap size, it could lead
to OutOfMemoryError sometimes. A lower value will lead to frequent persists which might
be bad for query performance and a higher value will limit number of persists but require
more jvm heap space and could lead to OOM.
'maxBytesInMemory' is an attempt to solve this problem. It limits the total number of bytes
kept in memory before persisting.
* The default value is 1/3(Runtime.maxMemory())
* To maintain the current behaviour set 'maxBytesInMemory' to -1
* If both 'maxRowsInMemory' and 'maxBytesInMemory' are present, both of them
will be respected i.e. the first one to go above threshold will trigger persist
* Address code review comments
* Fix the coding style according to druid conventions
* Add more javadocs
* Rename some variables/methods
* Other minor issues
* Address more code review comments
* Some refactoring to put defaults in IndexTaskUtils
* Added check for maxBytesInMemory in AppenderatorImpl
* Decrement bytes in abandonSegment
* Test unit test for multiple sinks in single appenderator
* Fix some merge conflicts after rebase
* Fix some style checks
* Merge conflicts
* Fix failing tests
Add back check for 0 maxBytesInMemory in OnHeapIncrementalIndex
* Address PR comments
* Put defaults for maxRows and maxBytes in TuningConfig
* Change/add javadocs
* Refactoring and renaming some variables/methods
* Fix TeamCity inspection warnings
* Added maxBytesInMemory config to HadoopTuningConfig
* Updated the docs and examples
* Added maxBytesInMemory config in docs
* Removed references to maxRowsInMemory under tuningConfig in examples
* Set maxBytesInMemory to 0 until used
Set the maxBytesInMemory to 0 if user does not set it as part of tuningConfing
and set to part of max jvm memory when ingestion task starts
* Update toString in KafkaSupervisorTuningConfig
* Use correct maxBytesInMemory value in AppenderatorImpl
* Update DEFAULT_MAX_BYTES_IN_MEMORY to 1/6 max jvm memory
Experimenting with various defaults, 1/3 jvm memory causes OOM
* Update docs to correct maxBytesInMemory default value
* Minor to rename and add comment
* Add more details in docs
* Address new PR comments
* Address PR comments
* Fix spelling typo
* Use mergeBuffer instead of processingBuffer in parallelCombiner
* Fix test
* address comments
* fix test
* Fix test
* Update comment
* address comments
* fix build
* Fix test failure
* Adding feature thetaSketchConstant to do some set operation in PostAggregator
* Updated review comments for PR #5551 - Adding thetaSketchConstant
* Fixed CI build issue
* Updated review comments 2 for PR #5551 - Adding thetaSketchConstant
* Add more indexing task status and error reporting
* PR comments, add support in AppenderatorDriverRealtimeIndexTask
* Use TaskReport instead of metrics/context
* Fix tests
* Use TaskReport uploads
* Refactor fire department metrics retrieval
* Refactor input row serde in hadoop task
* Refactor hadoop task loader names
* Truncate error message in TaskStatus, add errorMsg to task report
* PR comments
* Use the official aws-sdk instead of jet3t
* fix compile and serde tests
* address comments and fix test
* add http version string
* remove redundant dependencies, fix potential NPE, and fix test
* resolve TODOs
* fix build
* downgrade jackson version to 2.6.7
* fix test
* resolve the last TODO
* support proxy and endpoint configurations
* fix build
* remove debugging log
* downgrade hadoop version to 2.8.3
* fix tests
* remove unused log
* fix it test
* revert KerberosAuthenticator change
* change hadoop-aws scope to provided in hdfs-storage
* address comments
* address comments
* Future-proof some Guava usage
* Use a java-util EmptyIterator instead of Guava's
* Change some of the guava future handling to do manual async
transforms. Guava changes transform into transformAsync by deprecating
transform in ONLY Guava 19. Then its gone in 20
* Use `Collections.emptyIterator()`
* Pretty formatting
* Make listenable future transforms a thing in default druid
* Format fix
* Add forbidden guava apis
* Make the ListenableFutrues.transformAsync have comments
* Undo intellij bad pattern matching in comments
* Futrues --> Futures
* Add empty iterators forbidding
* Fix extra `A`
* Correct method signature
* Address review comments
* Finish Gian review comments
* Proper syntax from https://github.com/policeman-tools/forbidden-apis/wiki/SignaturesSyntax
* Fix round robining in router.
Say that ten times fast.
For query endpoints, AsyncQueryForwardingServlet called hostFinder.getDefaultServer()
to set a default server, followed by hostFinder.getServer(inputQuery) to override it
with query-specific routing. Since hostFinder is round-robin, this skips a server.
When there are only two servers, one server is _always_ skipped and the router sends
all queries to the same broker.
* Adjust spacing.
* SegmentMetadataQuery: Fix default interval handling.
PR #4131 introduced a new copy builder for segmentMetadata that did
not retain the value of usingDefaultInterval. This led to it being
dropped and the default-interval handling not working as expected.
Instead of using the default 1 week history when intervals are not
provided, the segmentMetadata query would query _all_ segments,
incurring an unexpected performance hit.
This patch fixes the bug and adds a test for the copy builder.
* Intervals
* Support for disabling bitmap indexes.
Can save space for columns where bitmap indexes are pointless (like
free-form text).
* Remove import.
* Fix CompactionTaskTest.
* Update for review comments.
* Review comments, tests.
* Fix test.
* Fix two improper casts in HavingSpecMetricComparator.
Fixes two things:
1. An improper double-to-long cast when comparing double metrics to any
kind of value, which was a regression from #4883.
2. An improper double-to-long cast when comparing a long/int metric to a
double/float value: the value was cast to long/int, drawing strange
conclusions like int 100 matching a havingSpec of equalTo(100.5).
* Add comments.
* Remove extraneous comment.
* Simplify code a bit.
* Properly set "identity" in query metrics.
This patch adds an "identity" field to QueryPlus and sets it in
QueryLifecycle when the query starts executing. This is important
because it allows it to be used for future QueryMetrics created
by that QueryPlus object.
We also add "identity" to the request-level QueryMetrics object
created in emitLogsAndMetrics.
* Remove unused method.
* Fix races in LookupSnapshotTaker, CoordinatorPollingBasicAuthenticatorCacheManager.
Both were susceptible to the following conditions:
1. Two JVMs on the same machine (perhaps two peons) could conflict by one reading while the
other was writing, or by writing to the file at the same time.
2. One JVM could partially write a file, then crash, leaving a truncated file.
* Use StringUtils.format
* Use both Joad Ids and Java IDs as Timezone to string readers
Change-Id: Ieb5c18559879f3f3a0104912ce2f0a354ad0aac3
* move the function to DateTimes and add org.joda.time.DateTimeZone#forID as part of forbidden api
Change-Id: Iff97fa044758019ed0c231587d10e31a9cc18da0
* exclude class and remove other usage
Change-Id: Ib458c2caaa1865535767e1009fbf017a92c8f615
* remove it from test classes
Change-Id: I9b576324f6c7e17a74bd8b13879232c9a8cd40b4
* remove unused
Change-Id: If1c5b70c26c2b7c83c20434cb72b2060653f5052
The behavior is configurable through druid.extensions.useExtensionClassloaderFirst.
It is useful when extensions want to load a dependency different from one provided
by Druid, for example a different version of geoip or protobuf.
Code changes:
- In the lookup-based extractionFns, inherit injective property from
the lookup itself if not specified.
Doc changes:
- Add a "Query execution" section to the lookups doc explaining how
injective lookups and their optimizations work.
- Remove scary warnings against using registeredLookup extractionFns.
They are necessary and important since they work with filters and
function cascades -- two things that the dimension specs do not do.
They deserve to be first class citizens.
- Move the "registeredLookup" fn above the "lookup" fn. It's probably
more commonly used, so the docs read better this way.
* timewarp and timezones
changes:
* `TimewarpOperator` will now compensate for daylight savings time shifts between date translation ranges for queries using a `PeriodGranularity` with a timezone defined
* introduces a new abstract query type `TimeBucketedQuery` for all queries which have a `Granularity` (100% not attached to this name). `GroupByQuery`, `SearchQuery`, `SelectQuery`, `TimeseriesQuery`, and `TopNQuery` all extend `TimeBucke
tedQuery`, cutting down on some duplicate code and providing a mechanism for `TimewarpOperator` (and anything else) that needs to be aware of granularity
* move precondition check to TimeBucketedQuery, add Granularities.nullToAll, add getTimezone to TimeBucketQuery
* formatting
* more formatting
* unused import
* changes:
* add 'getGranularity' and 'getTimezone' to 'Query' interface
* merge 'TimeBucketedQuery' into 'BaseQuery'
* fixup tests from resulting serialization changes
* dedupe
* fix after merge
* suppress warning
* Apply 'power of 2' optimization to BlockLayoutIndexedDoubleSupplier; slight optimization of buffer.get() in block layout indexed suppliers
* Fix byte order
* Deduplicate DataSegments contents (loadSpec's keys, dimensions and metrics lists as a whole) more aggressively; use ArrayMap instead of default LinkedHashMap for DataSegment.loadSpec, because they have only 3 entries on average; prune DataSegment.loadSpec on brokers
* Fix DataSegmentTest
* Refinements
* Try to fix
* Fix the second DataSegmentTest
* Nullability
* Fix tests
* Fix tests, unify to use TestHelper.getJsonMapper()
* Revert TestUtil as ServerTestHelper, fix tests
* Add newline
* Fix indexing tests
* Fix s3 tests
* Try to fix tests, remove lazy caching of ObjectMapper in TestHelper, rename TestHelper.getJsonMapper() to makeJsonMapper()
* Fix HDFS tests
* Fix HdfsDataSegmentPusherTest
* Capitalize constant names
* numeric quantiles sketch aggregator
* it seems that we need to synchronize all methods, which modify the state
* Seems like a false positive with -Pstrict
* code style fix
* code style fix
* use sketches-core-0.10.3
* moved cache ids to the central place
* better class names
* support large columns
* explained autodetection, added exception
* added comments regarding sketches moving on heap
* support reindexing
* implemented suggestions from jihoonson
* style fix
* use max(k, other.k) for better accuracy
* check for NilColumnValueSelector instead of null
* throw exceptions instead of providing no-op comparators
* SQL: Improve translation of time floor expressions.
The main change is to TimeFloorOperatorConversion.applyTimestampFloor.
- Prefer timestamp_floor expressions to timeFormat extractionFns, to
avoid turning things into strings when it isn't necessary.
- Collapse CAST(FLOOR(X TO Y) AS DATE) to FLOOR(X TO Y) if appropriate.
* Fix tests.
* ExpressionSelectors: Add caching selectors.
- SingleLongInputCaching selector for expressions on the __time column,
using a similar optimization to SingleScanTimeDimSelector
- SingleStringInputDimensionSelector for expressions on string columns
that return strings, using a similar optimization to ExtractionFn
based DimensionSelectors.
- SingleStringInputCaching selector for expressions on string columns
that return primitives.
Also, in the SQL planner, prefer expressions for time operations
rather than extractionFns.
* Code review comments.
* maxQueryTimeout property in runtime properties.
* extra line
* move withTimeoutAndMaxScatterGatherBytes method to QueryLifeCycle.
* Fix initialize method.
* remove unused import.
* doc update.
* some more details in doc about query failure..
* minor fix.
* decorating QueryRunner to set and verify context. Added by servers.
* remove whitespace.
* SQL: Improved behavior when implicitly casting strings to date/time literals.
- Handle all flavors of ISO8601 and SQL literals.
- Throw errors on other literals instead of silently transforming them to 0.
* Respect timeZone when format is null.
* Add retries for coordinator fetch and lookup start in LookupReferencesManager
* Fix LookupConfigTest
* Address comments
* Address more comments
* And address more comments
* Address comms
* Recognize 'not found' lookups in LookupReferencesManager.tryGetLookupListFromCoordinator(), by @egor-ryashin
* Add compaction task
* added doc
* use combining aggregators
* address comments
* add support for dimensionsSpec
* fix getUniqueDims and getUniqueMetics
* find unique dimensionsSpec
* fix compilation
* add unit test
* fix test
* fix test
* test for different dimension orderings and types, and doc for type and ordering
* add control for custom ordering and type
* update doc
* fix compile
* fix compile
* add segments param
* fix serde error
* fix build
* Fix havingSpec on complex aggregators.
- Uses the technique from #4883 on DimFilterHavingSpec too.
- Also uses Transformers from #4890, necessitating a move of that and other
related classes from druid-server to druid-processing. They probably make
more sense there anyway.
- Adds a SQL query test.
Fixes#4957.
* Remove unused import.
* Fix improper handling of empty arrays in StringDimensionIndexer.
This bug was able to introduce data errors: if the input rows to an
IncrementalIndex contained entirely empty arrays and single values, then
upon persisting to disk, the empty arrays would be replaced with the
lexicographically smallest single value, rather than nulls like they
should have been.
* Style fix.
* Add tests for bitmap indexes too.
* Introduce System wide property to select how to store double.
Set the default to store as float
Change-Id: Id85cca04ed0e7ecbce78624168c586dcc2adafaa
* fix tests
Change-Id: Ib42db724b8a8f032d204b58c366caaeabdd0d939
* Change the property name
Change-Id: I3ed69f79fc56e3735bc8f3a097f52a9f932b4734
* add tests and make default distribution store doubles as 64bits
Change-Id: I237b07829117ac61e247a6124423b03992f550f2
* adding mvn argument to parallel-test profile
Change-Id: Iae5d1328f901c4876b133894fa37e0d9a4162b05
* move property name and helper function to io.druid.segment.column.Column
Change-Id: I62ea903d332515de2b7ca45c02587a1b015cb065
* fix docs and clean style
Change-Id: I726abb8f52d25dc9dc62ad98814c5feda5e4d065
* fix docs
Change-Id: If10f4cf1e51a58285a301af4107ea17fe5e09b6d
* greater-than/less-than/equal-to havingSpec to call AggregatorFactory.finalizeComputation(..)
* fix the unit test and expect having to work on hyperUnique agg
* test fix
* fix style errors
* Changes for lookup synchronization
* Refactor of Lookup classes
* Minor refactors and doc update
* Change coordinator instance to be retrieved by DruidLeaderClient
* Wait before thread shutdown
* Make disablelookups flag true by default
* Update docs
* Rename flag
* Move executorservice shutdown to finally block
* Update LookupConfig
* Refactoring and doc changes
* Remove lookup config constructor
* Revert Lookupconfig constructor changes
* Add tests to LookupConfig
* Make executorservice local
* Update LRM
* Move ListeningScheduledExecutorService to ExecutorCompletionService
* Move exception to outer block
* Remove check to see future is done
* Remove unnecessary assignment
* Add logging
* QueryableIndexStorageAdapter: Lift column cache to Cursor sequence.
This is where it was before #4710, when its was moved to the individual
Cursors, leading to higher than expected memory usage. It could be
extreme for finer query granularities like "second".
* Comment.
* SQL: Upgrade to Calcite 1.14.0, some refactoring of internals.
This brings benefits:
- Ability to do GROUP BY and ORDER BY with ordinals.
- Ability to support IN filters beyond 19 elements (fixes#4203).
Some refactoring of druid-sql internals:
- Builtin aggregators and operators are implemented as SqlAggregators
and SqlOperatorConversions rather being special cases. This simplifies
the Expressions and GroupByRules code, which were becoming complex.
- SqlAggregator implementations are no longer responsible for filtering.
Added new functions:
- Expressions: strpos.
- SQL: TRUNCATE, TRUNC, LENGTH, CHAR_LENGTH, STRLEN, STRPOS, SUBSTR,
and DATE_TRUNC.
* Add missing @Override annotation.
* Adjustments for forbidden APIs.
* Adjustments for forbidden APIs.
* Disable GROUP BY alias.
* Doc reword.
* adding new post aggregators of test stats to druid-stats extension
* changes to address code review comments
* fix checkstyle violations using druid_intellij_formatting.xml after merge upstream/master
* add @Override annotation per CI log
* make changes per review comments/discussions
* remove some blocks per review comments
* Add identity to query metrics, logs.
Also fix a bug where unauthorized requests would not emit any logs or metrics,
and instead would log a "Tried to emit logs and metrics twice" warning.
Also rename QueryResource's "getServer" to "cancelQuery", because that's what
it does.
* Do not emit identity by default.
* Added org.joda.time.DateTime#(java.lang.String) to forbidden API.
* Added org.joda.time.DateTime#(java.lang.String, org.joda.time.format.DateTimeFormatter) to forbidden API.
* Add additional APIs that may create DateTime with default time zone
* Add helper function that accepts formatter to parse String.
* Add additional forbidden APIs
* Replace existing usage of forbidden APIs
* Use wrapper class to enforce Chronology on DateTimeFormatter.
* Creates constant UtcFormatter for constant ISODateTimeFormat.
* Move scan-query from a contrib extension into core.
Based on a proposal at: https://groups.google.com/d/topic/druid-development/ME_OatUDnbk/discussion
This patch also adds support for virtual columns to the Scan query,
and updates Druid SQL to use Scan instead of Select.
This patch also makes some behavioral changes to handling of the __time
column. In particular, it is now is returned as "__time" rather than
"timestamp"; it is no longer included if you do not specifically ask for
it in your "columns"; and it is returned as a long rather than a string.
Users can revert time handling to the legacy extension behavior by
setting "legacy" : true in their queries, or setting the property
druid.query.scan.legacy = true. This is meant to provide a migration
path for users that were formerly using the contrib extension.
* Adjustments from review.
* Add back Select query.
* Adjust SQL docs.
* Restore SelectQuery link.
* SQL: Full TRIM support.
- Support trimming arbitrary characters
- Support BOTH, LEADING, and TRAILING
* Remove unused import.
* Fix tests, add RTRIM / LTRIM.
* Remove unused imports.
* BTRIM and docs.
* Replace for with foreach.
* BufferHashGrouperTest: Better behavior with regard to large buffers.
1) Free buffers after each test
2) Avoid mmaping past the end of a file
* Use CloserRule.
When historical caching is enabled, and a select or topN query is
issued, and then a following query with "descending": true is set, the
cached query returns the ascending result (or vice versa), often
resulting in invalid paging identifiers.
The CacheKey for these queries doesn't include the "descending" flag;
this change adds it, and fixes the problem.
* Factor QueryableIndexColumnSelectorFactory and IncrementalIndexColumnSelectorFactory out of QueryableIndexStorageAdapter and IncrementalIndexStorageAdapter; Add Offset.getBaseReadableOffset(); Remove OffsetHolder interface; Replace Cursor extends ColumnSelectorFactory with composition; Reduce indirection in ColumnValueSelectors created by QueryableIndexColumnSelectorFactory
* Don't override clone() in FilteredOffset (the prev. implementation was broken); Some warnings fixed
* Simplify Cursors in QueryableIndexStorageAdapter
* Address comments
* Remove unused and unimplemented methods from GenericColumn interface
* Comments
* Add "round" option to cardinality and hyperUnique aggregators.
Also turn it on by default in SQL, to make math on distinct counts
work more as expected.
* Fix some compile errors.
* Fix test.
* Formatting.
* Add @ExtensionPoint and @PublicApi annotations.
* Clean up wording.
* Remove unused import.
* Remove unused imports.
* Only types can be extension points.
* Adjust annotations some more.
* Remove unused import.
* Make ServletFilterHolder an extension point.
* Add a couple extension points, and update docs.
* Fix dimension selectors with extractionFns on missing columns.
This patch properly applies the requested extractionFn to missing columns.
It's important when the extractionFn maps null to something other than null.
* Extract helper method.
* Change contracts of VirtualColumns and VirtualColumn methods based on review comments.
* Remove unused import.
* Remove unused method.
* Adjust helper function.
* Adjustments
* Replace Guava Enum.getIfPresent with builtin version.
This is useful for running in Hadoop environments that use Guava 11. Some
code is also simplified.
* Code review
* Don't use limit push down with having spec
* Throw exception when forcing limit push down with having
* Tests for having and limit push down
* Fix pool sizes in unit test
* Add HistoricalCursor.getReadableOffset() to access unwrapped offset in selectors, when the 'main' offset if FilteredOffset (fixes#4628)
* Stack overflow test
* NPE thrown when empty/null is passes to TimeDimExtractionFn
* Add @Nullable where ever applicable
* Add @Nullable to SearchQuerySpec.apply()
* Remove unused
Gian Merlino
Clint Wylie
Jihoon Son
Alexander Saydakov
Dylan Wylie
Surekha
kaijianding
Roman Leventov
Slim Bouguerra
scrawfor
Senthil Kumar L S
Niketh Sabbineni
Jonathan Wei
Kirill Kozlov
Atul Mohan
Charles Allen
Samarth Jain
Niraja Mishra
Nishant Bangarwa
Himanshu
Parag Jain
Chuanlei Ni
zhangxinyu1
Akash Dwivedi
chunghochen
Goh Wei Xiang
Andy Sloane
Asif Mansoor Amanullah
solimant