Skip to content

Iceberg integration plugin impl #22407

New issue

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

Sign up for GitHub

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

Already on GitHub? Sign in to your account

Draft
sachin-27 wants to merge 13 commits into
base: main
Choose a base branch
from
Draft

Iceberg integration plugin impl #22407

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
13 commits
Select commit Hold shift + click to select a range
054fbb3
creating new plugin and adding s3 related files and dependencies for …
sachin-27 Apr 20, 2026
dce7eac
moving all s3 repository files to jar
sachin-27 Apr 20, 2026
f33e2cf
enabling third party audit and renaming plugin
sachin-27 Apr 21, 2026
920ade4
updating build.gradle and removing s3 repository library jar
sachin-27 Apr 22, 2026
503af4d
removing plugin security policy
sachin-27 Apr 22, 2026
0f3adc1
adding core interfaces
sachin-27 Apr 23, 2026
2035f7a
adding core implementation
sachin-27 Apr 24, 2026
37e06e2
updating dependency list
sachin-27 Apr 24, 2026
9a276a1
adding plugin skeleton
sachin-27 Apr 24, 2026
2780b40
adding license files
sachin-27 Apr 27, 2026
8f3bf84
iceberg client implementation
sachin-27 Apr 29, 2026
4d0650e
iceber client publish shard action implementation
sachin-27 Apr 29, 2026
50731d4
Fix rebase drift and boot safety for catalog plugin
sachin-27 Jul 8, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
190 changes: 190 additions & 0 deletions plugins/iceberg-metadata-catalog/build.gradle
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,190 @@
/*
* SPDX-License-Identifier: Apache-2.0
*
* The OpenSearch Contributors require contributions made to
* this file be licensed under the Apache-2.0 license or a
* compatible open source license.
*/

apply plugin: 'opensearch.opensearchplugin'

opensearchplugin {
description = 'Apache Iceberg metadata catalog plugin: publishes OpenSearch index data to S3 Tables'
classname = 'org.opensearch.plugin.catalog.iceberg.IcebergCatalogPlugin'
}

dependencies {

// Iceberg
api "org.apache.iceberg:iceberg-api:1.6.1"
api "org.apache.iceberg:iceberg-core:1.6.1"
api("org.apache.iceberg:iceberg-aws:1.6.1") {
exclude group: 'org.apache.iceberg', module: 'iceberg-bundled-guava'
}

// Avro is required by iceberg-core's PartitionData (implements org.apache.avro.generic.IndexedRecord).
// Any code path that constructs a partitioned DataFile triggers Avro class loading. We pull it in
// explicitly so partitioned table writes work; we do not use Avro as a serialization format.
// commons-compress is declared as a compile-scope transitive by avro itself.
runtimeOnly "org.apache.avro:avro:1.11.1"
runtimeOnly "org.apache.commons:commons-compress:${versions.commonscompress}"

// AWS SDK v2 — core + S3 + Apache HTTP client
api "software.amazon.awssdk:s3:${versions.aws}"
api "software.amazon.awssdk:apache-client:${versions.aws}"
api "software.amazon.awssdk:sdk-core:${versions.aws}"
api "software.amazon.awssdk:aws-core:${versions.aws}"
api "software.amazon.awssdk:auth:${versions.aws}"
api "software.amazon.awssdk:identity-spi:${versions.aws}"
api "software.amazon.awssdk:http-client-spi:${versions.aws}"
api "software.amazon.awssdk:regions:${versions.aws}"
api "software.amazon.awssdk:utils:${versions.aws}"
api "software.amazon.awssdk:annotations:${versions.aws}"
api "software.amazon.awssdk:endpoints-spi:${versions.aws}"
api "software.amazon.awssdk:metrics-spi:${versions.aws}"

// STS is used from main code by CredentialsBuilder (IRSA + assume-role paths).
api "software.amazon.awssdk:sts:${versions.aws}"

// ============================================================================
// Runtime-only
// ============================================================================

// AWS SDK v2 runtime modules — signing, retries, protocols, checksums, auth wiring.
runtimeOnly "software.amazon.awssdk:http-auth:${versions.aws}"
runtimeOnly "software.amazon.awssdk:http-auth-aws:${versions.aws}"
runtimeOnly "software.amazon.awssdk:http-auth-spi:${versions.aws}"
runtimeOnly "software.amazon.awssdk:retries:${versions.aws}"
runtimeOnly "software.amazon.awssdk:retries-spi:${versions.aws}"
runtimeOnly "software.amazon.awssdk:checksums:${versions.aws}"
runtimeOnly "software.amazon.awssdk:checksums-spi:${versions.aws}"
runtimeOnly "software.amazon.awssdk:profiles:${versions.aws}"
runtimeOnly "software.amazon.awssdk:json-utils:${versions.aws}"
runtimeOnly "software.amazon.awssdk:protocol-core:${versions.aws}"
runtimeOnly "software.amazon.awssdk:aws-json-protocol:${versions.aws}"
runtimeOnly "software.amazon.awssdk:aws-xml-protocol:${versions.aws}"
runtimeOnly "software.amazon.awssdk:aws-query-protocol:${versions.aws}"
runtimeOnly "software.amazon.awssdk:signer:${versions.aws}"
runtimeOnly "software.amazon.awssdk:third-party-jackson-core:${versions.aws}"

// Apache HTTP transitives used by apache-client at runtime.
runtimeOnly "org.apache.httpcomponents:httpclient:${versions.httpclient}"
runtimeOnly "org.apache.httpcomponents:httpcore:${versions.httpcore}"
runtimeOnly "commons-codec:commons-codec:${versions.commonscodec}"
runtimeOnly "commons-logging:commons-logging:${versions.commonslogging}"

// Jackson (iceberg-core and SDK both use it at runtime).
runtimeOnly "com.fasterxml.jackson.core:jackson-core:${versions.jackson}"
runtimeOnly "com.fasterxml.jackson.core:jackson-databind:${versions.jackson_databind}"
runtimeOnly "com.fasterxml.jackson.core:jackson-annotations:${versions.jackson_annotations}"

// slf4j (iceberg uses slf4j for logging at runtime).
runtimeOnly "org.slf4j:slf4j-api:${versions.slf4j}"

// Iceberg runtime modules.
runtimeOnly "org.apache.iceberg:iceberg-common:1.6.1"
runtimeOnly("org.apache.iceberg:iceberg-bundled-guava:1.6.1") {
exclude group: 'org.apache.hadoop'
exclude group: 'org.slf4j'
}
runtimeOnly "com.github.ben-manes.caffeine:caffeine:2.9.3"
}

tasks.named("dependencyLicenses").configure {
// Iceberg artifacts share a common LICENSE/NOTICE.
mapping from: /iceberg-.*/, to: 'iceberg'

// Jackson artifacts share a common LICENSE/NOTICE.
mapping from: /jackson-.*/, to: 'jackson'
}
tasks.named("thirdPartyAudit").configure {
ignoreMissingClasses()
ignoreViolations(
// Caffeine uses sun.misc.Unsafe for atomic field updates and array access.
'com.github.benmanes.caffeine.base.UnsafeAccess',
'com.github.benmanes.caffeine.cache.BaseMpscLinkedArrayQueue',
'com.github.benmanes.caffeine.cache.BBHeader$ReadAndWriteCounterRef',
'com.github.benmanes.caffeine.cache.BBHeader$ReadCounterRef',
'com.github.benmanes.caffeine.cache.BLCHeader$DrainStatusRef',
'com.github.benmanes.caffeine.cache.FD',
'com.github.benmanes.caffeine.cache.FDA',
'com.github.benmanes.caffeine.cache.FDAR',
'com.github.benmanes.caffeine.cache.FDAW',
'com.github.benmanes.caffeine.cache.FDAWR',
'com.github.benmanes.caffeine.cache.FDR',
'com.github.benmanes.caffeine.cache.FDW',
'com.github.benmanes.caffeine.cache.FDWR',
'com.github.benmanes.caffeine.cache.FS',
'com.github.benmanes.caffeine.cache.FSA',
'com.github.benmanes.caffeine.cache.FSAR',
'com.github.benmanes.caffeine.cache.FSAW',
'com.github.benmanes.caffeine.cache.FSAWR',
'com.github.benmanes.caffeine.cache.FSR',
'com.github.benmanes.caffeine.cache.FSW',
'com.github.benmanes.caffeine.cache.FSWR',
'com.github.benmanes.caffeine.cache.FW',
'com.github.benmanes.caffeine.cache.FWA',
'com.github.benmanes.caffeine.cache.FWAR',
'com.github.benmanes.caffeine.cache.FWAW',
'com.github.benmanes.caffeine.cache.FWAWR',
'com.github.benmanes.caffeine.cache.FWR',
'com.github.benmanes.caffeine.cache.FWW',
'com.github.benmanes.caffeine.cache.FWWR',
'com.github.benmanes.caffeine.cache.PD',
'com.github.benmanes.caffeine.cache.PDA',
'com.github.benmanes.caffeine.cache.PDAR',
'com.github.benmanes.caffeine.cache.PDAW',
'com.github.benmanes.caffeine.cache.PDAWR',
'com.github.benmanes.caffeine.cache.PDR',
'com.github.benmanes.caffeine.cache.PDW',
'com.github.benmanes.caffeine.cache.PDWR',
'com.github.benmanes.caffeine.cache.PS',
'com.github.benmanes.caffeine.cache.PSA',
'com.github.benmanes.caffeine.cache.PSAR',
'com.github.benmanes.caffeine.cache.PSAW',
'com.github.benmanes.caffeine.cache.PSAWR',
'com.github.benmanes.caffeine.cache.PSR',
'com.github.benmanes.caffeine.cache.PSW',
'com.github.benmanes.caffeine.cache.PSWR',
'com.github.benmanes.caffeine.cache.PW',
'com.github.benmanes.caffeine.cache.PWA',
'com.github.benmanes.caffeine.cache.PWAR',
'com.github.benmanes.caffeine.cache.PWAW',
'com.github.benmanes.caffeine.cache.PWAWR',
'com.github.benmanes.caffeine.cache.PWR',
'com.github.benmanes.caffeine.cache.PWW',
'com.github.benmanes.caffeine.cache.PWWR',
'com.github.benmanes.caffeine.cache.StripedBuffer',
'com.github.benmanes.caffeine.cache.UnsafeAccess',
'com.github.benmanes.caffeine.cache.UnsafeRefArrayAccess',
'com.github.benmanes.caffeine.SCQHeader$HeadAndTailRef',
'com.github.benmanes.caffeine.SingleConsumerQueue',
'com.github.benmanes.caffeine.SingleConsumerQueue$Node',

// Iceberg's relocated Guava uses sun.misc.Unsafe for atomics and byte array access.
'org.apache.iceberg.relocated.com.google.common.hash.LittleEndianByteArray$UnsafeByteArray',
'org.apache.iceberg.relocated.com.google.common.hash.LittleEndianByteArray$UnsafeByteArray$1',
'org.apache.iceberg.relocated.com.google.common.hash.LittleEndianByteArray$UnsafeByteArray$2',
'org.apache.iceberg.relocated.com.google.common.primitives.UnsignedBytes$LexicographicalComparatorHolder$UnsafeComparator',
'org.apache.iceberg.relocated.com.google.common.primitives.UnsignedBytes$LexicographicalComparatorHolder$UnsafeComparator$1',
'org.apache.iceberg.relocated.com.google.common.util.concurrent.AbstractFuture$UnsafeAtomicHelper',
'org.apache.iceberg.relocated.com.google.common.util.concurrent.AbstractFuture$UnsafeAtomicHelper$1',

// Avro uses sun.misc.Unsafe for reflective field access in its specific-record path.
// We don't use Avro as a serialization format; we need it on the classpath only because
// iceberg-core's PartitionData implements org.apache.avro.generic.IndexedRecord. None of
// the FieldAccessUnsafe paths are exercised by our code.
'org.apache.avro.reflect.FieldAccessUnsafe',
'org.apache.avro.reflect.FieldAccessUnsafe$UnsafeBooleanField',
'org.apache.avro.reflect.FieldAccessUnsafe$UnsafeByteField',
'org.apache.avro.reflect.FieldAccessUnsafe$UnsafeCachedField',
'org.apache.avro.reflect.FieldAccessUnsafe$UnsafeCharField',
'org.apache.avro.reflect.FieldAccessUnsafe$UnsafeCustomEncodedField',
'org.apache.avro.reflect.FieldAccessUnsafe$UnsafeDoubleField',
'org.apache.avro.reflect.FieldAccessUnsafe$UnsafeFloatField',
'org.apache.avro.reflect.FieldAccessUnsafe$UnsafeIntField',
'org.apache.avro.reflect.FieldAccessUnsafe$UnsafeLongField',
'org.apache.avro.reflect.FieldAccessUnsafe$UnsafeObjectField',
'org.apache.avro.reflect.FieldAccessUnsafe$UnsafeShortField'
)
}
1 change: 1 addition & 0 deletions plugins/iceberg-metadata-catalog/licenses/annotations-2.32.29.jar.sha1
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
d70dcb2d74df899972ac888f1b306ddd7e83bee3
Loading
Loading