[FEA] Consume renamed cudf-spark-jni artifacts [databricks]

.github/workflows/mvn-verify-check/get-deps-sha1.sh

@@ -18,10 +18,10 @@ set -e
 
 scala_ver=${1:-"2.12"}
 base_URL="https://central.sonatype.com/repository/maven-snapshots/com/nvidia"
-project_jni="spark-rapids-jni"
+project_jni=$(mvn help:evaluate -q -pl dist -Dexpression=cudf-spark-jni.artifactId -DforceStdout)
 project_private="cudf-spark-private_${scala_ver}"
 
-jni_ver=$(mvn help:evaluate -q -pl dist -Dexpression=spark-rapids-jni.version -DforceStdout)
+jni_ver=$(mvn help:evaluate -q -pl dist -Dexpression=cudf-spark-jni.version -DforceStdout)
 private_ver=$(mvn help:evaluate -q -pl dist -Dexpression=cudf-spark-private.version -DforceStdout)
 
 get_latest_snapshot_version() {

.greptile/config.json

@@ -6,11 +6,11 @@
 
   "patternRepositories": [
     "rapidsai/cudf",
-    "NVIDIA/spark-rapids-jni",
+    "NVIDIA/cudf-spark-jni",
     "apache/spark"
   ],
 
-  "instructions": "Spark plugin for GPU-accelerated data processing using RAPIDS cuDF. Maven-based, Scala/Java with Python integration tests. Key modules: sql-plugin/, shuffle-plugin/, iceberg/, delta-lake/, tests/, integration_tests/. Version-specific shims at sql-plugin/src/main/spark{VERSION}/. Review rules are in .greptile/rules.md. Key files are listed in .greptile/files.json. Also reference AGENTS.md for build commands, project structure, and coding conventions. IMPORTANT: This project depends on cuDF (rapidsai/cudf) for GPU compute kernels and spark-rapids-jni (NVIDIA/spark-rapids-jni) for JNI bindings. When reviewing code that calls cuDF/JNI APIs, cross-reference the upstream repos to verify API usage correctness, especially for: column operations, memory allocation semantics, stream handling, and error codes. This project also shims against multiple Apache Spark versions (apache/spark). When reviewing shim code or GPU operator implementations, cross-reference apache/spark to verify that the GPU behavior matches the corresponding CPU implementation in the target Spark version, especially for: expression evaluation semantics, null handling, type coercion, and catalog/partition behavior.",
+  "instructions": "Spark plugin for GPU-accelerated data processing using RAPIDS cuDF. Maven-based, Scala/Java with Python integration tests. Key modules: sql-plugin/, shuffle-plugin/, iceberg/, delta-lake/, tests/, integration_tests/. Version-specific shims at sql-plugin/src/main/spark{VERSION}/. Review rules are in .greptile/rules.md. Key files are listed in .greptile/files.json. Also reference AGENTS.md for build commands, project structure, and coding conventions. IMPORTANT: This project depends on cuDF (rapidsai/cudf) for GPU compute kernels and cudf-spark-jni (NVIDIA/cudf-spark-jni) for JNI bindings. When reviewing code that calls cuDF/JNI APIs, cross-reference the upstream repos to verify API usage correctness, especially for: column operations, memory allocation semantics, stream handling, and error codes. This project also shims against multiple Apache Spark versions (apache/spark). When reviewing shim code or GPU operator implementations, cross-reference apache/spark to verify that the GPU behavior matches the corresponding CPU implementation in the target Spark version, especially for: expression evaluation semantics, null handling, type coercion, and catalog/partition behavior.",
 
   "rules": [
     {
@@ -117,7 +117,7 @@
     },
     {
       "id": "upstream-dependency-boundary",
-      "rule": "Code touching JNI calls (cudf/spark-rapids-jni) or private API interfaces (spark-rapids-private) may break from concurrent upstream SNAPSHOT changes. Review for assumptions about upstream behavior. Cross-reference rapidsai/cudf for correct function signatures, null handling semantics, and memory ownership rules at JNI boundaries.",
+      "rule": "Code touching JNI calls (cudf/cudf-spark-jni) or private API interfaces (cudf-spark-private) may break from concurrent upstream SNAPSHOT changes. Review for assumptions about upstream behavior. Cross-reference rapidsai/cudf for correct function signatures, null handling semantics, and memory ownership rules at JNI boundaries.",
       "scope": ["sql-plugin/src/**", "shuffle-plugin/src/**"],
       "severity": "high"
     },

.greptile/rules.md

@@ -17,6 +17,6 @@ shim layer architecture, see `AGENTS.md` at the repo root.
 - [ ] H5: Configuration — new RapidsConf without docs/defaults; should use .internal() if not user-visible; new features default off
 - [ ] H6: Magic numbers — unexplained numeric literals without named constants or comments
 - [ ] H7: Pre-merge CI gaps — only selected shims run unit tests; feature-gated tests need explicit enable; limited Scala 2.13 coverage
-- [ ] H8: Upstream dependencies — SNAPSHOT changes from spark-rapids-jni/cudf may break; verify API usage against upstream repos
+- [ ] H8: Upstream dependencies — SNAPSHOT changes from cudf-spark-jni/cudf may break; verify API usage against upstream repos
 - [ ] H9: Databricks coverage — change may behave differently on Databricks (filesystem/path semantics, plan-string/optimizer assertions, DBR Spark-fork runtime) but the PR title lacks [databricks] and the diff doesn't touch a *db* shim path or a path containing `databricks` (either auto-triggers DB CI). Recommend [databricks] or confirm DB validation (see AGENTS.md § PR title tags)
 - [ ] H10: Performance checklist — report `Performance: Not required` as a high-severity finding unless the PR is documentation-only or test-only, or its description gives a verifiable reason the change cannot affect runtime performance. A bug-fix label, small diff, or rarely used path is not by itself an exemption. When uncertain, flag

aggregator/pom.xml

@@ -166,8 +166,8 @@
                     <artifactSet>
                         <excludes>
                             <exclude>org.slf4j:*</exclude>
-                            <!-- excludes spark-rapids-jni and jucx, will package them in the dist submodule -->
-                            <exclude>com.nvidia:spark-rapids-jni:*</exclude>
+                            <!-- excludes cudf-spark-jni and jucx, will package them in the dist submodule -->
+                            <exclude>com.nvidia:${cudf-spark-jni.artifactId}:*</exclude>
                             <exclude>org.openucx:jucx:*</exclude>
                         </excludes>
                     </artifactSet>

api_validation/README.md

@@ -11,7 +11,7 @@ Validation fails when:
 
 # Dependencies
 
-It requires spark-rapids-jni, rapids-4-spark and Spark jars.
+It requires cudf-spark-jni, rapids-4-spark and Spark jars.
 
 # Running the script
 

api_validation/pom.xml

@@ -43,7 +43,7 @@
         </dependency>
         <dependency>
             <groupId>com.nvidia</groupId>
-            <artifactId>spark-rapids-jni</artifactId>
+            <artifactId>${cudf-spark-jni.artifactId}</artifactId>
             <classifier>${jni.classifier}</classifier>
             <scope>provided</scope>
         </dependency>

build/make-scala-version-build-files.sh

@@ -105,7 +105,7 @@ sed_i '/<java\.major\.version>/,/<spark\.version>\${spark[0-9]\+\.version}</s/<s
 
 # Update <scala.binary.version> in parent POM
 # Match any scala binary version to ensure idempotency
-sed_i '/<spark\-rapids\-jni\.version>/,/<scala\.binary\.version>[0-9]*\.[0-9]*</s/<scala\.binary\.version>[0-9]*\.[0-9]*</<scala.binary.version>'$TO_VERSION'</' \
+sed_i '/<cudf\-spark\-jni\.version>/,/<scala\.binary\.version>[0-9]*\.[0-9]*</s/<scala\.binary\.version>[0-9]*\.[0-9]*</<scala.binary.version>'$TO_VERSION'</' \
   "$TO_DIR/pom.xml"
 
 
@@ -126,5 +126,5 @@ if [[ -z "$SCALA_VERSION" ]]; then
   exit 1
 fi
 
-sed_i '/<spark\-rapids\-jni\.version>/,/<scala.version>[0-9]*\.[0-9]*\.[0-9]*</s/<scala\.version>[0-9]*\.[0-9]*\.[0-9]*</<scala.version>'$SCALA_VERSION'</' \
+sed_i '/<cudf\-spark\-jni\.version>/,/<scala.version>[0-9]*\.[0-9]*\.[0-9]*</s/<scala\.version>[0-9]*\.[0-9]*\.[0-9]*</<scala.version>'$SCALA_VERSION'</' \
   "$TO_DIR/pom.xml"

dist/pom.xml

@@ -445,7 +445,7 @@ self.log("... OK")
                                 <!-- Per-artifact markers refresh unpacked files when a source JAR is newer. -->
                                 <artifactItem>
                                     <groupId>com.nvidia</groupId>
-                                    <artifactId>spark-rapids-jni</artifactId>
+                                    <artifactId>${cudf-spark-jni.artifactId}</artifactId>
                                     <classifier>${jni.classifier}</classifier>
                                     <excludes>META-INF/**</excludes>
                                     <outputDirectory>${project.build.directory}/jni-deps</outputDirectory>

docs/additional-functionality/advanced_configs.md

@@ -71,7 +71,7 @@ Name | Description | Default Value | Applicable at
 <a name="shuffle.ucx.managementServerHost"></a>spark.rapids.shuffle.ucx.managementServerHost|The host to be used to start the management server|null|Startup
 <a name="shuffle.ucx.useWakeup"></a>spark.rapids.shuffle.ucx.useWakeup|When set to true, use UCX's event-based progress (epoll) in order to wake up the progress thread when needed, instead of a hot loop.|true|Startup
 <a name="sql.acceleratedColumnarToRow.enabled"></a>spark.rapids.sql.acceleratedColumnarToRow.enabled|When set to true (default) the GPU columnar-to-row transition uses the GPU transpose kernel (AcceleratedColumnarToRowIterator) for wide fixed-width / STRING schemas. Setting it to false forces the slower per-row CPU iterator (ColumnarToRowIterator). Mainly useful for troubleshooting and performance comparisons; production workloads should leave this on.|true|Runtime
-<a name="sql.allowMultipleJars"></a>spark.rapids.sql.allowMultipleJars|Allow multiple rapids-4-spark, spark-rapids-jni, and cudf jars on the classpath. Spark will take the first one it finds, so the version may not be expected. Possisble values are ALWAYS: allow all jars, SAME_REVISION: only allow jars with the same revision, NEVER: do not allow multiple jars at all.|SAME_REVISION|Startup
+<a name="sql.allowMultipleJars"></a>spark.rapids.sql.allowMultipleJars|Allow multiple rapids-4-spark, cudf-spark-jni, and cudf jars on the classpath. Spark will take the first one it finds, so the version may not be expected. Possible values are ALWAYS: allow all jars, SAME_REVISION: only allow jars with the same revision, NEVER: do not allow multiple jars at all.|SAME_REVISION|Startup
 <a name="sql.castDecimalToFloat.enabled"></a>spark.rapids.sql.castDecimalToFloat.enabled|Casting from decimal to floating point types on the GPU returns results that have tiny difference compared to results returned from CPU.|true|Runtime
 <a name="sql.castFloatToDecimal.enabled"></a>spark.rapids.sql.castFloatToDecimal.enabled|Casting from floating point types to decimal on the GPU returns results that have tiny difference compared to results returned from CPU.|true|Runtime
 <a name="sql.castFloatToIntegralTypes.enabled"></a>spark.rapids.sql.castFloatToIntegralTypes.enabled|Casting from floating point types to integral types on the GPU supports a slightly different range of values when using Spark 3.1.0 or later. Refer to the CAST documentation for more details.|true|Runtime

iceberg-common/pom.xml

@@ -42,7 +42,7 @@
     <dependencies>
         <dependency>
             <groupId>com.nvidia</groupId>
-            <artifactId>spark-rapids-jni</artifactId>
+            <artifactId>${cudf-spark-jni.artifactId}</artifactId>
             <classifier>${jni.classifier}</classifier>
         </dependency>
         <dependency>

jenkins/spark-nightly-build.sh

@@ -17,7 +17,7 @@
 
 set -ex
 
-## MVN_OPT : maven options environment, e.g. MVN_OPT='-Dspark-rapids-jni.version=xxx' to specify spark-rapids-jni dependency's version.
+## MVN_OPT : maven options environment, e.g. MVN_OPT='-Dcudf-spark-jni.version=xxx' to specify cudf-spark-jni dependency's version.
 MVN_SETTINGS=${MVN_SETTINGS:-"jenkins/settings.xml"}
 export MVN="mvn -s $MVN_SETTINGS -Dmaven.wagon.http.retryHandler.count=3 -DretryFailedDeploymentCount=3 ${MVN_OPT} -Psource-javadoc"
 

pom.xml

@@ -1061,7 +1061,8 @@
         <spark.version.classifier>spark${buildver}</spark.version.classifier>
         <cuda.version>cuda12</cuda.version>
         <jni.classifier>${cuda.version}</jni.classifier>
-        <spark-rapids-jni.version>26.10.0-SNAPSHOT</spark-rapids-jni.version>
+        <cudf-spark-jni.version>26.10.0-SNAPSHOT</cudf-spark-jni.version>
+        <cudf-spark-jni.artifactId>cudf-spark-jni</cudf-spark-jni.artifactId>
         <cudf-spark-private.version>26.10.0-SNAPSHOT</cudf-spark-private.version>
         <cudf-spark-private.artifactId>cudf-spark-private_${scala.binary.version}</cudf-spark-private.artifactId>
         <scala.binary.version>2.12</scala.binary.version>
@@ -1275,8 +1276,8 @@
             </dependency>
             <dependency>
               <groupId>com.nvidia</groupId>
-              <artifactId>spark-rapids-jni</artifactId>
-              <version>${spark-rapids-jni.version}</version>
+              <artifactId>${cudf-spark-jni.artifactId}</artifactId>
+              <version>${cudf-spark-jni.version}</version>
               <classifier>${jni.classifier}</classifier>
             </dependency>
             <dependency>
@@ -1572,7 +1573,7 @@ This will force full Scala code rebuild in downstream modules.
                                             failonerror="false">
                                             <arg value="${spark.rapids.source.basedir}/build/build-info"/>
                                             <arg value="${project.version}"/>
-                                            <arg value="${spark-rapids-jni.version}"/>
+                                            <arg value="${cudf-spark-jni.version}"/>
                                         </exec>
                                         <fail message="exec build-info.sh failed, exit code is ${build-info.exitCode}, error msg is ${build-info.errorMsg}">
                                             <condition>

scala2.13/aggregator/pom.xml

@@ -166,8 +166,8 @@
                     <artifactSet>
                         <excludes>
                             <exclude>org.slf4j:*</exclude>
-                            <!-- excludes spark-rapids-jni and jucx, will package them in the dist submodule -->
-                            <exclude>com.nvidia:spark-rapids-jni:*</exclude>
+                            <!-- excludes cudf-spark-jni and jucx, will package them in the dist submodule -->
+                            <exclude>com.nvidia:${cudf-spark-jni.artifactId}:*</exclude>
                             <exclude>org.openucx:jucx:*</exclude>
                         </excludes>
                     </artifactSet>

scala2.13/api_validation/pom.xml

@@ -43,7 +43,7 @@
         </dependency>
         <dependency>
             <groupId>com.nvidia</groupId>
-            <artifactId>spark-rapids-jni</artifactId>
+            <artifactId>${cudf-spark-jni.artifactId}</artifactId>
             <classifier>${jni.classifier}</classifier>
             <scope>provided</scope>
         </dependency>

scala2.13/dist/pom.xml

@@ -445,7 +445,7 @@ self.log("... OK")
                                 <!-- Per-artifact markers refresh unpacked files when a source JAR is newer. -->
                                 <artifactItem>
                                     <groupId>com.nvidia</groupId>
-                                    <artifactId>spark-rapids-jni</artifactId>
+                                    <artifactId>${cudf-spark-jni.artifactId}</artifactId>
                                     <classifier>${jni.classifier}</classifier>
                                     <excludes>META-INF/**</excludes>
                                     <outputDirectory>${project.build.directory}/jni-deps</outputDirectory>

scala2.13/iceberg-common/pom.xml

@@ -42,7 +42,7 @@
     <dependencies>
         <dependency>
             <groupId>com.nvidia</groupId>
-            <artifactId>spark-rapids-jni</artifactId>
+            <artifactId>${cudf-spark-jni.artifactId}</artifactId>
             <classifier>${jni.classifier}</classifier>
         </dependency>
         <dependency>

scala2.13/pom.xml

@@ -1061,7 +1061,8 @@
         <spark.version.classifier>spark${buildver}</spark.version.classifier>
         <cuda.version>cuda12</cuda.version>
         <jni.classifier>${cuda.version}</jni.classifier>
-        <spark-rapids-jni.version>26.10.0-SNAPSHOT</spark-rapids-jni.version>
+        <cudf-spark-jni.version>26.10.0-SNAPSHOT</cudf-spark-jni.version>
+        <cudf-spark-jni.artifactId>cudf-spark-jni</cudf-spark-jni.artifactId>
         <cudf-spark-private.version>26.10.0-SNAPSHOT</cudf-spark-private.version>
         <cudf-spark-private.artifactId>cudf-spark-private_${scala.binary.version}</cudf-spark-private.artifactId>
         <scala.binary.version>2.13</scala.binary.version>
@@ -1275,8 +1276,8 @@
             </dependency>
             <dependency>
               <groupId>com.nvidia</groupId>
-              <artifactId>spark-rapids-jni</artifactId>
-              <version>${spark-rapids-jni.version}</version>
+              <artifactId>${cudf-spark-jni.artifactId}</artifactId>
+              <version>${cudf-spark-jni.version}</version>
               <classifier>${jni.classifier}</classifier>
             </dependency>
             <dependency>
@@ -1572,7 +1573,7 @@ This will force full Scala code rebuild in downstream modules.
                                             failonerror="false">
                                             <arg value="${spark.rapids.source.basedir}/build/build-info"/>
                                             <arg value="${project.version}"/>
-                                            <arg value="${spark-rapids-jni.version}"/>
+                                            <arg value="${cudf-spark-jni.version}"/>
                                         </exec>
                                         <fail message="exec build-info.sh failed, exit code is ${build-info.exitCode}, error msg is ${build-info.errorMsg}">
                                             <condition>

scala2.13/shuffle-plugin/pom.xml

@@ -39,7 +39,7 @@
     <dependencies>
         <dependency>
             <groupId>com.nvidia</groupId>
-            <artifactId>spark-rapids-jni</artifactId>
+            <artifactId>${cudf-spark-jni.artifactId}</artifactId>
             <classifier>${jni.classifier}</classifier>
         </dependency>
         <dependency>

scala2.13/sql-plugin-api/pom.xml

@@ -38,7 +38,7 @@
     <dependencies>
         <dependency>
             <groupId>com.nvidia</groupId>
-            <artifactId>spark-rapids-jni</artifactId>
+            <artifactId>${cudf-spark-jni.artifactId}</artifactId>
             <classifier>${jni.classifier}</classifier>
         </dependency>
     </dependencies>
@@ -132,4 +132,4 @@
             </build>
         </profile>
     </profiles>
-</project>
+</project>

scala2.13/sql-plugin-columnar/pom.xml

@@ -41,7 +41,7 @@
     <dependencies>
         <dependency>
             <groupId>com.nvidia</groupId>
-            <artifactId>spark-rapids-jni</artifactId>
+            <artifactId>${cudf-spark-jni.artifactId}</artifactId>
             <classifier>${jni.classifier}</classifier>
         </dependency>
         <dependency>

scala2.13/sql-plugin-fileio/pom.xml

@@ -41,7 +41,7 @@
     <dependencies>
         <dependency>
             <groupId>com.nvidia</groupId>
-            <artifactId>spark-rapids-jni</artifactId>
+            <artifactId>${cudf-spark-jni.artifactId}</artifactId>
             <classifier>${jni.classifier}</classifier>
         </dependency>
         <dependency>

scala2.13/sql-plugin/pom.xml

@@ -45,7 +45,7 @@
         </dependency>
         <dependency>
             <groupId>com.nvidia</groupId>
-            <artifactId>spark-rapids-jni</artifactId>
+            <artifactId>${cudf-spark-jni.artifactId}</artifactId>
             <classifier>${jni.classifier}</classifier>
         </dependency>
         <dependency>

scala2.13/tests/pom.xml

@@ -69,7 +69,7 @@
         </dependency>
         <dependency>
             <groupId>com.nvidia</groupId>
-            <artifactId>spark-rapids-jni</artifactId>
+            <artifactId>${cudf-spark-jni.artifactId}</artifactId>
             <classifier>${jni.classifier}</classifier>
         </dependency>
         <!-- use aggregator jar because accessing internal classes -->

scala2.13/udf-compiler/pom.xml

@@ -39,7 +39,7 @@
     <dependencies>
         <dependency>
             <groupId>com.nvidia</groupId>
-            <artifactId>spark-rapids-jni</artifactId>
+            <artifactId>${cudf-spark-jni.artifactId}</artifactId>
             <classifier>${jni.classifier}</classifier>
         </dependency>
         <dependency>