Move both Iceberg and S3A CICD to snapshot builds (#186)

## Description of change

This commit makes sure we can use `-PsnapshotBuild=true` not only for
publishing artefacts but also to build them. This is for convenience:
for CICD, we always want to test the latest built artefact. Such an
artefact must:
1. Have a name that does not exist in any repositories (either local
Maven cache or public Maven central).
2. Have a name that does not change across versions to make it easy to
point automation to the latest artefact.

After this commit, Hadoop build should be healthy.

Commit was tested locally, by invoking manual builds and inspecting
artefacts. There is no mention of old artefact names like
`input-stream.jar`, `object-client.jar`, `common.jar` anymore in the
code base.

.github/workflows/build-upload.yml

@@ -36,27 +36,12 @@ jobs:
 
       - name: Build with Gradle
         run: |
-          ./gradlew build
+          ./gradlew -PsnapshotBuild=true build
 
       - uses: actions/upload-artifact@v4
         with:
-          path: "common/build/libs/common.jar"
-          name: "common.jar"
-
-      - uses: actions/upload-artifact@v4
-        with:
-          path: "input-stream/build/libs/input-stream-all.jar"
-          name: "input-stream-all.jar"
-
-      - uses: actions/upload-artifact@v4
-        with:
-          path: "input-stream/build/libs/input-stream.jar"
-          name: "input-stream.jar"
-
-      - uses: actions/upload-artifact@v4
-        with:
-          path: "object-client/build/libs/object-client.jar"
-          name: "object-client.jar"
+          path: "input-stream/build/libs/analyticsaccelerator-s3-SNAPSHOT.jar"
+          name: "analyticsaccelerator-s3-SNAPSHOT.jar"
 
   BuildHadoopAndUploadArtifact:
     name: Build Hadoop artifacts
@@ -73,10 +58,7 @@ jobs:
 
       - name: Build and publish to local Maven with Gradle
         run: |
-          ./gradlew publishToMavenLocal
-
-      - name: Publish uber jar to local Maven
-        run: mvn install:install-file -Dfile=input-stream/build/libs/input-stream-all.jar -DgroupId=com.amazon.connector.s3 -DartifactId=input-stream -Dversion=1.0.0 -Dpackaging=jar -DgeneratePom=true
+          ./gradlew -PsnapshotBuild=true publishToMavenLocal
 
       - name: Setup Hadoop SSH deploy key
         uses: webfactory/ssh-agent@v0.9.0
@@ -161,19 +143,7 @@ jobs:
     steps:
       - uses: actions/download-artifact@v4
         with:
-          name: "common.jar"
-
-      - uses: actions/download-artifact@v4
-        with:
-          name: "input-stream-all.jar"
-
-      - uses: actions/download-artifact@v4
-        with:
-          name: "input-stream.jar"
-
-      - uses: actions/download-artifact@v4
-        with:
-          name: "object-client.jar"
+          name: "analyticsaccelerator-s3-SNAPSHOT.jar"
 
       - uses: actions/download-artifact@v4
         with:
@@ -190,23 +160,11 @@ jobs:
           role-session-name: GitHub_to_AWS_via_FederatedOIDC
           aws-region: ${{ env.AWS_REGION }}
 
-      - name: Upload common JAR to S3a treatment bucket
-        run: aws s3 cp common.jar s3://${{ env.S3_BUCKET }}/s3a/common-1.0.0.jar
-
-      - name: Upload common JAR to S3FileIO treatment bucket
-        run: aws s3 cp common.jar s3://${{ env.S3_BUCKET }}/s3fileio/common.jar
-
-      - name: Upload input-stream JAR to S3a treatment bucket
-        run: aws s3 cp input-stream-all.jar s3://${{ env.S3_BUCKET }}/s3a/input-stream-1.0.0.jar
-
-      - name: Upload input-stream JAR to S3FileIO treatment bucket
-        run: aws s3 cp input-stream.jar s3://${{ env.S3_BUCKET }}/s3fileio/input-stream.jar
-
-      - name: Upload object-client JAR to S3a treatment bucket
-        run: aws s3 cp object-client.jar s3://${{ env.S3_BUCKET }}/s3a/object-client-1.0.0.jar
+      - name: Upload analyticsaccelerator JAR to S3a treatment bucket
+        run: aws s3 cp analyticsaccelerator-s3-SNAPSHOT.jar s3://${{ env.S3_BUCKET }}/s3a/analyticsaccelerator-s3-SNAPSHOT.jar
 
-      - name: Upload object-client JAR to S3FileIO treatment bucket
-        run: aws s3 cp object-client.jar s3://${{ env.S3_BUCKET }}/s3fileio/object-client.jar
+      - name: Upload analyticsaccelerator JAR to S3FileIO treatment bucket
+        run: aws s3 cp analyticsaccelerator-s3-SNAPSHOT.jar s3://${{ env.S3_BUCKET }}/s3fileio/analyticsaccelerator-s3-SNAPSHOT.jar
 
       - name: Upload Iceberg JAR to S3FileIO treatment bucket
         run: aws s3 cp iceberg-spark-runtime-3.5_2.12-1.6.0-SNAPSHOT.jar s3://${{ env.S3_BUCKET }}/s3fileio/iceberg-spark-runtime-3.5_2.12-1.6.0-SNAPSHOT.jar

input-stream/build.gradle.kts

@@ -10,7 +10,10 @@ import com.github.jk1.license.render.TextReportRenderer
 
 val group = "software.amazon.s3.analyticsaccelerator"
 val artefact = "analyticsaccelerator-s3"
-val currentVersion = "0.0.1"
+val currentVersionNumber = "0.0.1"
+
+val isSnapshot = findProperty("snapshotBuild") == "true"
+val currentVersion = if (isSnapshot) "SNAPSHOT" else currentVersionNumber;
 
 plugins {
     id("buildlogic.java-library-conventions")
@@ -282,8 +285,7 @@ publishing {
             groupId = group
             artifactId = artefact
 
-            val isSnapshot = findProperty("snapshotBuild") == "true"
-            version = if (isSnapshot) "SNAPSHOT" else currentVersion;
+            version = currentVersion
 
             pom {
                 name = "S3 Analytics Accelerator Library for Amazon S3"