Patch NiFi assembly to not zip binaries to save disk space

CHANGELOG.md

@@ -31,6 +31,7 @@ All notable changes to this project will be documented in this file.
 - pass platform argument to preflight check ([#626]).
 - nifi: provision stackable-bcrypt from Maven ([#663])
 - nifi: move /bin/stackable-bcrypt to /stackable/stackable-bcrypt and added softlink for backwards compatibility ([#678]).
+- nifi: patch nifi-assembly pom file to not zip binaries after the build to save disk space ([#685]).
 - hadoop: use java-devel as base layer for the builder stage ([#665])
 - hive: use java-devel as base layer for the builder stage ([#665])
 - zookeeper: use java-devel as base layer for the builder stage ([#665])
@@ -81,6 +82,7 @@ All notable changes to this project will be documented in this file.
 [#678]: https://github.com/stackabletech/docker-images/pull/678
 [#679]: https://github.com/stackabletech/docker-images/pull/679
 [#682]: https://github.com/stackabletech/docker-images/pull/682
+[#685]: https://github.com/stackabletech/docker-images/pull/685
 
 ## [24.3.0] - 2024-03-20
 

nifi/Dockerfile

@@ -17,6 +17,8 @@ RUN microdnf update && \
 USER stackable
 WORKDIR /stackable
 
+COPY --chown=stackable:stackable nifi/stackable/patches /stackable/patches
+
 RUN curl --fail -L 'https://repo.stackable.tech/repository/m2/tech/stackable/nifi/stackable-bcrypt/1.0-SNAPSHOT/stackable-bcrypt-1.0-20240508.153334-1-jar-with-dependencies.jar' \
     # This used to be located in /bin/stackable-bcrypt.jar. We create a softlink for /bin/stackable-bcrypt.jar in the main container for backwards compatibility.
     -o /stackable/stackable-bcrypt.jar && \
@@ -29,12 +31,18 @@ RUN curl --fail -L 'https://repo.stackable.tech/repository/m2/tech/stackable/nif
     # from /stackable/nifi-${PRODUCT}-src/nifi-assembly/target/nifi-${PRODUCT}-bin/nifi-${PRODUCT} (see later steps)
     # Therefore we add the suffix "-src" to be able to copy the binary and remove the unzipped sources afterwards.
     mv nifi-${PRODUCT} nifi-${PRODUCT}-src && \
+    # Apply patches
+    chmod +x patches/apply_patches.sh && \
+    patches/apply_patches.sh ${PRODUCT} && \
+    # Build NiFi
     cd /stackable/nifi-${PRODUCT}-src/ && \
     mvn clean install -DskipTests && \
     # Copy the binaries to the /stackable folder
     mv /stackable/nifi-${PRODUCT}-src/nifi-assembly/target/nifi-${PRODUCT}-bin/nifi-${PRODUCT} /stackable/nifi-${PRODUCT} && \
     # Remove the unzipped sources
-    rm -rf /stackable/nifi-${PRODUCT}-src 
+    rm -rf /stackable/nifi-${PRODUCT}-src && \
+    # Remove generated docs in binary
+    rm -rf /stackable/nifi-${PRODUCT}/docs
 
 # Add Iceberg extensions as they are not included by default and are important enough
 # They need to be build from source, as https://mvnrepository.com/artifact/org.apache.nifi/nifi-iceberg-processors-nar does not ship the org.apache.hadoop.fs.s3a.S3AFileSystem (see https://github.com/apache/nifi/pull/6368#issuecomment-1502175258)

nifi/stackable/patches/1.21.0/001-NIFI-no-zip-assembly.patch

@@ -0,0 +1,12 @@
+diff --git a/nifi-assembly/pom.xml b/nifi-assembly/pom.xml
+index 3206b0468d..5ae52932cc 100644
+--- a/nifi-assembly/pom.xml
++++ b/nifi-assembly/pom.xml
+@@ -66,7 +66,6 @@ language governing permissions and limitations under the License. -->
+                             <tarLongFileMode>posix</tarLongFileMode>
+                             <formats>
+                                 <format>dir</format>
+-                                <format>zip</format>
+                             </formats>
+                         </configuration>
+                     </execution>

nifi/stackable/patches/1.23.2/001-NIFI-no-zip-assembly.patch

@@ -0,0 +1,12 @@
+diff --git a/nifi-assembly/pom.xml b/nifi-assembly/pom.xml
+index b520c113ce..5cf04f421d 100644
+--- a/nifi-assembly/pom.xml
++++ b/nifi-assembly/pom.xml
+@@ -66,7 +66,6 @@ language governing permissions and limitations under the License. -->
+                             <tarLongFileMode>posix</tarLongFileMode>
+                             <formats>
+                                 <format>dir</format>
+-                                <format>zip</format>
+                             </formats>
+                         </configuration>
+                     </execution>

nifi/stackable/patches/1.25.0/001-NIFI-no-zip-assembly.patch

@@ -0,0 +1,12 @@
+diff --git a/nifi-assembly/pom.xml b/nifi-assembly/pom.xml
+index f2d236cd56..fdb3879c9f 100644
+--- a/nifi-assembly/pom.xml
++++ b/nifi-assembly/pom.xml
+@@ -66,7 +66,6 @@ language governing permissions and limitations under the License. -->
+                             <tarLongFileMode>posix</tarLongFileMode>
+                             <formats>
+                                 <format>dir</format>
+-                                <format>zip</format>
+                             </formats>
+                         </configuration>
+                     </execution>

nifi/stackable/patches/apply_patches.sh

@@ -0,0 +1,43 @@
+#!/usr/bin/env bash
+
+# Enable error handling and unset variable checking
+set -eu
+set -o pipefail
+
+# Check if $1 (VERSION) is provided
+if [ -z "${1-}" ]; then
+  echo "Please provide a value for VERSION as the first argument."
+  exit 1
+fi
+
+VERSION="$1"
+PATCH_DIR="patches/$VERSION"
+
+# Check if version-specific patches directory exists
+if [ ! -d "$PATCH_DIR" ]; then
+  echo "Patches directory '$PATCH_DIR' does not exist."
+  exit 1
+fi
+
+# Create an array to hold the patches in sorted order
+declare -a patch_files
+
+echo "Applying patches from ${PATCH_DIR}" now
+
+# Read the patch files into the array
+while IFS= read -r -d $'\0' file; do
+  patch_files+=("$file")
+done < <(find "$PATCH_DIR" -name "*.patch" -print0 | sort -zV)
+
+echo "Found ${#patch_files[@]} patches, applying now"
+
+# Iterate through sorted patch files
+for patch_file in "${patch_files[@]}"; do
+  echo "Applying $patch_file"
+  git apply --directory "nifi-${VERSION}-src" "$patch_file" || {
+    echo "Failed to apply $patch_file"
+    exit 1
+  }
+done
+
+echo "All patches applied successfully."