chore: CI improvements (#402)

* CI -> Moving lint to fail fast

* Removing node 20 from CI

* Improving CI

* Triggering CI for Kafka and core

* Trigger CI for sqs

* Fix lint

* Adding node 20 support note to readme

Author: CarlosGamero

.github/workflows/ci.common.yml

@@ -27,6 +27,9 @@ jobs:
       - name: Build TS
         run: npm run build -- --filter=${{ inputs.package_name }}
 
+      - name: Run lint
+        run: npm run lint -- --filter=${{ inputs.package_name }}
+
       - name: Docker start
         run: npm run docker:start:ci -- --filter=${{ inputs.package_name }}
 
@@ -35,6 +38,3 @@ jobs:
 
       - name: Docker stop
         run: npm run docker:stop:ci -- --filter=${{ inputs.package_name }}
-
-      - name: Run lint
-        run: npm run lint -- --filter=${{ inputs.package_name }}

.github/workflows/ci.yml

@@ -7,39 +7,69 @@ on:
   pull_request:
 
 jobs:
-  general:
-    strategy:
-      matrix:
-        node-version: [20.x, 22.x, 24.x]
-        package-name: [
-          '@message-queue-toolkit/amqp',
-          '@message-queue-toolkit/core',
-          '@message-queue-toolkit/metrics',
-          '@message-queue-toolkit/outbox-core',
-          '@message-queue-toolkit/redis-message-deduplication-store',
-          '@message-queue-toolkit/s3-payload-store',
-          '@message-queue-toolkit/schemas',
-          '@message-queue-toolkit/sns',
-          '@message-queue-toolkit/sqs',
-          '@message-queue-toolkit/gcp-pubsub',
-          '@message-queue-toolkit/gcs-payload-store'
-        ]
-    uses: ./.github/workflows/ci.common.yml
-    with:
-      node_version: ${{ matrix.node-version }}
-      package_name: ${{ matrix.package-name }}
+  changed-files-job:
+    name: Get changed packages
+    runs-on: ubuntu-latest
+    outputs:
+      packages: ${{ steps.detect.outputs.packages }}
+    steps:
+      - name: Get changed files
+        id: changed-files
+        uses: tj-actions/changed-files@ed68ef82c095e0d48ec87eccea555d944a631a4c # v46
+        with:
+          files: packages/**
+
+      - name: Detect changed packages
+        id: detect
+        env:
+          ALL_CHANGED_FILES: ${{ steps.changed-files.outputs.all_changed_files }}
+        run: |
+          declare -A PATH_TO_NAME=(
+            ["packages/amqp"]="@message-queue-toolkit/amqp"
+            ["packages/core"]="@message-queue-toolkit/core"
+            ["packages/gcp-pubsub"]="@message-queue-toolkit/gcp-pubsub"
+            ["packages/gcs-payload-store"]="@message-queue-toolkit/gcs-payload-store"
+            ["packages/kafka"]="@message-queue-toolkit/kafka"
+            ["packages/metrics"]="@message-queue-toolkit/metrics"
+            ["packages/outbox-core"]="@message-queue-toolkit/outbox-core"
+            ["packages/redis-message-deduplication-store"]="@message-queue-toolkit/redis-message-deduplication-store"
+            ["packages/s3-payload-store"]="@message-queue-toolkit/s3-payload-store"
+            ["packages/schemas"]="@message-queue-toolkit/schemas"
+            ["packages/sns"]="@message-queue-toolkit/sns"
+            ["packages/sqs"]="@message-queue-toolkit/sqs"
+          )
+
+          PACKAGES=()
+          for path in "${!PATH_TO_NAME[@]}"; do
+            if echo "$ALL_CHANGED_FILES" | grep -q "$path/"; then
+              PACKAGES+=("\"${PATH_TO_NAME[$path]}\"")
+            fi
+          done
 
-  kafka:
+          if [ ${#PACKAGES[@]} -eq 0 ]; then
+            echo 'packages=[]' >> $GITHUB_OUTPUT
+            echo "No packages changed"
+          else
+            JSON="[$(IFS=,; echo "${PACKAGES[*]}")]"
+            echo "packages=$JSON" >> $GITHUB_OUTPUT
+            echo "Changed packages: $JSON"
+          fi
+
+  general:
+    needs: [changed-files-job]
+    if: needs.changed-files-job.outputs.packages != '[]'
     strategy:
       matrix:
         node-version: [22.x, 24.x]
+        package-name: ${{ fromJson(needs.changed-files-job.outputs.packages) }}
     uses: ./.github/workflows/ci.common.yml
     with:
-      package_name: '@message-queue-toolkit/kafka'
       node_version: ${{ matrix.node-version }}
+      package_name: ${{ matrix.package-name }}
 
   automerge:
-    needs: [ general ]
+    needs: [general]
+    if: always() && (needs.general.result == 'success' || needs.general.result == 'skipped')
     runs-on: ubuntu-latest
     permissions:
       pull-requests: write

README.md

@@ -6,6 +6,9 @@
 
 Useful utilities, interfaces and base classes for message queue handling.
 
+> **Note:** Node.js 20 is no longer officially supported. It is likely to work, but we are no longer maintaining it.
+Please use a later version of Node.js.
+
 ## Overview
 
 `message-queue-toolkit ` is an abstraction over several different queue systems, which implements common deserialization,

packages/core/README.md

@@ -1,6 +1,7 @@
 # @message-queue-toolkit/core
 
-Core library for message-queue-toolkit. Provides foundational abstractions, utilities, and base classes for building message queue publishers and consumers.
+Core library for message-queue-toolkit. Provides foundational abstractions, utilities, and base classes for building
+message queue publishers and consumers.
 
 ## Table of Contents
 
@@ -74,13 +75,16 @@ type UserCreated = z.infer<typeof UserCreatedSchema>
 
 #### What is Message Type?
 
-The **message type** is a discriminator field that identifies what kind of event or command a message represents. It's used for:
+The **message type** is a discriminator field that identifies what kind of event or command a message represents. It's 
+used for:
 
 1. **Routing**: Directing messages to the appropriate handler based on their type
 2. **Schema validation**: Selecting the correct Zod schema to validate the message
 3. **Observability**: Tracking metrics and logs per message type
 
-In a typical event-driven architecture, a single queue or topic may receive multiple types of messages. For example, a `user-events` queue might receive `user.created`, `user.updated`, and `user.deleted` events. The message type tells the consumer which handler should process each message.
+In a typical event-driven architecture, a single queue or topic may receive multiple types of messages. For example, a
+`user-events` queue might receive `user.created`, `user.updated`, and `user.deleted` events. The message type tells the 
+consumer which handler should process each message.
 
 #### Configuration Options
 
@@ -126,7 +130,8 @@ const resolverConfig: MessageTypeResolverConfig = {
 }
 ```
 
-**Important:** The resolver function must always return a valid string. If the type cannot be determined, either return a default type or throw an error with a descriptive message.
+**Important:** The resolver function must always return a valid string. If the type cannot be determined, either return
+a default type or throw an error with a descriptive message.
 
 #### Real-World Examples by Platform
 
@@ -478,7 +483,8 @@ The third parameter to `addConfig` accepts these options:
 
 #### Explicit Message Type
 
-When using a custom resolver function (`messageTypeResolver: { resolver: fn }`), the message type cannot be automatically extracted from schemas at registration time. You must provide an explicit `messageType` for each handler:
+When using a custom resolver function (`messageTypeResolver: { resolver: fn }`), the message type cannot be automatically
+extracted from schemas at registration time. You must provide an explicit `messageType` for each handler:
 
 ```typescript
 const handlers = new MessageHandlerConfigBuilder<SupportedMessages, Context>()

packages/kafka/README.md

@@ -20,7 +20,8 @@ See [test consumer](test/consumer/PermissionConsumer.ts) for an example of imple
 
 ## Batch Processing
 
-Kafka supports batch processing for improved throughput. To enable it, set `batchProcessingEnabled` to `true` and configure `batchProcessingOptions`.
+Kafka supports batch processing for improved throughput. To enable it, set `batchProcessingEnabled` to `true` and 
+configure `batchProcessingOptions`.
 
 When batch processing is enabled, message handlers receive an array of messages instead of a single message.
 

packages/sqs/README.md

@@ -1,6 +1,7 @@
 # @message-queue-toolkit/sqs
 
-AWS SQS (Simple Queue Service) implementation for the message-queue-toolkit. Provides a robust, type-safe abstraction for publishing and consuming messages from both standard and FIFO SQS queues.
+AWS SQS (Simple Queue Service) implementation for the message-queue-toolkit. Provides a robust, type-safe abstraction
+for publishing and consuming messages from both standard and FIFO SQS queues.
 
 ## Table of Contents