Alfresco Node Processor

Giro... vedo nodi... faccio cose...

Do things with nodes.

A modern, threaded and easily customizable Spring Boot Application that - given a means for collecting nodes - do something with them.

Think about this as a template for your application.

Pull requests are welcome!

Features

• QueryNodeCollector collect nodes with Alfresco FTS queries
• NodeListCollector reads node IDs from a file
• NodeTreeCollector walks the repository tree
• DeleteNodeProcessor deletes or trashes nodes
• MoveNodeProcessor relocates nodes under a new parent
• AddAspectsAndSetPropertiesProcessor adds aspects and properties
• SetPermissionsProcessor applies permissions and inheritance
• DownloadNodeProcessor saves node content and metadata to the filesystem
• ChainingNodeProcessor executes multiple processors sequentially
• Queue based architecture with configurable consumer threads
• Easily extensible by implementing AbstractNodeCollector and AbstractNodeProcessor

Customize

If none of the predefined Collectors/Processors meet your needs, simply write your own by extending the abstract ones. Just inject the required handlers (e.g., NodesApi) and override the relevant methods.

Collecting nodes

QueryNodeCollector

The QueryNodeCollector takes an Alfresco FTS query, execute it on a separate thread and feed the queue:

"collector": {
  "name": "QueryNodeCollector",
  "args": {
    "query": "PATH:'/app:company_home/*' AND TYPE:'cm:folder'"
  }
}

the default page size for search is 100 and can be modified by passing an additional argument to the collector:

"batch-size": 1000

NodeListCollector

The NodeListCollector takes an input file containing a list of node-id with each id on a separate line, e.g.:

e72b6596-ec2e-4279-b490-3a03b119d8de
d78c0036-15c0-43cf-89e4-cd198d14b626
1a7ecc34-de06-45ed-85c0-76f8355f3724

and the path of the file need to be specified in the config:

  "collector": {
      "name": "NodeListCollector",
      "args": {
        "node-list-file": "/tmp/node-ids.txt"
      }
  }

NodeTreeCollector

Iteratively walk the tree starting from a folder node given either its id or its repository path:

"collector": {
  "name": "NodeTreeCollector",
  "args": {
    "path": "/"
  }
}

The collector automatically descends into folders. The default page size for listNodeChildren is 100 and can be modified by passing an additional argument to the collector:

"batch-size": 200

Processing nodes

DeleteNodeProcessor

Delete the collected nodes, set the permanent flag to true if you want to delete the nodes directly rather than move them into the trashcan:

"processor": {
  "name": "DeleteNodeProcessor",
  "args": {
    "permanent": true
  }
}

AddAspectsAndSetPropertiesProcessor

Add a list of aspects and apply a map of properties to the collected nodes:

"processor": {
  "name": "AddAspectsAndSetPropertiesProcessor",
  "args": {
    "properties": {
      "cm:publisher": "saidone",
      "cm:contributor": "saidone"
    },
    "aspects": [
      "cm:dublincore"
    ]
  }
}

SetPermissionsProcessor

Apply a list of permissions and set inheritance flag to the collected nodes:

"processor": {
  "name": "SetPermissionsProcessor",
  "args": {
    "permissions": {
      "isInheritanceEnabled": false,
      "locallySet": [
        {
          "authorityId": "GROUP_EVERYONE",
          "name": "Collaborator",
          "accessStatus": "ALLOWED"
        }
      ]
    }
  }
}

MoveNodeProcessor

Move collected nodes to a new folder identified either by its node-id or by the repository path:

"processor": {
  "name": "MoveNodeProcessor",
  "args": {
    "target-parent-id": "e72b6596-ec2e-4279-b490-3a03b119d8de"
  }
}

DownloadNodeProcessor

Download node content and metadata to a local directory in a format compatible with bulk import:

"processor": {
  "name": "DownloadNodeProcessor",
  "args": {
    "output-dir": "/tmp/export"
  }
}

ChainingNodeProcessor

Execute a list of processors sequentially on each node:

"processor": {
  "name": "ChainingNodeProcessor",
  "args": {
    "processors": [
      { "name": "LogNodeNameProcessor" },
      { "name": "VoidProcessor" }
    ]
  }
}

Custom processors

Custom processors can be easily created by extending the AbstractNodeProcessor and overriding the processNode method:

@Component
@Slf4j
public class LogNodeNameProcessor extends AbstractNodeProcessor {

    @Autowired
    private NodesApi nodesApi;

    @Override
    public void processNode(String nodeId, Config config) {
        var node = Objects.requireNonNull(nodesApi.getNode(nodeId, null, null, null).getBody()).getEntry();
        log.debug("node name --> {}", node.getName());
    }

}

Build

Java and Maven required

mvn package -DskipTests -Dlicense.skip=true

look at the build.sh or build.bat scripts for creating a convenient distribution package.

Application global config

Global configuration is stored in config/application.yml file, the relevant parameters are:

Parameter/env variable	Key in application.yml	Default value	Purpose
ALFRESCO_BASE_PATH	content.service.url	http://localhost:8080	scheme, host and port of the Alfresco server
ALFRESCO_USERNAME	content.service.security.basicAuth.username	admin	Alfresco user
ALFRESCO_PASSWORD	content.service.security.basicAuth.password	admin	password for the Alfresco user
QUEUE_SIZE	application.queue-size	1000	size of the node-uuid queue
CONSUMER_THREADS	application.consumer-threads	4	number of consumers that are executed simultaneously
CONSUMER_TIMEOUT	application.consumer-timeout	5000	milliseconds after which a consumer gives up waiting for data in the queue
READ_ONLY	application.read-only	true	when true, mutating operations on nodes are skipped

Testing

For integration tests just change configuration and point it to an existing Alfresco installation, or use alfresco.(sh|bat) script to start it with docker.

Run

$ java -jar anp.jar -c example-log-node-name.json

Further documentation

See Javadoc

License

Copyright (c) 2023-2025 Saidone

Distributed under the GNU General Public License v3.0