README.md

Clean up using AQL

AQL can be used to easily clean up unwanted artifacts, for example, artifacts with a combination of:

• low artifact maturity
• no recent downloads
• older than a certain threshold

Queries matching unwanted artifacts can passed into the JFrog CLI's delete command to easily tidy up.

Resources

The following will come in handy during the exercises:

• The AQL docs
• The Filespec reference, when using the JFrog CLI
• The JFrog CLI docs

You can wrap AQL in a Filespec, for example:

{
  "files": [
    {
      "aql": {
        "items.find": {
          "repo": "fofobo-generic-local"
        }
      }
    }
  ]
}

The following is an example of a JFrog CLI call to delete artifacts matching a given Filespec:

./jfrog rt d --spec=filespec.json

• rt stands for Artifactory
• d stand for delete
• --spec= defines where the filespec is located

Setup

• Run the setup script
• Download the JFrog CLI
• Run jf rt c to configure jfrog cli with the following information
  • Artifactory server ID: training
  • Artifactory URL: See setup script output
  • API key: See your profile in Artifactory

Tasks

Note: Remember to limit your Filespecs to your repositories only.

• Write a query that finds artifacts:
  • in your sandbox or dev repository
  • with less than 2 downloads
  • without the keep property set to true
• Run a search command using your query and verify the found artifacts match expectations
• Then finally, issue a delete command using your query
• Browse Artifactory and confirm the files have been cleaned up

Extra

We in eficode have created a tool that will help you with the Artifact deletions, based on AQL. The tools is based in the tools folder, where you can find the instructions to use it.