Advanced Task Drupal DB Cleaner

The advanced task drupal DB cleaner is designed to achieve two things

1. Return a report of matching tables in a Drupal which the cleaner considers removeable (nukeable)
2. DROP TABLES that match the pattern

The approach taken leverages https://github.com/uselagoon/advanced-task-toolbox, and follows the guidance of that project.

Docker files

The repo contains two dockerfiles which usa a FROM amazeeio/advanced-task-toolbox:latest

• drupclean_report.dockerfile : Builds and image for the reporting task
• drupclean_nuke.dockerfile : Builds and image for the removing (nuking) task

There is a docker-compose.yml for building the containers locally under development.

Advanced Task Tookbox

There are two scripts in the scripts/ directory, one for reporting and one for nuking. The docker build copies the relevant script to default.yml for the advanced task execution.

Drupclean Code

In the drupclean directory you will find the code that is called by the Advanced Task Toolbox scripts. The code is designed to be run as a drush scr call.

For development, there is also a drupclean_dev.php which will create 1000 "bad" tables that can be used to test the script.

Image building

When the code is pushed to Github, and action runs which will

1. Build the two images
2. Log into Dockerhub
3. Push the tagged images to Dockerhub as

• aio-advanced-task-drupal-db-cleaner:report-latest
• aio-advanced-task-drupal-db-cleaner:nuker-latest

Creating the Advanced Task in Lagoon

The relevant GraphQL template for the two Advanced Tasks are available in create_task_*.gql. Please note that these queries need at least the project: id to be set correctly. This could be changed to environment: if needed, or could be added to a whole group. Please see the Lagoon docs (http://docs.lagoon.sh) for more details on Advanced Tasks.

Helpful

• https://x-team.com/blog/drupal-field-name/