fucodo/seed

A tiny seeding helper for Neos Flow applications. It provides a CLI command to initialize a database with a set of SQL files and optional follow‑up commands, orchestrated via Flow settings. This is intended for first‑time bootstrap of an environment (e.g. CI, demo or developer machines).

Features

• Declarative jobs configured in Settings.yaml
• Import ordered SQL files (by numeric keys)
• Optional follow‑up shell/Flow commands after import
• Safe‑guard: skips when your database already contains flow_doctrine_migrationstatus

Requirements

• PHP compatible with your Neos/Flow version
• Neos Flow (package type neos-package)

Installation

Add the package via Composer (as part of your Flow distribution):

composer require fucodo/seed

The package key is fucodo.seed.

Configuration

Define one or more seeding jobs under fucodo.seed.jobs. Each job can

• be enabled/disabled
• list SQL files to import (in order)
• optionally list commands to execute after SQL import

Example minimal configuration (place into Configuration/Settings.fucodo.seed.yaml in your distribution or in a package):

fucodo:
  seed:
    jobs:
      default:
        enabled: true
        databaseImports:
          100:
            file: 'resource://fucodo.seed/Seed/000-examplefile.sql'

Example from a real package using resource URIs and a follow‑up command:

fucodo:
  seed:
    jobs:
      default:
        enabled: true
        databaseImports:
          100:
            file: 'resource://SBS.SingleSignOn/Seed/100-basic-structure.sql'

Notes:

• Keys (100, 101, 500, …) control the execution order; only their ordering matters.
• A databaseImports.NNN.enabled: false flag can be set to skip an entry temporarily.
• Files can be absolute paths or Flow resource URIs (resource://Vendor.Package/Path/To/File.sql).
• Files can also be urls, if file_get_contents() is allowed to make http requests.
• If a target database already has the table flow_doctrine_migrationstatus, the seeding job will abort early and not import again.

Usage

The package registers a CLI command in Flow. From your distribution root, run:

./flow seed:data

To run a specific job (defaults to default):

./flow seed:data --job yourJobName
# or positional in some shells / Flow versions
./flow seed:data yourJobName

Exit codes:

• 0: success (or skipped because database looked already initialized)
• 1: job disabled (or other StopCommand condition)

During execution you will see which files are imported. Empty files are reported and skipped.

How it works

Internally, SeedCommandController:

• Creates a Doctrine DBAL connection from Neos.Flow.persistence.backendOptions.
• Checks for the presence of the flow_doctrine_migrationstatus table.
• Iterates your configured databaseImports and runs each file with executeStatement.
• Prints a short progress line per step.

Best practices

• Place seed SQL files in your package under Resources/Seed/ and refer to them using the Flow resource:// scheme.
• Keep imports idempotent where possible; the safety check prevents re‑seeding but it is still good to make scripts re‑runnable.
• Use small, ordered files focused on a single concern (schema bootstrap, minimal data, fixtures, …).

Development status

This package is intentionally small. PRs improving documentation and robustness are welcome.

License

MIT (or the same as your project distribution).