pongee/database-schema-visualization

Database schema visualization

5.1.1 2026-07-28 15:49 UTC

Requires

• php: >=8.5
• ext-json: *
• symfony/console: ^7.0||^8.0
• twig/twig: ^3.0

Requires (Dev)

• phpstan/phpstan: ^2.0
• phpunit/phpunit: ^13.0
• rector/rector: ^2.0
• roave/security-advisories: dev-latest
• slevomat/coding-standard: ^8.0
• squizlabs/php_codesniffer: ^4.0

MIT

• database
• mysql
• schema
• cassandra
• docker
• visualization
• plantuml

Published on Packagist.org on 2026-07-28 15:49 UTC

README

Project goal

The aim of this project is to generate database documentation from sql schema.

Supported databases

• MySQL
• MariaDB
• Apache Cassandra (Basics)

Supported Output formats

• PNG, SVG image
• Plantuml raw text
• Json
• Markdown

Installation

$ composer require pongee/database-schema-visualization

Commands

Every command reads a schema file and writes the result to stdout, so redirect it into a file. Foreign keys are resolved automatically from the schema.

Each command exists for both MySQL (mysql:) and Apache Cassandra (cassandra:):

Command	Output
mysql:json / cassandra:json	JSON
mysql:plantuml / cassandra:plantuml	PlantUML raw text
mysql:markdown / cassandra:markdown	Markdown
mysql:image / cassandra:image	PNG / SVG image

Argument and options:

Name	Description
file	Path to the schema file. Required.
--type	Image format for the image commands: png (default) or svg.
--template	Twig template used for rendering. Defaults to the bundled template.

Usage

The command, argument and options are the same across every run mode below — see Commands for the full list.

Command line

$ php ./database-schema-visualization <command> [options] <file> > output

For example:

$ php ./database-schema-visualization mysql:image --type png ./example/schema/sakila.sql > ./example/output/sakila/sakila.png

Output:

Docker

The image is published to Docker Hub, built for both linux/amd64 and linux/arm64.

$ docker pull pongeepublic/database-schema-visualization:latest

Mount your schema into the container and pass the same command and options as on the CLI:

$ docker run --rm -v "$PWD/schema.sql:/app/schema.sql" \
    pongeepublic/database-schema-visualization mysql:image --type png schema.sql > diagram.png

List the available commands:

$ docker run --rm pongeepublic/database-schema-visualization list

PHP

Png export

<?php declare(strict_types=1);

use Pongee\DatabaseSchemaVisualization\DataObject\Sql\Database\Connection\ConnectionCollection;
use Pongee\DatabaseSchemaVisualization\Export\Plantuml;
use Pongee\DatabaseSchemaVisualization\Generator\ImageGenerator;
use Pongee\DatabaseSchemaVisualization\Generator\ImageType;
use Pongee\DatabaseSchemaVisualization\Parser\MysqlParser;

include __DIR__ . '/../../vendor/autoload.php';

$sqlSchema = '
  CREATE TABLE IF NOT EXISTS `foo` (
    `id` INT(10) UNSIGNED NOT NULL COMMENT "The id"
   ) ENGINE=innodb DEFAULT CHARSET=utf8;
';

$parser                     = new MysqlParser(); // or CassandraParser(), MariadbParser(); 
$plantumlExport             = new Plantuml(file_get_contents(__DIR__ . '/../../src/Template/Plantuml/v1.twig'));
$forcedConnectionCollection = new ConnectionCollection();
$imageGenerator             = new ImageGenerator(
    ImageType::Png,
    __DIR__ . '/../../bin/plantuml-mit-1.2026.6.jar'
);

$schema = $parser->run($sqlSchema, $forcedConnectionCollection);

print $imageGenerator->generate($plantumlExport->export($schema));

Json export

<?php declare(strict_types=1);

use Pongee\DatabaseSchemaVisualization\DataObject\Sql\Database\Connection\ConnectionCollection;
use Pongee\DatabaseSchemaVisualization\Export\Json;
use Pongee\DatabaseSchemaVisualization\Parser\MysqlParser;

include './vendor/autoload.php';

$sqlSchema = '
  CREATE TABLE IF NOT EXISTS `foo` (
    `id` INT(10) UNSIGNED NOT NULL COMMENT "The id"
   ) ENGINE=innodb DEFAULT CHARSET=utf8;
';

$parser                     = new MysqlParser();
$jsonExport                 = new Json();
$forcedConnectionCollection = new ConnectionCollection();

$schema = $parser->run($sqlSchema, $forcedConnectionCollection);

print $jsonExport->export($schema);

This will generate:

{
    "tables": {
        "foo": {
            "columns": [
                {
                    "name": "id",
                    "type": "INT",
                    "typeParameters": [
                        "10"
                    ],
                    "otherParameters": "UNSIGNED NOT NULL",
                    "comment": "The id"
                }
            ],
            "indexs": {
                "simple": [],
                "spatial": [],
                "fulltext": [],
                "unique": []
            },
            "primaryKey": []
        }
    },
    "connections": []
}
    

Markdown export

<?php declare(strict_types=1);

use Pongee\DatabaseSchemaVisualization\DataObject\Sql\Database\Connection\ConnectionCollection;
use Pongee\DatabaseSchemaVisualization\Export\Markdown;
use Pongee\DatabaseSchemaVisualization\Parser\MysqlParser;

include './vendor/autoload.php';

$sqlSchema = '
  CREATE TABLE IF NOT EXISTS `foo` (
    `id` INT(10) UNSIGNED NOT NULL COMMENT \'The id\'
   ) ENGINE=innodb DEFAULT CHARSET=utf8;
';

$parser                     = new MysqlParser();
$markdownExport             = new Markdown(file_get_contents('./src/Template/Markdown/v1.twig'));
$forcedConnectionCollection = new ConnectionCollection();

$schema = $parser->run($sqlSchema, $forcedConnectionCollection);

print $markdownExport->export($schema);

This will generate:

Schema

foo table

Columns

Name	Type	Parameters	Comment
id	INT ( 10 )	UNSIGNED NOT NULL	The id