loopback-connector-mysql

MySQL is a popular open-source relational database management system (RDBMS). The loopback-connector-mysql module provides the MySQL connector module for the LoopBack framework.

Installation

In your application root directory, enter this command to install the connector:

1npm install loopback-connector-mysql --save

Note: The MySQL connector requires MySQL 5.0+.

This installs the module from npm and adds it as a dependency to the application's package.json file.

If you create a MySQL data source using the data source generator as described below, you don't have to do this, since the generator will run npm install for you.

Creating a MySQL data source

For LoopBack 4 users, use the LoopBack 4 Command-line interface to generate a DataSource with MySQL connector to your LB4 application. Run lb4 datasource, it will prompt for configurations such as host, post, etc. that are required to connect to a MySQL database.

After setting it up, the configuration can be found under src/datasources/<DataSourceName>.datasource.ts, which would look like this:

1const config = {
2  name: 'db',
3  connector: 'mysql',
4  url: '',
5  host: 'localhost',
6  port: 3306,
7  user: 'user',
8  password: 'pass',
9  database: 'testdb',
10};

1"mydb": {
2  "name": "mydb",
3  "connector": "mysql",
4  "host": "myserver",
5  "port": 3306,
6  "database": "mydb",
7  "password": "mypassword",
8  "user": "admin"
9 }

Edit <DataSourceName>.datasources.ts to add any other additional properties that you require.

Properties

NOTE: In addition to these properties, you can use additional parameters supported by node-mysql.

Type mappings

See LoopBack 4 types (or LoopBack 3 types) for details on LoopBack's data types.

LoopBack to MySQL types

MySQL to LoopBack types

NOTE as of v3.0.0 of MySQL Connector, the following flags were introduced:

• treatCHAR1AsString default false - treats CHAR(1) as a String instead of a Boolean
• treatBIT1AsBit default true - treats BIT(1) as a Boolean instead of a Binary
• treatTINYINT1AsTinyInt default true - treats TINYINT(1) as a Boolean instead of a Number

Data mapping properties

Except the common database-specific properties we introduce in How LoopBack Models Map To Database Tables/Collections, the following are more detailed examples and MySQL-specific settings.

Table/Column Names

Besides the basic LoopBack types, as we introduced above, you can also specify additional MySQL-specific properties for a LoopBack model. It would be mapped to the database.

Use the mysql.<property> in the model definition or the property definition to configure the table/column definition.

For example, the following settings would allow you to have custom table name (Custom_User) and column name (custom_id and custom_name). Such mapping is useful when you'd like to have different table/column names from the model:

{% include code-caption.html content="user.model.ts" %}

1@model({
2  settings: { mysql: { schema: 'testdb', table: 'Custom_User'} },
3})
4export class User extends Entity {
5  @property({
6    type: 'number',
7    required: true,
8    id: true,
9    mysql: {
10      columnName: 'custom_id',
11    },
12  })
13  id: number;
14
15  @property({
16    type: 'string',
17    mysql: {
18      columnName: 'custom_name',
19    },
20  })
21  name?: string;

1{
2  "name": "User",
3  "options": {
4    "mysql": {
5      "schema": "testdb",
6      "table": "Custom_User"
7    }
8  },
9  "properties": {
10    "id": {
11      "type": "Number",
12      "required": true,
13      "mysql": {
14        "columnName": "custom_id",
15      }
16    },
17    "name": {
18      "type": "String",
19      "mysql": {
20        "columnName": "custom_name",
21      }
22    },
23  }
24}

Numeric Types

Except the names, you can also use the dataType column/property attribute to specify what MySQL column type to use. The following MySQL type-dataType combinations are supported:

• number
• integer
• tinyint
• smallint
• mediumint
• int
• bigint
• float
• double
• decimal

The following examples will be in LoopBack 4 style, but it's the same if you provide mysql.<property> to the LB3 property definition.

Floating-point types

For Float and Double data types, use the precision and scale options to specify custom precision. Default is (16,8).

1@property({
2  type: 'Number',
3  mysql: {
4    dataType: 'float',
5    precision: 20,
6    scale: 4
7  }
8})
9price: Number;

Fixed-point exact value types

For Decimal and Numeric types, use the precision and scale options to specify custom precision. Default is (9,2). These aren't likely to function as true fixed-point.

1@property({
2  type: 'Number',
3  mysql: {
4    dataType: 'decimal',
5    precision: 12,
6    scale: 8
7  }
8})
9price: Number;

Text types

Convert String / DataSource.Text / DataSource.JSON to the following MySQL types:

• varchar
• char
• text
• mediumtext
• tinytext
• longtext

1@property({
2  type: 'String',
3  mysql: {
4    dataType: 'char',
5    dataLength: 24 // limits the property length
6  },
7})
8userName: String;

Dat types

Convert JSON Date types to datetime or timestamp.

1@property({
2  type: 'Date',
3  mysql: {
4    dataType: 'timestamp',
5  },
6})
7startTime: Date;

Enum

See the Model ENUM property for details.

Default Clause/Constant

Use the default and dataType properties to have MySQL handle setting column DEFAULT value.

1@property({
2  type: 'String',
3  mysql: {
4    dataType: 'varchar',
5    default: 'pending'
6  }
7})
8status: String;
9
10@property({
11  type: 'Number',
12  mysql: {
13    dataType: 'int',
14    default: 42
15  }
16})
17maxDays: Number;
18
19@property({
20  type: 'boolean',
21  mysql: {
22    dataType: 'tinyint',
23    default: 1
24  }
25})
26isDone: Boolean;

For the date or timestamp types use CURRENT_TIMESTAMP or now.

1@property({
2  type: 'Date',
3  mysql: {
4    dataType: 'datetime',
5    default: 'CURRENT_TIMESTAMP'
6  }
7})
8last_modified: Date;

NOTE: The following column types do NOT supported MySQL Default Values:

• BLOB
• TEXT
• GEOMETRY
• JSON

Extended operators

MySQL connector supports the following MySQL-specific operators:

• match Please note extended operators are disabled by default, you must enable them at datasource level or model level by setting allowExtendedOperators to true.

Operator match

The match operator allows you to perform a full text search using the MATCH() .. AGAINST() operator in MySQL.

Three different modes of the MATCH clause are also available in the form of operators -

• matchbool for Boolean Full Text Search
• matchnl for Natural Language Full Text Search
• matchqe for Full-Text Searches with Query Expansion
• matchnlqe for Full-Text Searches with Query Expansion with the IN NATURAL LANGUAGE MODE WITH QUERY EXPANSION modifier.

By default, the match operator works in Natural Language mode.

Note The fields you are querying must be setup with a FULLTEXT index to perform full text search on them. Assuming a model such as this:

1@model({
2  settings: {
3    allowExtendedOperators: true,
4  }
5})
6class Post {
7  @property({
8    type: 'string',
9    mysql: {
10      index: {
11        kind: 'FULLTEXT'
12      }
13    },
14  })
15  content: string;
16}

You can query the content field as follows:

1const posts = await postRepository.find({
2  where: {
3    {
4      content: {match: 'someString'},
5    }
6  }
7});

Discovery and auto-migration

Model discovery

The MySQL connector supports model discovery that enables you to create LoopBack models based on an existing database schema. Once you defined your datasource:

• LoopBack 4 users could use the commend lb4 discover to discover models.
• For LB3 users, please check Discovering models from relational databases. (See database discovery API for related APIs information)

Auto-migration

The MySQL connector also supports auto-migration that enables you to create a database schema from LoopBack models. For example, based on the following model, the auto-migration method would create/alter existing Customer table in the database. Table Customer would have two columns: name and id, where id is also the primary key that has auto_increment set as it has definition of type: 'Number' and generated: true:

1@model()
2export class Customer extends Entity {
3  @property({
4    id: true,
5    type: 'Number',
6    generated: true,
7  })
8  id: number;
9
10  @property({
11    type: 'string',
12  })
13  name: string;
14}

Moreover, additional MySQL-specific properties mentioned in the Data mapping properties section work with auto-migration as well.

Auto-generated ids

For now LoopBack MySQL connector only supports auto-generated id (generated: true) for integer type as for MySQL, the default id type is integer. If you'd like to use other types such as string (uuid) as the id type, you can:

• use uuid that is generated by your LB application by setting defaultFn: uuid.

1  @property({
2    id: true,
3    type: 'string'
4    defaultFn: 'uuidv4',
5    // generated: true,  -> not needed
6  })
7  id: string;

• Alter the table in your database to use a certain function if you prefer having the database to generate the value.

1  @property({
2    id: true,
3    type: 'string'
4    generated: true,  // to indicate the value generates by the db
5    useDefaultIdType: false,  // needed
6  })
7  id: string;

Auto-migrate/Auto-update models with foreign keys

Foreign key constraints can be defined in the model definition.

Note: The order of table creation is important. A referenced table must exist before creating a foreign key constraint. The order can be specified using the optional SchemaMigrationOptions argument of migrateSchema:

await app.migrateSchema({
	models: [ 'Customer', 'Order' ]
});

Define your models and the foreign key constraints as follows:

{% include code-caption.html content="customer.model.ts" %}

1@model()
2export class Customer extends Entity {
3  @property({
4    id: true,
5    type: 'Number',
6    generated: true,
7  })
8  id: number;
9
10  @property({
11    type: 'string',
12  })
13  name: string;
14}

order.model.ts:

1@model({
2  settings: {
3    foreignKeys: {
4      fk_order_customerId: {
5        name: 'fk_order_customerId',
6        entity: 'Customer',
7        entityKey: 'id',
8        foreignKey: 'customerId',
9      },
10    },
11  })
12export class Order extends Entity {
13  @property({
14    id: true,
15    type: 'Number',
16    generated: true
17  })
18  id: number;
19
20  @property({
21    type: 'string'
22  })
23  name: string;
24
25  @property({
26    type: 'Number'
27  })
28  customerId: number;
29}

1({
2  "name": "Customer",
3  "options": {
4    "idInjection": false
5  },
6  "properties": {
7    "id": {
8      "type": "Number",
9      "id": 1
10    },
11    "name": {
12      "type": "String",
13      "required": false
14    }
15  }
16},
17{
18  "name": "Order",
19  "options": {
20    "idInjection": false,
21    "foreignKeys": {
22      "fk_order_customerId": {
23        "name": "fk_order_customerId",
24        "entity": "Customer",
25        "entityKey": "id",
26        "foreignKey": "customerId"
27      }
28    }
29  },
30  "properties": {
31    "id": {
32      "type": "Number"
33      "id": 1
34    },
35    "customerId": {
36      "type": "Number"
37    },
38    "description": {
39      "type": "String",
40      "required": false
41    }
42  }
43})

MySQL handles the foreign key integrity by the referential action specified by ON UPDATE and ON DELETE. You can specify which referential actions the foreign key follows in the model definition upon auto-migrate or auto-update operation. Both onDelete and onUpdate default to restrict.

Take the example we showed above, let's add the referential action to the foreign key customerId:

1@model({
2  settings: {
3    foreignKeys: {
4      fk_order_customerId: {
5        name: 'fk_order_customerId',
6        entity: 'Customer',
7        entityKey: 'id',
8        foreignKey: 'customerId',
9        onUpdate: 'restrict', // restrict|cascade|set null|no action|set default
10        onDelete: 'cascade'   // restrict|cascade|set null|no action|set default
11      },
12    },
13  })
14export class Order extends Entity {
15...

1{
2  "name": "Customer",
3  "options": {
4    "idInjection": false
5  },
6  "properties": {
7    "id": {
8      "type": "Number",
9      "id": 1
10    },
11    "name": {
12      "type": "String",
13      "required": false
14    }
15  }
16},
17{
18  "name": "Order",
19  "options": {
20    "idInjection": false,
21    "foreignKeys": {
22      "fk_order_customerId": {
23        "name": "fk_order_customerId",
24        "entity": "Customer",
25        "entityKey": "id",
26        "foreignKey": "customerId",
27        "onUpdate": "restrict",
28        "onDelete": "cascade"
29      }
30    }
31  },
32  "properties": {
33    "id": {
34      "type": "Number"
35      "id": 1
36    },
37    "customerId": {
38      "type": "Number"
39    },
40    "description": {
41      "type": "String",
42      "required": false
43    }
44  }
45}

1module.exports = function (app) {
2  var mysqlDs = app.dataSources.mysqlDS;
3  var Book = app.models.Order;
4  var Author = app.models.Customer;
5
6  // first autoupdate the `Customer` model to avoid foreign key constraint failure
7  mysqlDs.autoupdate('Customer', function (err) {
8    if (err) throw err;
9    console.log('\nAutoupdated table `Customer`.');
10
11    mysqlDs.autoupdate('Order', function (err) {
12      if (err) throw err;
13      console.log('\nAutoupdated table `Order`.');
14      // at this point the database table `Order` should have one foreign key `customerId` integrated
15    });
16  });
17};

Breaking Changes with GeoPoint since 5.x

Prior to loopback-connector-mysql@5.x, MySQL connector was saving and loading GeoPoint properties from the MySQL database in reverse. MySQL expects values to be POINT(X, Y) or POINT(lng, lat), but the connector was saving them in the opposite order(i.e. POINT(lat,lng)).

Use the geopoint type to achieve so:

1  @property({
2    type: 'geopoint'
3  })
4  name: GeoPoint;

If you have an application with a model that has a GeoPoint property using previous versions of this connector, you can migrate your models using the following programmatic approach:

1'use strict';
2module.exports = function (app) {
3  function findAndUpdate() {
4    var teashop = app.models.teashop;
5    //find all instances of the model we'd like to migrate
6    teashop.find({}, function (err, teashops) {
7      teashops.forEach(function (teashopInstance) {
8        //what we fetch back from the db is wrong, so need to revert it here
9        var newLocation = {
10          lng: teashopInstance.location.lat,
11          lat: teashopInstance.location.lng,
12        };
13        //only update the GeoPoint property for the model
14        teashopInstance.updateAttribute('location', newLocation, function (
15          err,
16          inst,
17        ) {
18          if (err) console.log('update attribute failed', err);
19          else console.log('updateAttribute successful');
20        });
21      });
22    });
23  }
24
25  findAndUpdate();
26};

1{
2  "name": "teashop",
3  "base": "PersistedModel",
4  "idInjection": true,
5  "options": {
6    "validateUpsert": true
7  },
8  "properties": {
9    "name": {
10      "type": "string",
11      "default": "storename"
12    },
13    "location": {
14      "type": "geopoint"
15    }
16  },
17  "validations": [],
18  "relations": {},
19  "acls": [],
20  "methods": {}
21}

Running tests

Own instance

If you have a local or remote MySQL instance and would like to use that to run the test suite, use the following command:

• Linux

1MYSQL_HOST=<HOST> MYSQL_PORT=<PORT> MYSQL_USER=<USER> MYSQL_PASSWORD=<PASSWORD> MYSQL_DATABASE=<DATABASE> CI=true npm test

• Windows

1SET MYSQL_HOST=<HOST> SET MYSQL_PORT=<PORT> SET MYSQL_USER=<USER> SET MYSQL_PASSWORD=<PASSWORD> SET MYSQL_DATABASE=<DATABASE> SET CI=true npm test

Docker

If you do not have a local MySQL instance, you can also run the test suite with very minimal requirements.

• Assuming you have Docker installed, run the following script which would spawn a MySQL instance on your local:

1source setup.sh <HOST> <PORT> <USER> <PASSWORD> <DATABASE>

where <HOST>, <PORT>, <USER>, <PASSWORD> and <DATABASE> are optional parameters. The default values are localhost, 3306, root, pass and testdb respectively.

• Run the test:

1npm test