Mysql Dump

Create a backup of a MySQL database.

Installation

$ npm install mysqldump

If you're using this package in typescript, you should also

$ npm install @types/node

Usage

import mysqldump from 'mysqldump';
// or const mysqldump = require('mysqldump')

// dump the result straight to a file
mysqldump({
    connection: {
        host: 'localhost',
        user: 'root',
        password: '123456',
        database: 'my_database',
    },
    dumpToFile: './dump.sql',
});

// dump the result straight to a compressed file
mysqldump({
    connection: {
        host: 'localhost',
        user: 'root',
        password: '123456',
        database: 'my_database',
    },
    dumpToFile: './dump.sql.gz',
    compressFile: true,
});

// return the dump from the function and not to a file
const result = await mysqldump({
    connection: {
        host: 'localhost',
        user: 'root',
        password: '123456',
        database: 'my_database',
    },
});

Result

The returned result contains the dump property, which is split into schema and data.

export default interface DumpReturn {
    /**
     * The result of the dump
     */
    dump : {
        /**
         * The concatenated SQL schema dump for the entire database.
         * Null if configured not to dump.
         */
        schema : string | null
        /**
         * The concatenated SQL data dump for the entire database.
         * Null if configured not to dump.
         */
        data : string | null
        /**
         * The concatenated SQL trigger dump for the entire database.
         * Null if configured not to dump.
         */
        trigger : string | null
    }
    tables : Table[]
}

Options

All the below options are documented in the typescript declaration file:

export interface ConnectionOptions {
    /**
     * The database host to connect to.
     * Defaults to 'localhost'.
     */
    host?: string;
    /**
     * The port on the host to connect to.
     * Defaults to 3306.
     */
    port?: number;
    /**
     * The database to dump.
     */
    database: string;
    /**
     * The DB username to use to connect.
     */
    user: string;
    /**
     * The password to use to connect.
     */
    password: string;
    /**
     * The charset to use for the connection.
     * Defaults to 'UTF8_GENERAL_CI'.
     */
    charset?: string;
    /**
     * SSL configuration options.
     * Passing 'Amazon RDS' will use Amazon's RDS CA certificate.
     *
     * Otherwise you can pass the options which get passed to tls.createSecureContext.
     * See: https://nodejs.org/api/tls.html#tls_tls_createsecurecontext_options
     */
    ssl?: 'Amazon RDS' | null | {
        /**
         * Optionally override the trusted CA certificates. Default is to trust the well-known CAs curated by Mozilla.
         */
        ca?: string | Buffer;
        /**
         * Optional cert chains in PEM format.
         */
        cert?: string | Buffer;
        /**
         * Optional cipher suite specification, replacing the default.
         */
        ciphers?: string;
        /**
         * Optional PEM formatted CRLs (Certificate Revocation Lists).
         */
        crl?: string | Array<string>;
        /**
         * Attempt to use the server's cipher suite preferences instead of the client's.
         */
        honorCipherOrder?: boolean;
        /**
         * Optional private keys in PEM format.
         */
        key?: string | Buffer;
        /**
         * Optional shared passphrase used for a single private key and/or a PFX.
         */
        passphrase?: string;
        /**
         * Optional PFX or PKCS12 encoded private key and certificate chain.
         */
        pfx?: string | Buffer;
        /**
         * DO NOT USE THIS OPTION UNLESS YOU REALLY KNOW WHAT YOU ARE DOING!!!
         * Set to false to allow connection to a MySQL server without properly providing the appropraite CA to trust.
         */
        rejectUnauthorized?: boolean;
    };
}
export interface SchemaDumpOptions {
    /**
     * True to include autoincrement values in schema, false otherwise.
     * Defaults to true.
     */
    autoIncrement?: boolean;
    /**
     * True to include engine values in schema, false otherwise.
     * Defaults to true.
     */
    engine?: boolean;
    /**
     * True to run a sql formatter over the output, false otherwise.
     * Defaults to true.
     */
    format?: boolean;
    /**
     * Options for table dumps
     */
    table?: {
        /**
         * Guard create table calls with an "IF NOT EXIST"
         * Defaults to true.
         */
        ifNotExist?: boolean;
        /**
         * Drop tables before creation (overrides `ifNotExist`).
         * Defaults to false.
         */
        dropIfExist?: boolean;
        /**
         * Include the `DEFAULT CHARSET = x` at the end of the table definition
         * Set to true to include the value form the DB.
         * Set to false to exclude it altogether.
         * Set to a string to explicitly set the charset.
         * Defaults to true.
         */
        charset?: boolean | string;
    };
    view?: {
        /**
         * Uses `CREATE OR REPLACE` to define views.
         * Defaults to true.
         */
        createOrReplace?: boolean;
        /**
         * Include the `DEFINER = {\`user\`@\`host\` | CURRENT_USER}` in the view definition or not
         * Defaults to false.
         */
        definer?: boolean;
        /**
         * Include the `ALGORITHM = {UNDEFINED | MERGE | TEMPTABLE}` in the view definition or not
         * Defaults to false.
         */
        algorithm?: boolean;
        /**
         * Incldue the `SQL SECURITY {DEFINER | INVOKER}` in the view definition or not
         * Defaults to false.
         */
        sqlSecurity?: boolean;
    };
}
export interface TriggerDumpOptions {
    /**
     * The temporary delimiter to use between statements.
     * Set to false to not use delmiters
     * Defaults to ';;'.
     */
    delimiter?: string | false;
    /**
     * Drop triggers before creation.
     * Defaults to false.
     */
    dropIfExist?: boolean;
    /**
     * Include the `DEFINER = {\`user\`@\`host\` | CURRENT_USER}` in the view definition or not
     * Defaults to false.
     */
    definer?: boolean;
}
export interface DataDumpOptions {
    /**
     * True to run a sql formatter over the output, false otherwise.
     * Defaults to true.
     */
    format?: boolean;
    /**
     * Include file headers in output
     * Defaults to true.
     */
    verbose?: boolean;
    /**
     * Use a read lock during the data dump (see: https://dev.mysql.com/doc/refman/5.7/en/replication-solutions-backups-read-only.html)
     * Defaults to false.
     */
    lockTables?: boolean;
    /**
     * Dump data from views.
     * Defaults to false.
     */
    includeViewData?: boolean;
    /**
     * Maximum number of rows to include in each multi-line insert statement
     * Defaults to 1 (i.e. new statement per row).
     */
    maxRowsPerInsertStatement?: number;
    /**
     * True to return the data in a function, false to not.
     * This is useful in databases with a lot of data.
     *
     * We stream data from the DB to reduce the memory footprint.
     * However note that if you want the result returned from the function,
     * this will result in a larger memory footprint as the string has to be stored in memory.
     *
     * Defaults to false if dumpToFile is truthy, or true if not dumpToFile is falsey.
     */
    returnFromFunction?: boolean;
    /**
     * A map of tables to additional where strings to add.
     * Use this to limit the number of data that is dumped.
     * Defaults to no limits
     */
    where?: {
        [k: string]: string;
    };
}
export interface DumpOptions {
    /**
     * The list of tables that you want to dump.
     * Defaults to all tables (signalled by passing an empty array).
     */
    tables?: Array<string>;
    /**
     * True to use the `tables` options as a blacklist, false to use it as a whitelist.
     * Defaults to false.
     */
    excludeTables?: boolean;
    /**
     * Explicitly set to false to not include the schema in the dump.
     * Defaults to including the schema.
     */
    schema?: false | SchemaDumpOptions;
    /**
     * Explicitly set to false to not include data in the dump.
     * Defaults to including the data.
     */
    data?: false | DataDumpOptions;
    /**
     * Explicitly set to false to not include triggers in the dump.
     * Defaults to including the triggers.
     */
    trigger?: false | TriggerDumpOptions;
}
export interface Options {
    /**
     * Database connection options
     */
    connection: ConnectionOptions;
    /**
     * Dump configuration options
     */
    dump?: DumpOptions;
    /**
     * Set to a path to dump to a file.
     * Exclude to just return the string.
     */
    dumpToFile?: string | null;
    /**
     * Should the output file be compressed (gzip)?
     * Defaults to false.
     */
    compressFile?: boolean;
}
export interface ColumnList {
    /**
     * Key is the name of the column
     */
    [k: string]: {
        /**
         * The type of the column as reported by the underlying DB.
         */
        type: string;
        /**
         * True if the column is nullable, false otherwise.
         */
        nullable: boolean;
    };
}
export interface Table {
    /**
     * The name of the table.
     */
    name: string;
    /**
     * The raw SQL schema dump for the table.
     * Null if configured to not dump.
     */
    schema: string | null;
    /**
     * The raw SQL data dump for the table.
     * Null if configured to not dump.
     */
    data: string | null;
    /**
     * The list of column definitions for the table.
     */
    columns: ColumnList;
    /**
     * An ordered list of columns (for consistently outputing as per the DB definition)
     */
    columnsOrdered: Array<string>;
    /**
     * True if the table is actually a view, false otherwise.
     */
    isView: boolean;
    /**
     * A list of triggers attached to the table
     */
    triggers: Array<string>;
}
export interface DumpReturn {
    /**
     * The result of the dump
     */
    dump: {
        /**
         * The concatenated SQL schema dump for the entire database.
         * Null if configured not to dump.
         */
        schema: string | null;
        /**
         * The concatenated SQL data dump for the entire database.
         * Null if configured not to dump.
         */
        data: string | null;
        /**
         * The concatenated SQL trigger dump for the entire database.
         * Null if configured not to dump.
         */
        trigger: string | null;
    };
    tables: Array<Table>;
}
export default function main(inputOptions: Options): Promise<DumpReturn>;

export as namespace mysqldump;

The MIT License

Contributing

Local Installation

Make sure to first install all the required development dependencies:

yarn
// or
npm install .

Linting

We use eslint in conjunction with typescript-eslint-parser for code linting.

PRs are required to pass the linting with no errors and preferrably no warnings.

Testing

Tests can be run via the test script - yarn test / npm test.

Additionally it's required that you do a build and run your test against the public package to ensure the build doesn't cause regressions - yarn run test-prod / npm run test-prod.

PRs are required to maintain the 100% test coverage, and all tests must pass successfully.

mysqldump versions

mysqldump dependencies