Skip to content

Latest commit

 

History

History
1022 lines (815 loc) · 26.1 KB

README.md

File metadata and controls

1022 lines (815 loc) · 26.1 KB
Raw

sharp-db

Build Status Code Coverage ISC License

Classes for running SQL and building select queries for MySQL in Node

Installation

npm install sharp-db

Table of Contents

Db

Connection

Connection can be configured with environmental variables or in the constructor.

Option ENV name Default
host DB_HOST 127.0.0.1
port DB_PORT 3306
user DB_USER root
password DB_PASSWORD empty string
database DB_DATABASE undefined
charset DB_CHARSET utf8mb4

See node's mysqljs for other options.

Instantiation

Connect to MySQL server

const { Db } = require('sharp-db');
// read options from ENV
const db1 = Db.factory();

// specify options in constructor
const db2 = new Db({
    host: '127.0.0.1',
    user: 'root',
    password: '',
    port: 3306,
});

// instance that was last created
const db2Again = Db.factory();

// Don't forget to close the connection when done
db1.end();
db2.end();

Auto factory and end

You can use await Db.withInstance(db => /* do stuff with db */) to get an instance, do something, and then close the connection.

const { Db } = require('sharp-db');

// read options from ENV, instantiate and call db.end() automatically
const { error, domain } = await Db.withInstance(async db => {
    const sql = 'SELECT email FROM users WHERE id = 5';
    const { results: email } = await db.selectValue(sql);
    return {
        domain: email.split('@').pop(),
    };
});

Db.withInstance() will return one of the following:

  1. { error } where error is an object from mysql2. Full docs. Summary:

    • error.sqlMessage The textual description of the error
    • error.code The string error code such as PROTOCOL_CONNECTION_LOST
    • error.errno The associated number code
    • error.fatal True if the error caused the connection to close
    • error.sql The full SQL of the failed query
    • error.sqlState The five-character SQLSTATE code

  2. Whatever value is returned from the handler. We suggest always returning an object, expecting the caller to check .error.

WARNING: Your handler function must return a promise that resolves AFTER your query has returned a result. Failing to do so will result in db.end() being called before your query is run. You may see an Error similar to the following:

[ERROR] -1 (N/A): Can't add new command when connection is in closed state

For example:

// WILL FAIL:
const { error } = Db.withInstance(db => {
    db.insertInto('users', user);
});

// WILL SUCCEED:
const { error, results } = Db.withInstance(db => {
    return db.insertInto('users', user);
});
// ALSO OK:
const { error, newUserId } = Db.withInstance(async db => {
	const { insertId } = await db.insertInto('users', user);
	return {
		newUserId: insertId,
	};
});

SSH Tunneling

Connect to MySQL through an SSH tunnel

const { Db } = require('sharp-db');
const db = Db.factory({
    // MySQL connection as first argument
    host: '127.0.0.1',
    port: 3306,
    user: 'root',
    password: '',
}, {
    // SSH connection as second argument
    host: 'example.com',
    port: 22,
    user: 'ubuntu',
    privateKey: '~/.ssh/example.com.pem',
});

SSH Tunnel Options

Option ENV name Default
host DB_SSH_HOST "localhost"
port DB_SSH_PORT 22
user DB_SSH_USER none
password DB_SSH_PASSWORD none
privateKey DB_SSH_PRIVATE_KEY none
localPort DB_SSH_LOCAL_PORT 12346

See all options in ssh2's npm package.

Basic use

All code examples below assume the Db instance has been stored in db.

Plain select queries

const { Db } = require('sharp-db');
const db = Db.factory();
const { query, results, fields } = await db.select('SELECT * FROM users');
// query is the final query executed after value binding
// results is an Array of objects representing the query results
// fields is an Array of objects representing the columns that were returned

// Don't forget to close the connection when done
db.destroy();

Relevant properties of each fields item:

Item Description Example
characterSet Character set constant 45
encoding Character set name utf8
name Name of column my_column
columnLength Number of bytes of field 400
columnType Data type constant 253
flags Field flag constant 33

Bindings

Question-mark and colon-prefixed bindings are supported.

Binding with Question Marks

const sql = 'SELECT * FROM users WHERE is_active = ? AND department_id = ?';
const { results: users } = await db.select(sql, true, 5);

Named Bindings

const sql = 'SELECT * FROM users WHERE is_active = :isActive AND department_id = :departmentId';
const { results: users } = await db.select(sql, {
    isActive: true,
    departmentId: 5,
});

Binding data types

const { results: users } = await db.select(sql, {
    isActive: true,          // Boolean
    departmentId: 5,         // Number
    createdAt: '2020-02-14', // Strings
    statusCode: [1, 2, 3],   // Arrays e.g. IN(1, 2, 3)
    deletedAt: null,         // null e.g. NULL
});

Methods

selectFirst(sql, ...bindValues)

Get only the first row.

const { results: row } = await db.selectFirst(sql);

Example results: { id: 1, name: "John" }

selectValue(sql, ...bindValues)

Get only the first column of the first row.

const { results: value } = await db.selectValue(sql);

Example results: "John"

selectHash(sql, ...bindValues)

Get an Object with column-value pairs.

const { results: hash } = await db.selectHash(sql);

Example results: { "1": "John", "2": "Jane" }

selectList(sql, ...bindValues)

Get an Array of values for the first column of the first row.

const { results: list } = await db.selectList(sql);

Example results: ["John", "Jane"]

selectExists(sql, ...bindValues)

Return true if query returns any rows.

const { results: doesExist } = await db.selectExists(sql);

Example results: true

selectIndexed(indexColumn, sql, ...bindValues)

Return an Object where every result row is indexed by the given field.

const { results: usersById } = await db.selectIndexed('id', sql);

Example results:

results = {
  "1": { id: 1, name: "John" },
  "2": { id: 2, name: "Jane" },
}

selectGrouped(groupColumn, sql, ...bindValues)

Return an Object where every result row is indexed by the given field.

const { results: usersGroupedByOrg } = await db.selectGrouped('org', sql);

Example results:

results = {
    "Marketing": [
        { id: 1, name: "John", org: "Marketing" },
        { id: 2, name: "Jane", org: "Marketing" },
    ],
    "Finance": [
        { id: 3, name: "Jose", org: "Finance" },
    ],
}

selectOrCreate(table, criteria[, newValues])

Select a record or create a new record. Good when normalizing data that is frequently referenced.

For example, say I have a table hits with a column url_id and a table urls with columns id and url.

I want to add a new hit record with a given URL. You might write this:

const newHit = {
	date: '2021-10-15 17:43:24',
	url: 'https://example.com',
}

const { results } = await db.selectOrCreate('urls', { url: newHit.url });

await db.insert('hits', {
	date: newHit.date,
	url_id: results.id,
});

Useful Query Options

SQL can actually be an Object with options.

const options = {
    sql: `
        SELECT users.*, avatars.*
        FROM users
        INNER JOIN avatars ON avatars.user_id = users.id
        WHERE users.is_active = ?
    `,
    // kill query if not completed within 30 seconds
    timeout: 30000,
    // return records with keys `users` and `avatars` with their own fields nested underneath
    nestTables: true,
    // you can also bind values here using question marks
    values: [true],
};
const { results } = await db.select(options);

nestTables Example

Given a query of:

SELECT users.*, avatars.*
FROM users
INNER JOIN avatars ON avatars.user_id = users.id
WHERE users.is_active = ?

nesting tables will return a data structure such as:

results = [
	{
		users: {
			id: 1,
			name: 'John Doe',
			is_active: true,
		},
		avatars: {
			id: 101,
			user_id: 1,
			url: 'http://example.com/john.png'
		}
	},
	{
		users: {
			id: 2,
			name: 'Jane Doe',
			is_active: true,
		},
		avatars: {
			id: 102,
			user_id: 2,
			url: 'http://example.com/jane.png'
		}
	}
]

selectFrom(table, fields, values)

Build and run a simple select statement.

const { results } = await db.selectFrom('users', ['fname','lname'], {
    'id >': 5,
    is_active: true,
    department_id: [1,2],
});

insert(sql, ...bindVars)

Run an insert statement; return the id of the new record if applicable.

const { insertId } = await db.insert("INSERT INTO users SET name='John', email='[email protected]'");

insertInto(table, values)

Build and run an insert statement; return the id of the new record if applicable.

const { insertId } = await db.insertInto('users', {
    name: 'John',
    email: '[email protected]',
});

insertExtended(table, rows)

Build and run an extended insert statement; return the id of the last record if applicable.

const { insertId } = await db.insertExtended('users', [
    { name: 'John', email: '[email protected]' },
    { name: 'Jane', email: '[email protected]' },
]);

insertIntoOnDuplicateKeyUpdate(table, insert, update)

Build and run an insert statement; return the id of the new record if applicable.

const { insertId, affectedRows } = await db.insertIntoOnDuplicateKeyUpdate(
    'users',
    {
        sso_ref: 'A123456',
        name: 'Jane Doe',
        created_at: '2020-02-02',
    },
    {
        name: 'Jane Doe Carter',
        modified_at: '2020-02-02',
    }
);

update(sql, ...bindValues)

Run an update statement; return the number of affected rows.

const { affectedRows } = await db.update(
    "UPDATE users SET name = ? WHERE id = ?",
    'Jane Doe Carter',
    5
);

updateTable(table, set, where)

Build and run an update statement; return the number of affected rows.

const { affectedRows } = await db.updateTable(
    'users',
    { name: 'Jane Doe Carter' },
    { id: 5 }
);

delete(sql, ...bindValues)

Run a delete statement; return the number of affected rows.

const { affectedRows } = await db.delete(
    "DELETE FROM users WHERE id = ? LIMIT 1",
    5
);

deleteFrom(table, where, limit)

Build and run a delete statement; return the number of affected rows.

const { affectedRows } = await db.deleteFrom('users', { id: 5 }, 1);

query(sql, ...bindValues)

Run any type of statement.

const { query, results, fields } = await db.query(
    'SELECT * FROM users'
);

multiQuery(sql, ...bindValues)

Run multiple statements delimited by semicolon.

const { query, results, fields } = await db.query(
    'SELECT * FROM users; SELECT * FROM tags'
);

Solutions to Common Problems

Connection is in closed state

Error: Can't add new command when connection is in closed state

Make sure you use await your results before closing your connection.

ECONNRESET or error event

Error: read ECONNRESET or Emitted 'error' event on Client instance

Your SSH connection may have timed out. To keep connection alive, you can send keepalive packets.

const sshConfig = {
	// ...
	// How often (in milliseconds) to send SSH-level keepalive packets to the server (in a similar way as OpenSSH's ServerAliveInterval config option). Set to 0 to disable. Default: 0
	keepaliveInterval: 30,
	// How many consecutive, unanswered SSH-level keepalive packets that can be sent to the server before disconnection (similar to OpenSSH's ServerAliveCountMax config option). Default: 3
	keepaliveCountMax: 120,
}
const db = new Db(mysqlConfig, sshConfig);

Select

A Select object represents a SQL SELECT query and allows dynamically adding clauses including JOIN, WHERE, ORDER BY, LIMIT, OFFSET.

Select.parse()

The easiest way to define a base query is to use Select.parse(sql) and then add criteria as needed.

const { Select } = require('sharp-db');
const query = Select.parse(`
    SELECT u.id, u.fname, u.lname, u.email, p.phone
    FROM users
    LEFT JOIN phone_numbers p ON p.user_id = u.id
      AND p.type = 'main'
    WHERE u.is_active = 1
`);
if (email) {
    query.where('u.email', email);
}
if (areaCode) {
    query.where('p.phone', 'LIKE ?%', areaCode);
}
query.sort(sortField);
query.limit(limitTo);

You can also define binding in the base query itself.

const query = Select.parse(`
    SELECT u.id, u.fname, u.lname, u.email, a.city, a.zip
    FROM users
    LEFT JOIN addresses a ON a.user_id = u.id
    WHERE a.state = :state
`);
query.bind('state', state);

And you can bind multiple values at once.

const query = Select.parse(`
    SELECT u.id, u.fname, u.lname, u.email, a.city, a.zip
    FROM users
    LEFT JOIN addresses a ON a.user_id = u.id
    WHERE a.state = :state
      AND a.city IN (:city)
`);
query.bind({ state, city });

Building the Query

The following are the most common methods for building queries.

  • query.columns(columnNames) - Add column names to fetch
  • query.column(columnName) - Add a column name to fetch
  • query.table(tableName) - Specify the table in the FROM clause
  • query.from(tableName) - Same as above
  • query.innerJoin(expression) - Add an INNER JOIN expression
  • query.leftJoin(expression) - Add a LEFT JOIN expression
  • query.fullJoin(expression) - Add a FULL JOIN expression
  • query.rightJoin(expression) - Add a RIGHT JOIN expression
  • query.crossJoin(expression) - Add a CROSS JOIN expression
  • query.leftOuterJoin(expression) - Add a LEFT OUTER JOIN expression
  • query.fullOuterJoin(expression) - Add a FULL OUTER JOIN expression
  • query.rightOuterJoin(expression) - Add a RIGHT OUTER JOIN expression
  • query.groupBy(column) - Group by a column or expression
  • query.where(column, operator, value) - Require column satisfy operator
  • query.where(column, value) - Require column equal a value
  • query.where(expression) - Add an arbitrary WHERE expression
  • query.where(columnValuePairs) - Add multiple conditions
  • query.whereBetween(column, twoValueArray) - Require value BETWEEN, < or >
  • query.orWhere(conditions) - Specify multiple where()s joined by OR
  • query.having(column, operator, value) - Having column satisfy operator
  • query.having(column, value) - Having column equal value
  • query.having(column, value) - Having column equal value
  • query.having(expression) - Having an arbitrary expression
  • query.orHaving(expressions) - Multiple having()s joined by OR
  • query.orderBy(column) - Add ORDER BY clause
  • query.sortField(column, mapNames) - Add ORDER BY clause with mapNames
  • query.limit(num) - Limit by the given number
  • query.offset(num) - Specify an offset
  • query.page(num) - Automatically calculate offset based on limit and page

Fetching Data

The methods to fetch data mirror those of Db.

  • query.fetch() - equivalent to db.select()
  • query.fetchFirst() - equivalent to db.selectFirst()
  • query.fetchHashed() - equivalent to db.selectHashed()
  • query.fetchList() - equivalent to db.selectList()
  • query.fetchValue() - equivalent to db.selectValue()
  • query.fetchIndexed(byField) - equivalent to db.selectIndexed(byField)
  • query.fetchGrouped(byField) - equivalent to db.selectGrouped(byField)

Counting Results

One powerful feature of Select is that it can construct a count query to fetch the number of results that would have been returned if there were no LIMIT.

const query = Select.parse('SELECT id, name FROM users LIMIT 5');
const { results: users } = await query.fetch();
const { results: count } = await query.foundRows();
// will run the following query:
// SELECT COUNT(*) AS foundRows FROM users

Specifying the Db Instance to Use

There are three ways to specify the Db instance to fetch data with:

  1. query = Select.parse(sql, db)
  2. query = new Select(db)
  3. query.db = db

If no instance is specified, Db.factory() is used.

Dependent Data

A Select object can splice in sibling or child data for each row.

withSiblingData(propertyName, siblingSql)

Example:

const query = Select.parse('SELECT id, name FROM users');
query.withSiblingData(
    'homeAddress',
    Select.parse(`
        SELECT * FROM addresses
        WHERE addresses.user_id IN(:id)
        AND addresses.type = 'home'
        AND addresses.deleted_at IS NULL
    `),
);
query.withSiblingData(
    'workAddress',
    Select.parse(`
        SELECT * FROM addresses
        WHERE addresses.user_id IN(:id)
        AND addresses.type = 'work'
        AND addresses.deleted_at IS NULL
    `),
);
const { results } = await query.fetch();

...and results for example may equal:

results = [
    {
        id: 1,
        name: 'John',
        homeAddress: {
            id: 11,
            type: 'home',
            is_active: 1,
            user_id: 1,
            street: '123 Any St.',
            city: 'Any Town',
            state: 'CA'
        },
        workAddress: {
            id: 12,
            type: 'work',
            is_active: 1,
            user_id: 1,
            street: '123 Commerce Dr.',
            city: 'Any Town',
            state: 'CA',
        },
    },
    {
        id: 2,
        name: 'Jane',
        // rows without sibling data will be null
        homeAddress: null,
        workAddress: {
            id: 12,
            type: 'work',
            is_active: 1,
            user_id: 2,
            street: '123 Tower Blvd.',
            city: 'Any Town',
            state: 'CA',
        },
    }
]

withChildData(propertyName, childSql)

Example:

const query = Select.parse('SELECT id, headline, published_by FROM posts');
query.withChildData(
    'theComments',
    Select.parse('SELECT * FROM comments WHERE comments.post_id IN(:id)')
);
query.withChildData(
    'theTags',
    Select.parse(`
        SELECT posts_tags.post_id, tags.* FROM tags
        INNER JOIN posts_tags ON posts_tags.tag_id = tags.id
        WHERE posts_tags.post_id IN(:id)
    `)
);
query.withSiblingData(
    'thePublisher',
    Select.parse('SELECT id, name FROM users WHERE user_id IN(:published_by)')
);
const { results } = await query.fetch();

...and results for example may equal:

results = [
    {
        id: 1,
        headline: 'Turmoil in China',
        published_by: 1001,
        theComments: [
            {
                id: 11,
                post_id: 1,
                user_id: 101,
                text: 'Sad to hear',
            },
            {
                id: 12,
                post_id: 1,
                user_id: 102,
                text: 'Hope it works out',
            },
        ],
        theTags: [
            {
                id: 101,
                post_id: 1,
                name: 'China',
            },
            {
                id: 102,
                post_id: 1,
                name: 'Crisis',
            },
        ],
        thePublisher: {
            id: 1001,
            name: 'John',
        },
    },
    {
        id: 2,
        headline: 'Syria at War',
        // records with missing child data will hae empty arrays
        theComments: [],
        theTags: [],
        thePublisher: null,
    }
]

Other methods

Select has a few other useful methods.

  • query.getClone() - Get an exact copy of this query object
  • query.unjoin(table) - Remove a join expression
  • query.escape(value) - Escape a raw value
  • query.escapeQuoteless(value) - Escape a value but avoid wrapping in quotes
  • query.toString() - Get prettified SQL
  • query.normalized() - Get raw SQL (all whitespace is spaces)
  • query.toBoundSql() - Get raw SQL with bindings replaced
  • query.reset(field) - Reset a single aspect of the query (e.g. where, having)
  • query.reset() - Reset query to an empty state

Select.parse() Limitations

Select.parse() uses regular expressions and is not a true parser. The intent is to be fast and useful for 99% of situations.

Below are some limitations illustrated by example.

Nested Subqueries

Most subqueries can be parsed but sub-subqueries don't work.

// WILL NOT WORK
const query = Select.parse(`
SELECT * FROM categories_posts WHERE category_id IN(
    SELECT id FROM categories WHERE client_id IN(
        SELECT client_id FROM affiliations WHERE name LIKE :name
    )
)`);
// WILL WORK
const subquery = Select.parse(`SELECT id FROM categories WHERE client_id IN(
    SELECT client_id FROM affiliations WHERE name LIKE :name
)`);
subquery.bind({ name: 'DogeCoin' });
const query = Select.parse(`SELECT * FROM categories_posts WHERE`);
query.where(`category_id IN(${subquery})`);

Keywords in Strings

If you need to use keywords in strings, use bindings.

-- WILL NOT WORK
SELECT id, CONCAT('WHERE ', expr) FROM users WHERE name = :name;
-- WILL WORK
SELECT id, CONCAT(:binding, expr) FROM users WHERE name = :name;

Nested OR and AND Clauses

Nested logic can't be parsed properly.

-- WILL NOT WORK
SELECT * FROM users
WHERE (
    fname = :fname AND (
        lname LIKE '%john' OR lname LIKE 'john%'
    ) OR (
        id > 0 AND is_active IS NOT NULL
    )
)
// WILL WORK
const query = Select.parse(`SELECT * FROM users`);
query.orWhere([
	"fname = :fname AND (lname LIKE '%john' OR lname LIKE 'john%')",
	'id > 0 AND is_active IS NOT NULL',
]);

DataBroker

DataBroker is useful for inserting and deleting data that will needs to be removed and restored.

Use in Integration Tests

With integration tests, it may be useful to insert test data, run assertions and then clean up the test data.

Insertions

Use the .insert() method to add records and then call .cleanup() to remove those records.

Example:

const { DataBroker, Db } = require('sharp-db');
const broker = new DataBroker(Db.factory(config));
const userId = await broker.insert('users', {
	name: 'John',
	is_active: true,
});
// the new user ID is also available at broker.ids
expect(broker.ids.users[0]).toBe(userId);
// ... integration test using userId ...
// then clean up all data
await broker.cleanup();

Example with composite key:

const { DataBroker, Db } = require('sharp-db');
const broker = new DataBroker(Db.factory(config));
const userId = await broker.insert('posts_images', {
	post_id: 1,
	image_id: 2,
	sort: 1,
}, { compositeKey: ['post_id', 'image_id'] });
// the new user ID is also available at broker.ids
expect(broker.ids.posts_images[0]).toEqual({
	post_id: 1,
	image_id: 2,
});
// ... integration test using userId ...
// then clean up all data
await broker.cleanup();

Deletions

Example:

const { DataBroker, Db } = require('sharp-db');
const broker = new DataBroker(Db.factory(config));
// affectedRows will be the count of records deleted
const affectedRows = await broker.delete('users', { status_id: 5 });
// the deleted records are available at broker.deleted
expect(broker.deleted).toHaveLength(affectedRows);
// ... integration test ...
// then restore all the deleted all data
await broker.cleanup();

SqlBuilder

Function list

The SqlBuilder objects builds SQL for Db methods such as selectFrom(). Below is a full list of methods if you want to build SQL outside of Db.

  • SqlBuilder.quote(identifier)
  • SqlBuilder.escape(value)
  • SqlBuilder.selectFrom(table, fields, criteria, extra)
  • SqlBuilder.selectBy(table, column, value)
  • SqlBuilder.insertInto(table, row)
  • SqlBuilder.insertIntoOnDuplicateKeyUpdate(table, insert, update)
  • SqlBuilder.insertExtended(table, rows)
  • SqlBuilder.updateTable(table, set, where)
  • SqlBuilder.deleteFrom(table, where, limit)
  • SqlBuilder.exportRows(table, rows, options)
  • SqlBuilder.buildWhere(field, value)
  • SqlBuilder.buildWheres(wheres)

QueryLogger

Logging queries

Example:

const { Db, QueryLogger } = require('sharp-db');
const logger = new QueryLogger();
const db = Db.factory();
logger.watch(db);
// ... run queries
logger.getLastQuery(); // last query
logger.getQueries(); // all queries
logger.clear(); // clear all logs
logger.unwatch(db); // stop capturing logs