JSPM

  • Created
  • Published
  • Downloads 24163
  • Score
    100M100P100Q163883F
  • License ISC

Client for ClickHouse

Package Exports

  • clickhouse

This package does not declare an exports field, so the exports above have been automatically detected and optimized by JSPM instead. If any package subpath is missing, it is recommended to post an issue to the original package (clickhouse) to support the "exports" field. If that is not possible, create a JSPM override to customize the exports field for this package.

Readme

clickhouse

NodeJS client for ClickHouse. Send query over HTTP interface.

Install:

npm i clickhouse

Example:

const { ClickHouse } = require('clickhouse');

const clickhouse = new ClickHouse();

or with all options:

const clickhouse = new ClickHouse({
    url: 'http://localhost',
    port: 8123,
    debug: false,
    basicAuth: null,
    isUseGzip: false,
    format: "json", // "json" || "csv" || "tsv"
    config: {
        session_id                              : 'session_id if neeed',
        session_timeout                         : 60,
        output_format_json_quote_64bit_integers : 0,
        enable_http_compression                 : 0,
        database                                : 'my_database_name',
    },
    
    // This object merge with request params (see request lib docs)
    reqParams: {
        ...
    }
});

or change

basicAuth: null

to

basicAuth: {
    username: 'default',
    password: '',
},

Exec query:

const queries = [
    'DROP TABLE IF EXISTS session_temp',

    `CREATE TABLE session_temp (
        date Date,
        time DateTime,
        mark String,
        ips Array(UInt32),
        queries Nested (
            act String,
            id UInt32
        )
    )
    ENGINE=MergeTree(date, (mark, time), 8192)`,

    'OPTIMIZE TABLE ukit.loadstat PARTITION 201807 FINAL'
];

for(const query of queries) {
    const r = await clickhouse.query(query).toPromise();

    console.log(query, r);
}

Exec by callback way:

clickhouse.query(query).exec(function (err, rows) {
    ...
});

Stream:

clickhouse.query(`SELECT number FROM system.numbers LIMIT 10`).stream()
    .on('data', function() {
        const stream = this;

        stream.pause();

        setTimeout(() => {
            stream.resume();
        }, 1000);
    })
    .on('error', err => {
        ...
    })
    .on('end', () => {
        ...
    });

or async stream:

// async iteration
for await (const row of clickhouse.query(sql).stream()) {
    console.log(row);
}

As promise:

const rows = await clickhouse.query(query).toPromise();

// use query with external data
const rows = await clickhouse.query('SELECT * AS count FROM temp_table', {
    external: [
        {
            name: 'temp_table',
            data: e._.range(0, rowCount).map(i => `str${i}`)
        },
    ]
}).toPromise();

Set session:

clickhouse.sessionId = '...';
const r = await clickhouse.query(
    `CREATE TEMPORARY TABLE test_table
    (_id String, str String)
    ENGINE=Memory`
).toPromise();

In case your application requires specific sessions to manage specific data then you can send session_id with each query.

let mySessionId = 'some_randome_string';
const r = await clickhouse.query(
    `CREATE TEMPORARY TABLE test_table
    (_id String, str String)
    ENGINE=Memory`, {}, {sessionId: mySessionId}
).toPromise();

Insert stream:

const ws = clickhouse.insert('INSERT INTO session_temp').stream();
for(let i = 0; i <= 1000; i++) {
    await ws.writeRow(
        [
            e._.range(0, 50).map(
                j => `${i}:${i * 2}:${j}`
            ).join('-')
        ]
    );
}

//wait stream finish
const result = await ws.exec();

Pipe readable stream to writable stream (across transform):

const rs = clickhouse.query(query).stream();

const tf = new stream.Transform({
    objectMode : true,
    transform  : function (chunk, enc, cb) {
        cb(null, JSON.stringify(chunk) + '\n');
    }
});

clickhouse.sessionId = Date.now();
const ws = clickhouse.insert('INSERT INTO session_temp2').stream();

const result = await rs.pipe(tf).pipe(ws).exec();

Changelogs:

  • 2020-11-02 (v2.2.0) Backward Incompatible Change
    • port from url more important then port from config
  • 2020-04-17 (v2.1.0)
    • Fix query with totals. For json formats work perfect, but for another - doesn't
  • 2019-02-13
    • Add compatibility with user and username options
  • 2019-02-07
    • Add TLS/SSL Protocol support
    • Add async iteration over SELECT

Links