Google BigQuery Client Library for Node.js

The samples/ directory contains various Node.js implementation examples for BigQuery operations. Common tasks covered include:

• Table Management: createTable, deleteTable, copyTable, addEmptyColumn, addColumnLoadAppend, addColumnQueryAppend.
• Dataset Management: createDataset, deleteDataset.
• Job Management: createJob, cancelJob.
• Advanced Table Types: createTablePartitioned, createTableRangePartitioned, createTableClustered.
• Views and Routines: createView, createRoutine, createRoutineDDL.
• Data Extraction: extractTableToGCS, extractTableJSON, extractTableCompressed.
• Authentication: clientJSONCredentials, authViewTutorial.

Use the quickstart sample to get a minimal working example of the BigQuery Node.js client in action.

To run the sample:

node samples/quickstart.js

To get started with the Google BigQuery Node.js client library, follow these prerequisite steps:

1. Select or create a Cloud Platform project.
2. Enable the Google BigQuery API in your selected project.
3. Set up authentication to allow your local workstation to access the API (e.g., using Application Default Credentials).

Note: This repository is deprecated. All content and history have moved to google-cloud-node.

Install the BigQuery client library using npm:

npm install @google-cloud/bigquery

The BigQuery Node.js client library is compatible with all current active and maintenance versions of Node.js.

If you are using an end-of-life version of Node.js, it is recommended to update to an actively supported LTS version. While legacy versions are supported on a best-effort basis, they are not tested in continuous integration, may lack security patches, and may have outdated dependencies.

If you must use a legacy version of Node.js, you can install specific client library versions using npm dist-tags. The naming convention for these tags is legacy-(version).

npm install @google-cloud/bigquery@legacy-8

To run the performance benchmarks for the BigQuery client, execute the bench.js script using Node.js and provide a JSON file containing the queries to be benchmarked as an argument.

Note on Accuracy: Because the BigQuery service caches requests, you should run the benchmark at least twice and disregard the results from the first run to ensure you are measuring actual processing performance rather than cache hits.

node bench.js queries.json

Standard JavaScript numbers cannot safely represent all INT64 values. To maintain precision, use bigquery.int() to wrap values in a BigQueryInt object. You can provide an integerTypeCastFunction via IntegerTypeCastOptions to define how the value is converted.

If you attempt to decode an INT64 that exceeds Number.MAX_SAFE_INTEGER without custom casting, the library will throw an error.

const {BigQuery} = require('@google-cloud/bigquery');
const bigquery = new BigQuery();

const largeIntegerValue = Number.MAX_SAFE_INTEGER + 1;

const options = {
  integerTypeCastFunction: value => value.split(), // Example custom function
};

const bqInteger = bigquery.int(largeIntegerValue, options);
const customValue = bqInteger.valueOf();

The Dataset class allows you to interact with a specific BigQuery dataset. You can obtain a Dataset instance using BigQuery#createDataset or BigQuery#dataset.

Common tasks include creating, getting, checking existence, and deleting datasets, as well as managing child resources like tables, routines, and models.

const {BigQuery} = require('@google-cloud/bigquery');
const bigquery = new BigQuery();
const dataset = bigquery.dataset('institutions');

The Job class represents a BigQuery job and can be used to check the status of a running job or fetch results from a completed one.

Jobs are event emitters. The status of a job is polled continuously, but polling only begins after you register a complete listener. You should also register an error listener to catch issues that impede the job.

To stop the job from polling for updates, remove all listeners using removeAllListeners().

const {BigQuery} = require('@google-cloud/bigquery');
const bigquery = new BigQuery();

const job = bigquery.job('job-id');

// Polling starts once this listener is registered
job.on('complete', (metadata) => {
  // The job is complete.
});

job.on('error', (err) => {
  // An error occurred during the job.
});

// To stop polling:
job.removeAllListeners();