vega-loader

What is vega-loader?

The vega-loader npm package is a utility for loading data in Vega visualizations. It provides a flexible and extensible way to fetch data from various sources, including URLs, files, and inline data. This package is essential for handling data input in Vega specifications, enabling users to integrate diverse data sources seamlessly.

What are vega-loader's main functionalities?

Loading Data from URLs

This feature allows users to load data from a specified URL. The loader fetches the data and returns it as a promise, which can then be processed as needed.

const vega = require('vega-loader');
const loader = vega.loader();
loader.load('https://api.example.com/data.json').then(data => {
  console.log(JSON.parse(data));
});

Loading Inline Data

This feature allows users to load inline data directly. The loader can handle different data formats such as JSON, CSV, and TSV, making it versatile for various data input scenarios.

const vega = require('vega-loader');
const loader = vega.loader();
const inlineData = '[{"name": "Alice", "age": 30}, {"name": "Bob", "age": 25}]';
loader.load({type: 'json', value: inlineData}).then(data => {
  console.log(data);
});

Customizing Loaders

This feature allows users to customize the loader with options such as base URL and headers. This is useful for scenarios where data fetching requires authentication or specific configurations.

const vega = require('vega-loader');
const customLoader = vega.loader({
  baseURL: 'https://api.example.com/',
  headers: { 'Authorization': 'Bearer token' }
});
customLoader.load('data.json').then(data => {
  console.log(JSON.parse(data));
});

Other packages similar to vega-loader

axios

Axios is a popular promise-based HTTP client for the browser and Node.js. It provides a simple API for making HTTP requests and handling responses. Compared to vega-loader, axios is more general-purpose and can be used for a wider range of HTTP request scenarios beyond just loading data for visualizations.

d3-fetch

d3-fetch is a module in the D3.js library that provides functions for fetching data. It supports various formats such as JSON, CSV, and TSV. While d3-fetch is similar to vega-loader in terms of data loading capabilities, it is more tightly integrated with the D3.js ecosystem and is often used in conjunction with other D3 modules for data visualization.

node-fetch

Node-fetch is a lightweight module that brings window.fetch to Node.js. It allows server-side code to make HTTP requests using the Fetch API. Compared to vega-loader, node-fetch is more focused on providing a Fetch API implementation for Node.js and does not include data parsing or visualization-specific features.

README

vega-loader

Network request and file loading utilities.

API Reference

• File Loading
• Data Format Parsing

File Loading

# vega.loader([options]) <>

Creates a new loader instance with default options. A loader object provides methods for loading files from the network or disk, and for sanitizing requested URLs and filenames. If provided, the key-value pairs in the options object will be passed as default options to the various loader methods.

The options object can include the following entries:

• baseURL: A base URL prefix to append to provided uri values. This can be useful for applications that load multiple data sets from the same domain.
• mode: A string explicitly indicating the loading mode. One of 'file' (server-side only) or 'http'. If set to 'file' mode, later uri parameters may safely omit a 'file://' prefix.
• defaultProtocol: The default protocol to use for protocol-relative uri values (e.g., '//vega.github.io'). Defaults to 'http'.
• target: The browser target attribute for hyperlinks. Only applies when sanitizing uri values for use as a hyperlink.
• headers: An object of key-values indicating custom request headers, used only when loading via HTTP.

# loader.load(uri[, options]) <>

Loads a file from either the network or disk, and returns a Promise for asyncronously accessing the loaded content. This method does not perform any parsing, it simply returns the loaded data as either a Buffer or String instance, depending on the execution environment. To subsequently parse loaded data, use the read method.

The uri argument is a value indicating the file to load. This is typically either an absolute or relative URL string. If running server-side via node.js, this argument might also be a file path (e.g., 'file:///path/to/file.txt').

If provided, the options argument will be combined with any default options passed to the loader constructor. In the case of identical property names, values from the options argument for this method will be used.

var loader = vega.loader();
loader.load('data.json').then(function(data) {
  // do something with loaded data
}).catch(function(error) {
  // error handling here
});

# loader.sanitize(uri, options) <>

URI sanitizer function, which takes a uri and options object as input, and returns a Promise that resolves to a return object that includes a sanitized URL under the href property. This method is used internally by load to ensure the URL is valid and to add additional protocol and hostname information, if needed. This method accepts the same options object accepted by load and returns a Promise. If sanitization is successful, the Promise resolves to a return object containing the URL string as (href), along with a non-enumerable boolean localFile flag, indicating if the file should be loaded from the local filesystem. The Promise rejects if the uri is invalid or disallowed. This method is over-writable for clients who wish to implement custom sanitization.

If provided, the options argument will be combined with any default options passed to the loader constructor. In the case of identical property names, values from the options argument for this method will be used.

# loader.http(url, options) <>

Function used internally by load for servicing HTTP requests. This method is over-writable for clients who wish to implement custom HTTP request handling. Uses d3-request by default.

If provided, the options argument will be combined with any default options passed to the loader constructor. In the case of identical property names, values from the options argument for this method will be used.

# loader.file(filename) <>

Function used internally by load for local file system requests. This method is over-writable for clients who wish to implement custom file loading. Uses the node.js fs module by default.

Data Format Parsing

# vega.read(data, schema[, dateParse]) <>

Parse loaded data according to a given format schema. The data argument should be either a String or Buffer instance, typically the result of calling load.

The schema object contents may depend on the data format (see below). Common options include:

• type: The data format type, such as json, csv, tsv, or topojson.
• property: For JSON types, specifies a property of the loaded JSON to reference. This is useful if a loaded JSON file contains multiple data sets and one would like to parse data under a specific property.
• parse: When set to 'auto' (the default), the method will perform type inference (using the inferTypes method) to determine data types of each field. Alternatively, callers can specify parsing rules by providing an object mapping field names to data types (for example: {'timestamp': 'date', 'price': 'number'}). The valid data type options are 'boolean', 'integer', 'number', 'date', and 'string'.

The 'date' data type also accepts an optional format string ('date:format'). If provided, the optional dateParse function is used to generate date-time parsers for a date format string. If dateParse is unspecified, the d3-time-format library is used by default. Date-time format strings may be quoted (date:'%A'), but quoting is not required. In addition, parsing of date-time format strings to UTC time is supported ('utc:format').

// read loaded csv data, automatically infer value types
var data = null;
loader.load('data/stocks.csv').then(function(data) {
  data = vega.read(csv_data, {type: 'csv', parse: 'auto'});
});

// read loaded csv data, using provided value types
var data = null;
loader.load('data/stocks.csv').then(function(data) {
  data = vega.read(data, {
    type: 'csv',
    parse: {'date': 'date', 'price': 'number'}
  });
});

// read loaded topojson data, extract mesh of countries
var topojson = null;
loader.load('data/world-110m.json').then(function(data) {
  topojson = vega.read(data, {type: 'topojson', mesh: 'countries'});
});

# vega.inferType(values[, field]) <>

Given an array of values, infers their data type as one of 'boolean', 'integer', 'number', 'date', or 'string'. An optional field accessor can be used to first extract values from the input array, and is equivalent to first calling values.map(field).

# vega.inferTypes(data, fields) <>

Given an array of data objects and a list of string-typed field names (fields), infers the data type for each field. Returns an object that maps field names to inferred types, determined using the inferType method.

# vega.typeParsers <>

An object containing a set of parsing functions for converting input values to a specified data type. All parsing functions return null if the input is null, undefined or the empty string ('').

The supported functions are:

• typeParsers.boolean(value): Parse the input value to a Boolean.
• typeParsers.integer(value): Parse the input value to an integer Number.
• typeParsers.number(value): Parse the input value to a Number.
• typeParsers.date(value[, parser]): Parse the input value to a Date. If provided, the parser function is used to interpret the value; otherwise Date.parse is used.
• typeParsers.string(value): Parse the input value to a String. If value is not already string-typed, it is coerced to a String.

# vega.formats(name[, format]) <>

Registry function for data format parsers. If invoked with two arguments, adds a new format parser with the provided name. Otherwise, returns an existing parser with the given name. The method signature of a format parser is:

• format(data, options)

A format parser that accepts two arguments, the input data to parse (e.g., a block of CSV text) and a set of format-specific options. The following data formats are registered by default:

• dsv: Delimiter-separated values format. Each line of text is a record, with each field separated by a delimiter string. Accepts a delimiter option indicating the delimiter string used to separate field values.
• csv: Comma-separated values format. A dsv instance with a comma (,) delimiter.
• tsv: Tab-separated values format. A dsv instance with a tab (\t) delimiter.
• json: JavaScript Object Notation (JSON) format. Accepts a property option, indicating a sub-property of the parsed JSON to return; useful if a data array is nested within a larger object.
• topojson: TopoJSON format for compressed encoding of geographic data. Requires either a feature option indicating the name of the geographic feature to extract (e.g., extracts individual paths for all countries), or a mesh option indicating a feature name for which a single mesh should be extracted (e.g., all country boundaries in a single path). Please see the TopoJSON documentation for more.