You can also stream a file to a PUT or POST request. This method will also check the file extension against a mapping of file extensions to content-types (in this case `application/json`) and use the proper `content-type` in the PUT request (if the headers don’t already provide one).
Request emits a "response" event when a response is received. The `response` argument will be an instance of [http.IncomingMessage](https://nodejs.org/api/http.html#http_class_http_incomingmessage).
You can also `pipe()` from `http.ServerRequest` instances, as well as to `http.ServerResponse` instances. The HTTP method, headers, and entity-body data will be sent. Which means that, if you don't really care about security, you can do:
```js
http.createServer(function (req, resp) {
if (req.url === '/doodle.png') {
const x = request('http://mysite.com/doodle.png')
req.pipe(x)
x.pipe(resp)
}
})
```
And since `pipe()` returns the destination stream in ≥ Node 0.5.x you can do one line proxying. :)
Also, none of this new functionality conflicts with requests previous features, it just expands them.
```js
const r = request.defaults({'proxy':'http://localproxy.com'})
http.createServer(function (req, resp) {
if (req.url === '/doodle.png') {
r.get('http://google.com/doodle.png').pipe(resp)
}
})
```
You can still use intermediate proxies, the requests will still follow HTTP forwards, etc.
[back to top](#table-of-contents)
---
## Promises & Async/Await
`request` supports both streaming and callback interfaces natively. If you'd like `request` to return a Promise instead, you can use an alternative interface wrapper for `request`. These wrappers can be useful if you prefer to work with Promises, or if you'd like to use `async`/`await` in ES2017.
Several alternative interfaces are provided by the request team, including:
Also, [`util.promisify`](https://nodejs.org/api/util.html#util_util_promisify_original), which is available from Node.js v8.0 can be used to convert a regular function that takes a callback to return a promise instead.
[back to top](#table-of-contents)
---
## Forms
`request` supports `application/x-www-form-urlencoded` and `multipart/form-data` form uploads. For `multipart/related` refer to the `multipart` API.
For `multipart/form-data` we use the [form-data](https://github.com/form-data/form-data) library by [@felixge](https://github.com/felixge). For the most cases, you can pass your upload form data via the `formData` option.
// Pass optional meta-data with an 'options' object with style: {value: DATA, options: OPTIONS}
// Use case: for some types of streams, you'll need to provide "file"-related information manually.
// See the `form-data` README for more information about options: https://github.com/form-data/form-data
custom_file: {
value: fs.createReadStream('/dev/urandom'),
options: {
filename: 'topsecret.jpg',
contentType: 'image/jpeg'
}
}
};
request.post({url:'http://service.com/upload', formData: formData}, function optionalCallback(err, httpResponse, body) {
if (err) {
return console.error('upload failed:', err);
}
console.log('Upload successful! Server responded with:', body);
});
```
For advanced cases, you can access the form-data object itself via `r.form()`. This can be modified until the request is fired on the next cycle of the event-loop. (Note that this calling `form()` will clear the currently set form data for that request.)
```js
// NOTE: Advanced use-case, for normal use see 'formData' usage above
const r = request.post('http://service.com/upload', function optionalCallback(err, httpResponse, body) {...})
See the [form-data README](https://github.com/form-data/form-data) for more information & examples.
#### multipart/related
Some variations in different HTTP implementations require a newline/CRLF before, after, or both before and after the boundary of a `multipart/related` request (using the multipart option). This has been observed in the .NET WebAPI version 4.0. You can turn on a boundary preambleCRLF or postamble by passing them as `true` to your request options.
* Pass `transport_method : 'query'` or `transport_method : 'body'` in the OAuth
options object.
*`transport_method` defaults to `'header'`
To use [Request Body Hash](https://oauth.googlecode.com/svn/spec/ext/body_hash/1.0/oauth-bodyhash.html) you can either
* Manually generate the body hash and pass it as a string `body_hash: '...'`
* Automatically generate the body hash by passing `body_hash: true`
[back to top](#table-of-contents)
---
## Proxies
If you specify a `proxy` option, then the request (and any subsequent
redirects) will be sent via a connection to the proxy server.
If your endpoint is an `https` url, and you are using a proxy, then
request will send a `CONNECT` request to the proxy server *first*, and
then use the supplied connection to connect to the endpoint.
That is, first it will make a request like:
```
HTTP/1.1 CONNECT endpoint-server.com:80
Host: proxy-server.com
User-Agent: whatever user agent you specify
```
and then the proxy server make a TCP connection to `endpoint-server`
on port `80`, and return a response that looks like:
```
HTTP/1.1 200 OK
```
At this point, the connection is left open, and the client is
communicating directly with the `endpoint-server.com` machine.
See [the wikipedia page on HTTP Tunneling](https://en.wikipedia.org/wiki/HTTP_tunnel)
for more information.
By default, when proxying `http` traffic, request will simply make a
standard proxied `http` request. This is done by making the `url`
section of the initial line of the request a fully qualified url to
the endpoint.
For example, it will make a single request that looks like:
```
HTTP/1.1 GET http://endpoint-server.com/some-url
Host: proxy-server.com
Other-Headers: all go here
request body or whatever
```
Because a pure "http over http" tunnel offers no additional security
or other features, it is generally simpler to go with a
straightforward HTTP proxy in this case. However, if you would like
to force a tunneling proxy, you may set the `tunnel` option to `true`.
You can also make a standard proxied `http` request by explicitly setting
`tunnel : false`, but **note that this will allow the proxy to see the traffic
to/from the destination server**.
If you are using a tunneling proxy, you may set the
`proxyHeaderWhiteList` to share certain headers with the proxy.
You can also set the `proxyHeaderExclusiveList` to share certain
headers only with the proxy and not with destination host.
By default, this set is:
```
accept
accept-charset
accept-encoding
accept-language
accept-ranges
cache-control
content-encoding
content-language
content-length
content-location
content-md5
content-range
content-type
connection
date
expect
max-forwards
pragma
proxy-authorization
referer
te
transfer-encoding
user-agent
via
```
Note that, when using a tunneling proxy, the `proxy-authorization`
header and any headers from custom `proxyHeaderExclusiveList` are
*never* sent to the endpoint server, but only to the proxy server.
### Controlling proxy behaviour using environment variables
The following environment variables are respected by `request`:
*`HTTP_PROXY` / `http_proxy`
*`HTTPS_PROXY` / `https_proxy`
*`NO_PROXY` / `no_proxy`
When `HTTP_PROXY` / `http_proxy` are set, they will be used to proxy non-SSL requests that do not have an explicit `proxy` configuration option present. Similarly, `HTTPS_PROXY` / `https_proxy` will be respected for SSL requests that do not have an explicit `proxy` configuration option. It is valid to define a proxy in one of the environment variables, but then override it for a specific request, using the `proxy` configuration option. Furthermore, the `proxy` configuration option can be explicitly set to false / null to opt out of proxying altogether for that request.
`request` is also aware of the `NO_PROXY`/`no_proxy` environment variables. These variables provide a granular way to opt out of proxying, on a per-host basis. It should contain a comma separated list of hosts to opt out of proxying. It is also possible to opt of proxying when a particular destination port is used. Finally, the variable may be set to `*` to opt out of the implicit proxy configuration of the other environment variables.
Here's some examples of valid `no_proxy` values:
*`google.com` - don't proxy HTTP/HTTPS requests to Google.
*`google.com:443` - don't proxy HTTPS requests to Google, but *do* proxy HTTP requests to Google.
*`google.com:443, yahoo.com:80` - don't proxy HTTPS requests to Google, and don't proxy HTTP requests to Yahoo!
`request` supports making requests to [UNIX Domain Sockets](https://en.wikipedia.org/wiki/Unix_domain_socket). To make one, use the following URL scheme:
```js
/* Pattern */ 'http://unix:SOCKET:PATH'
/* Example */ request.get('http://unix:/absolute/path/to/unix.socket:/request/path')
```
Note: The `SOCKET` path is assumed to be absolute to the root of the host file system.
[back to top](#table-of-contents)
---
## TLS/SSL Protocol
TLS/SSL Protocol options, such as `cert`, `key` and `passphrase`, can be
set directly in `options` object, in the `agentOptions` property of the `options` object, or even in `https.globalAgent.options`. Keep in mind that, although `agentOptions` allows for a slightly wider range of configurations, the recommended way is via `options` object directly, as using `agentOptions` or `https.globalAgent.options` would not be applied in the same way in proxied environments (as data travels through a TLS connection instead of an http/https agent).
// Or use `pfx` property replacing `cert` and `key` when using private key, certificate and CA certs in PFX or PKCS12 format:
// pfx: fs.readFileSync(pfxFilePath),
passphrase: 'password',
securityOptions: 'SSL_OP_NO_SSLv3'
}
};
request.get(options);
```
It is able to force using SSLv3 only by specifying `secureProtocol`:
```js
request.get({
url: 'https://api.some-server.com/',
agentOptions: {
secureProtocol: 'SSLv3_method'
}
});
```
It is possible to accept other certificates than those signed by generally allowed Certificate Authorities (CAs).
This can be useful, for example, when using self-signed certificates.
To require a different root certificate, you can specify the signing CA by adding the contents of the CA's certificate file to the `agentOptions`.
The certificate the domain presents must be signed by the root certificate specified:
```js
request.get({
url: 'https://api.some-server.com/',
agentOptions: {
ca: fs.readFileSync('ca.cert.pem')
}
});
```
The `ca` value can be an array of certificates, in the event you have a private or internal corporate public-key infrastructure hierarchy. For example, if you want to connect to https://api.some-server.com which presents a key chain consisting of:
1. its own public key, which is signed by:
2. an intermediate "Corp Issuing Server", that is in turn signed by:
3. a root CA "Corp Root CA";
you can configure your request as follows:
```js
request.get({
url: 'https://api.some-server.com/',
agentOptions: {
ca: [
fs.readFileSync('Corp Issuing Server.pem'),
fs.readFileSync('Corp Root CA.pem')
]
}
});
```
[back to top](#table-of-contents)
---
## Support for HAR 1.2
The `options.har` property will override the values: `url`, `method`, `qs`, `headers`, `form`, `formData`, `body`, `json`, as well as construct multipart data and read files from disk when `request.postData.params[].fileName` is present without a matching `value`.
A validation step will check if the HAR Request format matches the latest spec (v1.2) and will skip parsing if not matching.
```js
const request = require('request')
request({
// will be ignored
method: 'GET',
uri: 'http://www.google.com',
// HTTP Archive Request Object
har: {
url: 'http://www.mockbin.com/har',
method: 'POST',
headers: [
{
name: 'content-type',
value: 'application/x-www-form-urlencoded'
}
],
postData: {
mimeType: 'application/x-www-form-urlencoded',
params: [
{
name: 'foo',
value: 'bar'
},
{
name: 'hello',
value: 'world'
}
]
}
}
})
// a POST request will be sent to http://www.mockbin.com
// with body an application/x-www-form-urlencoded body:
// foo=bar&hello=world
```
[back to top](#table-of-contents)
---
## request(options, callback)
The first argument can be either a `url` or an `options` object. The only required option is `uri`; all others are optional.
-`uri` || `url` - fully qualified uri or a parsed url object from `url.parse()`
-`baseUrl` - fully qualified uri string used as the base url. Most useful with `request.defaults`, for example when you want to do many requests to the same domain. If `baseUrl` is `https://example.com/api/`, then requesting `/end/point?test=true` will fetch `https://example.com/api/end/point?test=true`. When `baseUrl` is given, `uri` must also be a string.
-`method` - http method (default: `"GET"`)
-`headers` - http headers (default: `{}`)
---
-`qs` - object containing querystring values to be appended to the `uri`
-`qsParseOptions` - object containing options to pass to the [qs.parse](https://github.com/hapijs/qs#parsing-objects) method. Alternatively pass options to the [querystring.parse](https://nodejs.org/docs/v0.12.0/api/querystring.html#querystring_querystring_parse_str_sep_eq_options) method using this format `{sep:';', eq:':', options:{}}`
-`qsStringifyOptions` - object containing options to pass to the [qs.stringify](https://github.com/hapijs/qs#stringifying) method. Alternatively pass options to the [querystring.stringify](https://nodejs.org/docs/v0.12.0/api/querystring.html#querystring_querystring_stringify_obj_sep_eq_options) method using this format `{sep:';', eq:':', options:{}}`. For example, to change the way arrays are converted to query strings using the `qs` module pass the `arrayFormat` option with one of `indices|brackets|repeat`
-`useQuerystring` - if true, use `querystring` to stringify and parse
querystrings, otherwise use `qs` (default: `false`). Set this option to
`true` if you need arrays to be serialized as `foo=bar&foo=baz` instead of the
default `foo[0]=bar&foo[1]=baz`.
---
-`body` - entity body for PATCH, POST and PUT requests. Must be a `Buffer`, `String` or `ReadStream`. If `json` is `true`, then `body` must be a JSON-serializable object.
-`form` - when passed an object or a querystring, this sets `body` to a querystring representation of value, and adds `Content-type: application/x-www-form-urlencoded` header. When passed no options, a `FormData` instance is returned (and is piped to request). See "Forms" section above.
-`formData` - data to pass for a `multipart/form-data` request. See
[Forms](#forms) section above.
-`multipart` - array of objects which contain their own headers and `body`
attributes. Sends a `multipart/related` request. See [Forms](#forms) section
above.
- Alternatively you can pass in an object `{chunked: false, data: []}` where
`chunked` is used to specify whether the request is sent in
[chunked transfer encoding](https://en.wikipedia.org/wiki/Chunked_transfer_encoding)
In non-chunked requests, data items with body streams are not allowed.
-`preambleCRLF` - append a newline/CRLF before the boundary of your `multipart/form-data` request.
-`postambleCRLF` - append a newline/CRLF at the end of the boundary of your `multipart/form-data` request.
-`json` - sets `body` to JSON representation of value and adds `Content-type: application/json` header. Additionally, parses the response body as JSON.
-`jsonReviver` - a [reviver function](https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/JSON/parse) that will be passed to `JSON.parse()` when parsing a JSON response body.
-`jsonReplacer` - a [replacer function](https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/JSON/stringify) that will be passed to `JSON.stringify()` when stringifying a JSON request body.
---
-`auth` - a hash containing values `user` || `username`, `pass` || `password`, and `sendImmediately` (optional). See documentation above.
-`oauth` - options for OAuth HMAC-SHA1 signing. See documentation above.
-`hawk` - options for [Hawk signing](https://github.com/hueniverse/hawk). The `credentials` key must contain the necessary signing info, [see hawk docs for details](https://github.com/hueniverse/hawk#usage-example).
-`aws` - `object` containing AWS signing information. Should have the properties `key`, `secret`, and optionally `session` (note that this only works for services that require session as part of the canonical string). Also requires the property `bucket`, unless you’re specifying your `bucket` as part of the path, or the request doesn’t use a bucket (i.e. GET Services). If you want to use AWS sign version 4 use the parameter `sign_version` with value `4` otherwise the default is version 2. If you are using SigV4, you can also include a `service` property that specifies the service name. **Note:** you need to `npm install aws4` first.
-`httpSignature` - options for the [HTTP Signature Scheme](https://github.com/joyent/node-http-signature/blob/master/http_signing.md) using [Joyent's library](https://github.com/joyent/node-http-signature). The `keyId` and `key` properties must be specified. See the docs for other options.
---
-`followRedirect` - follow HTTP 3xx responses as redirects (default: `true`). This property can also be implemented as function which gets `response` object as a single argument and should return `true` if redirects should continue or `false` otherwise.
-`followOriginalHttpMethod` - by default we redirect to HTTP method GET. you can enable this property to redirect to the original HTTP method (default: `false`)
-`maxRedirects` - the maximum number of redirects to follow (default: `10`)
-`removeRefererHeader` - removes the referer header when a redirect happens (default: `false`). **Note:** if true, referer header set in the initial request is preserved during redirect chain.
---
-`encoding` - encoding to be used on `setEncoding` of response data. If `null`, the `body` is returned as a `Buffer`. Anything else **(including the default value of `undefined`)** will be passed as the [encoding](http://nodejs.org/api/buffer.html#buffer_buffer) parameter to `toString()` (meaning this is effectively `utf8` by default). (**Note:** if you expect binary data, you should set `encoding: null`.)
-`gzip` - if `true`, add an `Accept-Encoding` header to request compressed content encodings from the server (if not already present) and decode supported content encodings in the response. **Note:** Automatic decoding of the response content is performed on the body data returned through `request` (both through the `request` stream and passed to the callback function) but is not performed on the `response` stream (available from the `response` event) which is the unmodified `http.IncomingMessage` object which may contain compressed data. See example below.
-`jar` - if `true`, remember cookies for future use (or define your custom cookie jar; see examples section)
---
-`agent` - `http(s).Agent` instance to use
-`agentClass` - alternatively specify your agent's class name
-`agentOptions` - and pass its options. **Note:** for HTTPS see [tls API doc for TLS/SSL options](http://nodejs.org/api/tls.html#tls_tls_connect_options_callback) and the [documentation above](#using-optionsagentoptions).
-`forever` - set to `true` to use the [forever-agent](https://github.com/request/forever-agent) **Note:** Defaults to `http(s).Agent({keepAlive:true})` in node 0.12+
-`pool` - an object describing which agents to use for the request. If this option is omitted the request will use the global agent (as long as your options allow for it). Otherwise, request will search the pool for your custom agent. If no custom agent is found, a new agent will be created and added to the pool. **Note:**`pool` is used only when the `agent` option is not specified.
- A `maxSockets` property can also be provided on the `pool` object to set the max number of sockets for all agents created (ex: `pool: {maxSockets: Infinity}`).
- Note that if you are sending multiple requests in a loop and creating
multiple new `pool` objects, `maxSockets` will not work as intended. To
work around this, either use [`request.defaults`](#requestdefaultsoptions)
with your pool options or create the pool object with the `maxSockets`
property outside of the loop.
-`timeout` - integer containing number of milliseconds, controls two timeouts.
- **Read timeout**: Time to wait for a server to send response headers (and start the response body) before aborting the request.
- **Connection timeout**: Sets the socket to timeout after `timeout` milliseconds of inactivity. Note that increasing the timeout beyond the OS-wide TCP connection timeout will not have any effect ([the default in Linux can be anywhere from 20-120 seconds][linux-timeout])
-`localAddress` - local interface to bind for network connections.
-`proxy` - an HTTP proxy to be used. Supports proxy Auth with Basic Auth, identical to support for the `url` parameter (by embedding the auth info in the `uri`)
-`strictSSL` - if `true`, requires SSL certificates be valid. **Note:** to use your own certificate authority, you need to specify an agent that was created with that CA as an option.
-`undefined` (default) - `true` if the destination is `https`, `false` otherwise
-`true` - always tunnel to the destination by making a `CONNECT` request to
the proxy
-`false` - request the destination as a `GET` request.
-`proxyHeaderWhiteList` - a whitelist of headers to send to a
tunneling proxy.
-`proxyHeaderExclusiveList` - a whitelist of headers to send
exclusively to a tunneling proxy and not to destination.
---
-`time` - if `true`, the request-response cycle (including all redirects) is timed at millisecond resolution. When set, the following properties are added to the response object:
-`elapsedTime` Duration of the entire request/response in milliseconds (*deprecated*).
-`responseStartTime` Timestamp when the response began (in Unix Epoch milliseconds) (*deprecated*).
-`timingStart` Timestamp of the start of the request (in Unix Epoch milliseconds).
-`timings` Contains event timestamps in millisecond resolution relative to `timingStart`. If there were redirects, the properties reflect the timings of the final request in the redirect chain:
-`socket` Relative timestamp when the [`http`](https://nodejs.org/api/http.html#http_event_socket) module's `socket` event fires. This happens when the socket is assigned to the request.
-`lookup` Relative timestamp when the [`net`](https://nodejs.org/api/net.html#net_event_lookup) module's `lookup` event fires. This happens when the DNS has been resolved.
-`connect`: Relative timestamp when the [`net`](https://nodejs.org/api/net.html#net_event_connect) module's `connect` event fires. This happens when the server acknowledges the TCP connection.
-`response`: Relative timestamp when the [`http`](https://nodejs.org/api/http.html#http_event_response) module's `response` event fires. This happens when the first bytes are received from the server.
-`end`: Relative timestamp when the last bytes of the response are received.
-`timingPhases` Contains the durations of each request phase. If there were redirects, the properties reflect the timings of the final request in the redirect chain:
-`wait`: Duration of socket initialization (`timings.socket`)
-`dns`: Duration of DNS lookup (`timings.lookup` - `timings.socket`)
-`tcp`: Duration of TCP connection (`timings.connect` - `timings.socket`)
-`firstByte`: Duration of HTTP server response (`timings.response` - `timings.connect`)
-`download`: Duration of HTTP download (`timings.end` - `timings.response`)
-`har` - a [HAR 1.2 Request Object](http://www.softwareishard.com/blog/har-12-spec/#request), will be processed from HAR format into options overwriting matching values *(see the [HAR 1.2 section](#support-for-har-12) for details)*
-`callback` - alternatively pass the request's callback in the options object
The callback argument gets 3 arguments:
1. An `error` when applicable (usually from [`http.ClientRequest`](http://nodejs.org/api/http.html#http_class_http_clientrequest) object)
2. An [`http.IncomingMessage`](https://nodejs.org/api/http.html#http_class_http_incomingmessage) object (Response object)
3. The third is the `response` body (`String` or `Buffer`, or JSON object if the `json` option is supplied)
[back to top](#table-of-contents)
---
## Convenience methods
There are also shorthand methods for different HTTP METHODs and some other conveniences.
### request.defaults(options)
This method **returns a wrapper** around the normal request API that defaults
to whatever options you pass to it.
**Note:** `request.defaults()`**does not** modify the global request API;
instead, it **returns a wrapper** that has your default settings applied to it.
**Note:** You can call `.defaults()` on the wrapper that is returned from
`request.defaults` to add/override defaults that were previously defaulted.
For example:
```js
//requests using baseRequest() will set the 'x-token' header
const baseRequest = request.defaults({
headers: {'x-token': 'my-token'}
})
//requests using specialRequest() will include the 'x-token' header set in
//baseRequest and will also include the 'special' header
const specialRequest = baseRequest.defaults({
headers: {special: 'special value'}
})
```
### request.METHOD()
These HTTP method convenience functions act just like `request()` but with a default method already set for you:
- *request.get()*: Defaults to `method: "GET"`.
- *request.post()*: Defaults to `method: "POST"`.
- *request.put()*: Defaults to `method: "PUT"`.
- *request.patch()*: Defaults to `method: "PATCH"`.
- *request.del() / request.delete()*: Defaults to `method: "DELETE"`.
- *request.head()*: Defaults to `method: "HEAD"`.
- *request.options()*: Defaults to `method: "OPTIONS"`.
### request.cookie()
Function that creates a new cookie.
```js
request.cookie('key1=value1')
```
### request.jar()
Function that creates a new cookie jar.
```js
request.jar()
```
### response.caseless.get('header-name')
Function that returns the specified response header field using a [case-insensitive match](https://tools.ietf.org/html/rfc7230#section-3.2)
```js
request('http://www.google.com', function (error, response, body) {
// print the Content-Type header even if the server returned it as 'content-type' (lowercase)
For backwards-compatibility, response compression is not supported by default.
To accept gzip-compressed responses, set the `gzip` option to `true`. Note
that the body data passed through `request` is automatically decompressed
while the response object is unmodified and will contain compressed data if
the server sent a compressed response.
```js
const request = require('request')
request(
{ method: 'GET'
, uri: 'http://www.google.com'
, gzip: true
}
, function (error, response, body) {
// body is the decompressed response body
console.log('server encoded the data as: ' + (response.headers['content-encoding'] || 'identity'))
console.log('the decoded data is: ' + body)
}
)
.on('data', function(data) {
// decompressed data as it is received
console.log('decoded chunk: ' + data)
})
.on('response', function(response) {
// unmodified http.IncomingMessage object
response.on('data', function(data) {
// compressed data as it is received
console.log('received ' + data.length + ' bytes of compressed data')
})
})
```
Cookies are disabled by default (else, they would be used in subsequent requests). To enable cookies, set `jar` to `true` (either in `defaults` or `options`).
```js
const request = request.defaults({jar: true})
request('http://www.google.com', function () {
request('http://images.google.com')
})
```
To use a custom cookie jar (instead of `request`’s global cookie jar), set `jar` to an instance of `request.jar()` (either in `defaults` or `options`)
