http-basic

http-basic

Simple wrapper arround http.request/https.request

Installation

npm install http-basic

Usage

var request = require('http-basic');

var options = {followRedirects: true, gzip: true, cache: 'memory'};

var req = request('GET', 'http://example.com', options, function (err, res) {
  if (err) throw err;
  console.dir(res.statusCode);
  res.body.resume();
});
req.end();

method:

The http method (e.g. GET, POST, PUT, DELETE etc.)

url:

The url as a string (e.g. http://example.com). It must be fully qualified and either http or https.

options:

• headers - (default {}) http headers
• agent - (default: false) controlls keep-alive (see http://nodejs.org/api/http.html#http_http_request_options_callback)
• followRedirects - (default: false) - if true, redirects are followed (note that this only affects the result in the callback)
• maxRedirects - (default: Infinity) - limit the number of redirects allowed.
• allowRedirectHeaders (default: null) - an array of headers allowed for redirects (none if null).
• gzip (default: false) - automatically accept gzip and deflate encodings. This is kept completely transparent to the user.
• cache - (default: null) - 'memory' or 'file' to use the default built in caches or you can pass your own cache implementation.
• timeout (default: false) - times out if no response is returned within the given number of milliseconds.
• socketTimeout (default: false) - calls req.setTimeout internally which causes the request to timeout if no new data is seen for the given number of milliseconds.
• retry (default: false) - retry GET requests. Set this to true to retry when the request errors or returns a status code greater than or equal to 400 (can also be a function that takes (err, req, attemptNo) => shouldRetry)
• retryDelay (default: 200) - the delay between retries (can also be set to a function that takes (err, res, attemptNo) => delay)
• maxRetries (default: 5) - the number of times to retry before giving up.

callback:

The callback is called with err as the first argument and res as the second argument. res is an http-response-object. It has the following properties:

• statusCode - a number representing the HTTP Status Code
• headers - an object representing the HTTP headers
• body - a readable stream respresenting the request body.
• url - the URL that was requested (in the case of redirects, this is the final url that was requested)

returns:

If the method is GET, DELETE or HEAD, it returns undefined.

Otherwise, it returns a writable stream for the body of the request.

Implementing a Cache

A Cache is an object with two methods:

• getResponse(url, callback) - retrieve a cached response object
• setResponse(url, response) - cache a response object

A cached response object is an object with the following properties:

• statusCode - Number
• headers - Object (key value pairs of strings)
• body - Stream (a stream of binary data)
• requestHeaders - Object (key value pairs of strings)
• requestTimestamp - Number

getResponse should call the callback with an optional error and either null or a cached response object, depending on whether the url can be found in the cache. Only GETs are cached.

setResponse should just swallow any errors it has (or resport them using console.warn).

A cache may also define any of the methods from lib/cache-utils.js to override behaviour for what gets cached. It is currently still only possible to cache "get" requests, although this could be changed.

License

MIT