nise (偽)
fake XHR and Server
This module has been extracted from Sinon.JS and can be used standalone.
Fake XMLHttpRequest
Provides a fake implementation of XMLHttpRequest
and provides
several interfaces for manipulating objects created by it.
Also fakes native XMLHttpRequest
and ActiveXObject
(when available, and only for XMLHTTP
progids). Helps with testing requests made with XHR
.
var fakeXhr = require("nise").fakeXhr;
var sinon = require("sinon");
{
setUp: function () {
this.xhr = fakeXhr.useFakeXMLHttpRequest();
var requests = this.requests = [];
this.xhr.onCreate = function (xhr) {
requests.push(xhr);
};
},
tearDown: function () {
this.xhr.restore();
},
"test should fetch comments from server" : function () {
var callback = sinon.spy();
myLib.getCommentsFor("/some/article", callback);
assertEquals(1, this.requests.length);
this.requests[0].respond(200, { "Content-Type": "application/json" },
'[{ "id": 12, "comment": "Hey there" }]');
assert(callback.calledWith([{ id: 12, comment: "Hey there" }]));
}
}
useFakeXMLHttpRequest
var xhr = fakeXhr.useFakeXMLHttpRequest();
Causes fakeXhr to replace the native XMLHttpRequest
object in browsers that support it with a custom implementation which does not send actual requests.
In browsers that support ActiveXObject
, this constructor is replaced, and fake objects are returned for XMLHTTP
progIds. Other progIds, such as XMLDOM
are left untouched.
The native XMLHttpRequest
object will be available at fakeXhr.xhr.XMLHttpRequest
xhr.onCreate = function (xhr) {};
By assigning a function to the onCreate
property of the returned object from useFakeXMLHttpRequest()
you can subscribe to newly created FakeXMLHttpRequest
objects. See below for the fake xhr object API.
Using this observer means you can still reach objects created by e.g. jQuery.ajax
(or other abstractions/frameworks).
xhr.restore();
Restore original function(s).
FakeXMLHttpRequest
String request.url
The URL set on the request object.
String request.method
The request method as a string.
An object of all request headers, i.e.:
{
"Accept": "text/html, */*",
"Connection": "keep-alive"
}
String request.requestBody
The request body
int request.status
The request's status code.
undefined
if the request has not been handled (see respond
below)
String request.statusText
Only populated if the respond
method is called (see below).
boolean request.async
Whether or not the request is asynchronous.
String request.username
Username, if any.
String request.password
Password, if any.
Document request.responseXML
When using respond
, this property is populated with a parsed document if response headers indicate as much (see the spec)
The value of the given response header, if the request has been responded to (see respond
).
All response headers as an object.
Filtered requests
When using Sinon.JS for mockups or partial integration/functional testing, you might want to fake some requests, while allowing others to go through to the backend server. With filtered FakeXMLHttpRequest
s (new in v1.3.0), you can.
FakeXMLHttpRequest.useFilters
Default false
.
When set to true
, Sinon will check added filters if certain requests should be "unfaked"
FakeXMLHttpRequest.addFilter(fn)
Add a filter that will decide whether or not to fake a request.
The filter will be called when xhr.open
is called, with the exact same arguments (method
, url
, async
, username
, password
). If the filter returns true
, the request will not be faked.
Simulating server responses
request.setStatus(status);
Sets response status (status
and statusText
properties).
Status should be a number, the status text is looked up from fakeXhr.FakeXMLHttpRequest.statusCodes
.
Sets response headers (e.g. { "Content-Type": "text/html", /* ... */ }
, updates the readyState
property and fires onreadystatechange
.
request.setResponseBody(body);
Sets the respond body, updates the readyState
property and fires onreadystatechange
.
Additionally, populates responseXML
with a parsed document if response headers indicate as much.
request.respond(status, headers, body);
Calls the above three methods.
request.error();
Simulates a network error on the request. The onerror
handler will be called and the status
will be 0
.
Boolean request.autoRespond
When set to true
, causes the server to automatically respond to incoming requests after a timeout.
The default timeout is 10ms but you can control it through the autoRespondAfter
property.
Note that this feature is intended to help during mockup development, and is not suitable for use in tests.
Number request.autoRespondAfter
When autoRespond
is true
, respond to requests after this number of milliseconds. Default is 10.
Fake server
High-level API to manipulate FakeXMLHttpRequest
instances.
For help with handling JSON-P please refer to our notes below
var fakeServer = require("nise").fakeServer;
var sinon = require("sinon");
{
setUp: function () {
this.server = fakeServer.create();
},
tearDown: function () {
this.server.restore();
},
"test should fetch comments from server" : function () {
this.server.respondWith("GET", "/some/article/comments.json",
[200, { "Content-Type": "application/json" },
'[{ "id": 12, "comment": "Hey there" }]']);
var callback = sinon.spy();
myLib.getCommentsFor("/some/article", callback);
this.server.respond();
sinon.assert.calledWith(callback, [{ id: 12, comment: "Hey there" }]);
assert(server.requests.length > 0)
}
}
var server = fakeServer.create([config]);
Creates a new server.
This function also calls useFakeXMLHttpRequest()
.
create
accepts optional properties to configure the fake server. See options below for configuration parameters.
var server = fakeServerWithClock.create();
Creates a server that also manages fake timers.
This is useful when testing XHR
objects created with e.g. jQuery 1.3.x, which uses a timer to poll the object for completion, rather than the usual onreadystatechange
.
server.configure(config);
Configures the fake server.
See options below for configuration parameters.
server.respondWith(response);
Causes the server to respond to any request not matched by another response with the provided data. The default catch-all response is [404, {}, ""]
.
response
can be one of three things:
- A
String
or ArrayBuffer
representing the response body - An
Array
with status, headers and response body, e.g. [200, { "Content-Type": "text/html", "Content-Length": 2 }, "OK"]
- A
Function
.
Default status is 200 and default headers are none.
When the response is a Function
, it will be passed the request object. You
must manually call respond on it to complete the
request.
server.respondWith(url, response);
Responds to all requests to given URL, e.g. /posts/1
.
server.respondWith(method, url, response);
Responds to all method
requests to the given URL with the given response.
method
is an HTTP verb.
server.respondWith(urlRegExp, response);
URL may be a regular expression, e.g. /\\/post\\//\\d+
If the response is a Function
, it will be passed any capture groups from the regular expression along with the XMLHttpRequest object:
server.respondWith(/\/todo-items\/(\d+)/, function (xhr, id) {
xhr.respond(200, { "Content-Type": "application/json" }, '[{ "id": ' + id + " }]");
});
server.respondWith(method, urlRegExp, response);
Responds to all method
requests to URLs matching the regular expression.
server.respond();
Causes all queued asynchronous requests to receive a response.
If none of the responses added through respondWith
match, the default response is [404, {}, ""]
.
Synchronous requests are responded to immediately, so make sure to call respondWith
upfront.
If called with arguments, respondWith
will be called with those arguments before responding to requests.
server.autoRespond = true;
If set, will automatically respond to every request after a timeout.
The default timeout is 10ms but you can control it through the autoRespondAfter
property.
Note that this feature is intended to help during mockup development, and is not suitable for use in tests. For synchronous immediate responses, use respondImmediately
instead.
server.autoRespondAfter = ms;
Causes the server to automatically respond to incoming requests after a timeout.
server.respondImmediately = true;
If set, the server will respond to every request immediately and synchronously.
This is ideal for faking the server from within a test without having to call server.respond()
after each request made in that test.
As this is synchronous and immediate, this is not suitable for simulating actual network latency in tests or mockups. To simulate network latency with automatic responses, see server.autoRespond
and server.autoRespondAfter
.
array server.requests
You can inspect the server.requests
to verify request ordering, find unmatched requests or check that no requests has been done.
server.requests
is an array of all the FakeXMLHttpRequest
objects that have been created.
Boolean server.fakeHTTPMethods
If set to true
, server will find _method
parameter in POST body and recognize that as the actual method.
Supports a pattern common to Ruby on Rails applications. For custom HTTP method faking, override server.getHTTPMethod(request)
.
server.getHTTPMethod(request)
Used internally to determine the HTTP method used with the provided request.
By default this method simply returns request.method
. When server.fakeHTTPMethods
is true, the method will return the value of the _method
parameter if the method is "POST".
This method can be overridden to provide custom behavior.
server.restore();
Restores the native XHR constructor.
Fake server options
These options are properties on the server object and can be set directly
server.autoRespond = true
You can also pass options with an object literal to fakeServer.create
and .configure
.
Boolean autoRespond
If set, will automatically respond to every request after a timeout.
The default timeout is 10ms but you can control it through the autoRespondAfter
property.
Note that this feature is intended to help during mockup development, and is not suitable for use in tests.
For synchronous immediate responses, use respondImmediately
instead.
Number autoRespondAfter (ms)
Causes the server to automatically respond to incoming requests after a timeout.
Boolean respondImmediately
If set, the server will respond to every request immediately and synchronously.
This is ideal for faking the server from within a test without having to call server.respond()
after each request made in that test.
As this is synchronous and immediate, this is not suitable for simulating actual network latency in tests or mockups. To simulate network latency with automatic responses, see server.autoRespond
and server.autoRespondAfter
.
boolean fakeHTTPMethods
If set to true
, server will find _method
parameter in POST
body and recognize that as the actual method.
Supports a pattern common to Ruby on Rails applications.
For custom HTTP method faking, override server.getHTTPMethod(request)