Awesome
spyfs
Spy on filesystem calls. Create file system mocks. Use for testing.
Install:
npm install --save spyfs
Create a new file system that spies:
import * as fs from 'fs';
import {spy} from 'spyfs';
const sfs = spy(fs);
Now you can use sfs
for all your filesystem operations.
const data = sfs.readFileSync('./package.json').toString();
Subscribe to all actions happening on that filesystem:
sfs.subscribe(action => {
// ...
});
Every time somebody uses sfs
, the subscription callback will be called.
You will receive a single argument: an action
which is a Promise
object
containing all the information about the performed filesystem operation and its result.
You can also subscribe by providing a listener at creation:
const sfs = spy(fs, action => {
// ...
});
Want to spy on real filesystem?
Overwrite the real fs
module using fs-monkey
to spy on all filesystem
calls:
import {patchFs} from 'fs-monkey';
patchFs(sfs);
Use async/await
spyfs
returns actions which are instances of the Promise
constructor,
so you can use asynchronous functions for convenience:
const sfs = spy(fs, async function(action) {
console.log(await action); // prints directory files...
});
sfs.readdir('/', () => {});
Use with memfs
You can use spyfs
with any fs-like object, including memfs
:
import {fs} from 'memfs';
import {spy} from 'spyfs';
const sfs = spy(fs, async function(action) {
console.log(await action); // bar
});
sfs.writeFile('/foo', 'bar', () => {});
sfs.readFile('/foo', 'utf8', () => {});
Action properties
spyfs
actions have extra properties that tell you more about the action
being executed:
action.method
(string) - name of filesystem method called.action.isAsync
(boolean) - whether the filesystem method called was asynchronous.action.args
(Array) - list of arguments with which the method was called (sans callback).
const sfs = spy(fs, action => {
console.log(action.method); // readdir, readdirSync
console.log(action.isAsync); // true, false
console.log(action.args); // [ '/' ], [ '/' ]
});
sfs.readdir('/', () => {});
sfs.readdirSync('/');
Subscribe to events
The returned filesystem object is also an event emitter, you can subscribe
to specific filesystem actions using the .on()
method, in that case you
will receive only actions for that method:
sfs.on('readdirSync', async function(action) {
console.log(action.args, await action);
});
sfs.readdirSync('/');
Listening for action
event is equivalent to subscribing using .subscribe()
.
sfs.on('action', listener);
sfs.subscribe(listener);
Mock responses
You can overwrite what is returned by the filesystem call at runtime or even throw your custom errors, this way you can mock any filesystem call:
For example, prohibit readFileSync
for /usr/foo/.bashrc
file:
sfs.on('readFileSync', ({args, reject}) => {
if(args[0] === '/usr/foo/.bashrc')
reject(new Error("Cant't touch this!"));
});
Sync mocking
action.resolve(result)
Returns to the user result
as successfully executed action, below
all operations readFileSync
will return '123'
:
sfs.on('readFileSync', action => {
action.resolve('123');
});
action.reject(error)
Throws error
:
sfs.on('statSync', action => {
action.reject(new Error('This filesystem does not support stat'));
});
action.exec()
Executes an action user was intended to perform and returns back result only to you. This method can throw.
sfs.on('readFileSync', action => {
const result = action.exec();
if(result.length > 100) action.reject(new Error('File too long'));
});
action.result
result
is a reference to the action
for your convenience:
Async mocking
Just like sync mocking actions support resolve
, reject
and exec
methods,
but, in addition, async mocking also has pause
and unpause
methods.
action.resolve(results)
Successfully executes user's filesystem call. results
is an array, because
some Node's async filesystem calls return more than one result.
sfs.on('readFile', ({resolve}) => {
resolve(['123']);
});
action.reject(error)
Fails filesystem call and returns your specified error.
sfs.on('readFile', ({reject}) => {
reject(new Error('You cannot touch this file!'));
});
action.pause()
Pauses the async filesystem call until you un-pause it.
sfs.on('readFile', ({pause}) => {
pause();
// The readFile operation will never end,
// if you don't unpause it.
});
Pausing is useful if you want to perform some other async IO before yielding back to the original filesystem operation.
action.unpause()
Un-pauses previously pauses filesystem operation:
sfs.on('readFile', ({pause, unpause}) => {
// This effectively does nothing:
pause();
unpause();
});
action.exec()
Executes user's intended filesystem call and returns the result only to you.
Unlike the sync version exec
, async exec
returns a promise.
You should use it together with pause()
and unpause()
.
sfs.on('readFile', ({exec, pause, unpause, reject}) => {
pause();
exec().then(result => {
if(result.length < 100) {
reject(new Error('File too small'));
} else {
unpause();
}
});
});
Use async/await
with exec()
:
sfs.on('readFile', async function({exec, pause, unpause, reject}) {
pause();
let result = await exec();
if(result.length < 100) {
reject(new Error('File too small'));
} else {
unpause();
}
});
action.result
result
is a reference to the action
for your convenience:
Spy
constructor
Create spying filesystems manually:
import {Spy} from 'spyfs';
new Spy(fs[, listener])
fs
is the file system to spy on. Note that Spy
will not overwrite or
in any way modify your original filesystem, but rather it will create a
new object for you.
listener
is an optional callback that will be set using the .subscribe()
method, see below.
sfs.subscribe(listener)
Subscribe to all filesystem actions:
const sfs = new Spy(fs);
sfs.subscribe(action => {
});
It is equivalent to calling sfs.on('action', listener)
.
sfs.unsubscribe(listener)
Unsubscribes your listener. It is equivalent to calling sfs.off('action', listener)
.
sfs.on(method, listener)
Subscribes to a specific filesystem call.
sfs.on('readFile', listener);
sfs.off(method, listener)
Unsubscribes from a specific filesystem call.
License
This is free and unencumbered software released into the public domain.
Anyone is free to copy, modify, publish, use, compile, sell, or distribute this software, either in source code form or as a compiled binary, for any purpose, commercial or non-commercial, and by any means.
In jurisdictions that recognize copyright laws, the author or authors of this software dedicate any and all copyright interest in the software to the public domain. We make this dedication for the benefit of the public at large and to the detriment of our heirs and successors. We intend this dedication to be an overt act of relinquishment in perpetuity of all present and future rights to this software under copyright law.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
For more information, please refer to http://unlicense.org/