For the full story, check out my other question.
Basically, I had asked if it were more efficient to use named functions in the socket handlers for the following code:
var app = require('express').createServer()
var io = require('socket.io').listen(app);
app.listen(8080);
// Some unrelated stuff
io.sockets.on('connection', function (socket) {
socket.on('action1', function (data) {
// logic for action1
});
socket.on('action2', function (data) {
// logic for action2
});
socket.on('disconnect', function(){
// logic for disconnect
});
});
The overall answer was yes (see the above link for more details), but the following ment was posted by ThiefMaster:
I'm not familiar with V8 internals but it might be smart enough to pile the function once and re-use it everytime, just with a different scope attached.
So now that's my question. Is V8 smart enough to pile anonymous functions once and reuse them with different scopes in situations where anonymous functions ordinarily lead to several function instances being created? For example, above I would expect the handler for the connection
event to be created once but the handlers for action1
, action2
, and disconnect
to be created for each connection. In the other question this was solved with named functions but I am more interested if this is necessary in V8 or if it will do some optimizations.
For the full story, check out my other question.
Basically, I had asked if it were more efficient to use named functions in the socket handlers for the following code:
var app = require('express').createServer()
var io = require('socket.io').listen(app);
app.listen(8080);
// Some unrelated stuff
io.sockets.on('connection', function (socket) {
socket.on('action1', function (data) {
// logic for action1
});
socket.on('action2', function (data) {
// logic for action2
});
socket.on('disconnect', function(){
// logic for disconnect
});
});
The overall answer was yes (see the above link for more details), but the following ment was posted by ThiefMaster:
I'm not familiar with V8 internals but it might be smart enough to pile the function once and re-use it everytime, just with a different scope attached.
So now that's my question. Is V8 smart enough to pile anonymous functions once and reuse them with different scopes in situations where anonymous functions ordinarily lead to several function instances being created? For example, above I would expect the handler for the connection
event to be created once but the handlers for action1
, action2
, and disconnect
to be created for each connection. In the other question this was solved with named functions but I am more interested if this is necessary in V8 or if it will do some optimizations.
- 1 Note that this has nothing to do with the function being named or anonymous; it has to do with whether you create a new function on each connection. (See my answer for more on that.) – T.J. Crowder Commented Apr 15, 2012 at 7:47
1 Answer
Reset to default 9Yes. I asked a very similar question (related in my case to creating functions from within a constructor function) on the V8 mailing list. I got the reply that the function's code is "...normally reused...", even though there's a separate function object each time (as required by the spec).
Note, though, that your question has nothing to do with whether the function is named or anonymous. The function in your example could have a name:
io.sockets.on('connection', function handleConnection(socket) {
socket.on('action1', function (data) {
// logic for action1
});
socket.on('action2', function (data) {
// logic for action2
});
socket.on('disconnect', function(){
// logic for disconnect
});
});
That uses a named function expression, which is perfectly valid and handled correctly by V8. (Sadly, it's not handled correctly by IE8 and earlier, which create two pletely different functions at totally different times. But as you're using V8, you don't have to worry about that.)