with() Problem/Bug
Categories: Flash
Table of Contents
This is yet another bug I found while doing some Flash-dev, it has to do with the with() statement.
If you use with(this){…} it will give you erratic/unpredicted behavior, for instance take this code:
ob = {};
ob.test = function() {
with(this) {
__x = 1;
trace("test: inside: "+__x);
}
trace("test: outside: "+this.__x);
}
ob.test();
You would think that output would look like this:
test: inside: 1 test: outside: 1
But, instead it comes out as:
test: inside: 1 test: outside: undefined
Seems as though Flash does not allow you to set variables when using this in a with() statement. You can get the .fla i used to make the test cases for this bug here.
Thanks Macromedia for wasting another hour of my time! 😛
You could have saved yourself an hour by reading the reference information for the with() statement…lol.
“To set a variable inside a with statement, you must have declared the variable outside the with statement, or you must enter the full path to the Timeline on which you want the variable to live. If you set a variable in a with statement without declaring it, the with statement will look for the value according to the scope chain. If the variable doesn’t already exist, the new value will be set on the Timeline from which the with statement was called.”
BTW: This works…
with (this) {
this.__x = 1;
trace(“inside”+__x);
}
trace(“outside”+this.__x);
Because of the scope of how you define the variables. Since __x isn’t defined, using a with(this) statement doesn’t define it within the object itself it is instead defining it within the timeline scope of the ‘this’ object.
Example in your test file…
You have…
ob3.test2 = function() {
with (this) {
__x = 50;
trace(“test2: inside “+__x);
}
trace(“test2: outside “+this.__x);
};
Now, remove __x = 50 from that statement. You will see it will return 1 now. Because you set it to 1 in the first object test() prototype.
Bah, I’m just babbling, but in essense, it isn’t a bug at all, it’s just how Flash works with the with() statement.
Yeah, I guess your right. I should of looked at documentation for it.
I would consider this a bug though.