I have some models with their ActiveQuery defined with some default conditions.
class ProfileQuery extends \yii\db\ActiveQuery{
public function init(){
$this->andOnCondition(['not',[Profile::tableName().'.status'=>2]]);
parent::init();
}
...
The problem is if in another model search I use a join with an alias the init() function of ProfileQuery still tries to search tablename.status
Is there a way to get the current alias while inside the init() function?
This is not possible right now. Limited alias support was problem since the beginning of Yii 2, but recently it was postponed to Yii 3 and most likely will not be fixed in Yii 2: https://github.com/yiisoft/active-record/issues/33
There were some attempts to fix this, but none of them has been finalized and merged. If you really need it, you may use some code from these PRs and implement it yourself in custom component or fork:
https://github.com/yiisoft/yii2/pull/10253
https://github.com/yiisoft/yii2/pull/10813
https://github.com/yiisoft/yii2/pull/11646
Since Yii 2.0.16 you may also use ActiveQuery::getTableNameAndAlias() but it probably will be useless in init().
Related
I'm having a slight problem with CPPDepend's ability to detect virtual function usage. Consider the following scenario. Two classes, CParentClass and CChildClass, where CChildClass is derived from CParentClass. The CParentClass has a virtual function Test and CChildClass overrides the base class version of Test.
When it comes to usage, for various reasons I want to do something like the following:-
CChildClass * pMyChild = new CChildClass();
CParentClass * pParentClass = (CParentClass*)pMyChild;
int B = pParentClass->Test();
delete pParentClass;
This results in pMyChild's Test function being called, as desired, yet CPPDepend doesn't detect this and claims that the code is never reached. If I add the word "virtual" to the Test function header in CChildClass (in addition to the one already in CParentClass) then CPPDepend claims everything is ok.
Can anyone shed some light on this for me please as it feels wrong that I should have to put virtual in the derived class function as well as the base class function.
A similar issue can be seen with CDialog destructors in derived classes. Without the virtual in the derived class destructor declaration, CPPDepend complains.
Thanks for any help you can give.
Regards
Neil.
CppDepend do a static analysis not a dynamic one, and give the dependencies from a static point of view and it's more interesting. Indeed what's important is the dependency related to the design choices, for example in your case the object is declared as CParentClass, so the method is coupled with the contract of CParentClass, and in the runtime it could invoke a method from child classes.
I was reading a tutorial and the user there was starting setters and getters like so
function get f():Number;
function set f(value:Number):void;
inside an interface, then saving it in the main file inside a variable
var testNode:INode;
and referencing them as
testNode.f;
I thought this was really handy, specially for when you have a lot of variables. Instead of having to create two functions for each private value.
However, I tried to do the same without instancing the get and set inside an interface (because I had no need for this) and I get an error saying the function doesn't have a body.
Why is that? Is there any way to write get and set in such a clean, short manner? So far I've been writing these as
public function get someVar():SomeClass {
return _someVar;
}
public function set someVar(newValue:SomeClass):void {
_someVar = newValue;
}
but it's a bit of a hassle when I have several private variables.
An interface is just a model for your class, it does not allows you to skip the definition of a function (or a getter/setter). You must declare the getter and the setter in any class implementing your interface. What you are looking for is an extension (inheritage).
If you define a class Mother.as with a getter and a setter, the class Child.as which extends Mother.as doesn't need to redefine the getter and setter.
Furthermore, if you don't plan on writing anything else inside the getter and setter, you should use a public variable. These are less resource consuming than useless get/set.
I'm looking for an elegant solution to setting the Model argument for a View that is a stage instance.
My View constructor looks like this:
public function View($model:Model, $controller:IController=null){ ... }
My Subclass constructor:
public function ViewSubClass($model:Model, $controller:IController=null){ ... }
The idea is that the subclass will be a UI element that I'd rather just have on stage and not have to position it manually. I've thrown around the possibility of having placeholder elements that get removed and using their positions to attach the real UI elements but it seems a little hackish to me.
I suppose I could always set the default value of the $model argument in View to null, but not all views will be stage instances so I don't exactly want to do that either.
Any thoughts are appreciated.
Flash has no way to populate constructor parameters. This is why I always suggest avoiding constructor parameters for View. If this is your own Class, I'd suggest to go ahead and fix the parent Class so it doesn't need this.
If it is not, you can use
public function VewSubClass() {
super(null, null);
}
I would be concerned here that you actually need the Model and controller, so I would make setters available on the View subclass and then populate those variables once the instance arrives on stage.
public function set model(value:Model):void {
_model = value;
//do whatever the super constructor did based on receiving the model
}
public function set controller(value:IController):void {
_controller = value;
//etc.
}
Note that it probably shouldn't be necessary for the View to know about the controller, and I wouldn't suggest that the entire model be given to the View either--just the few properties it needs. So if the Class is yours, that puts you into a better position to correct these (IMO) architectural problems.
Based on the signature you've provided, I wouldn't be at all surprised to find that the model and controller internal storage are private, rather than protected, which means you're pretty much faced with a rewrite anyway (assuming the model and controller are needed on the View).
I solved the issue by changing the ViewSubClass constructor to:
public function ViewSubClass ($model:Model=null, $controller:IController=null){ ... }
Not sure why I was thinking that the constructor had to match the superclass' constructor perfectly. I'm still open to more elegant solutions if there are any. Thanks for looking.
-Veo
I have a class containing, among other things, a drop down menu. With the aim of saving space, and since the contents of the menu will never change, I've made a static DataProvider for the whole class that populates each instances menu. I was hoping to populate the list with actual functions like so:
tmpArr.push({label:"Details...", funct:openDetailsMenu, args:""});
and then assign tmpArr to the DataProvider. Because the DataProvider is static the function that contains that code also needs to be static, but the functions in the array are non-static. At first it didn't seem like a problem, because when the user clicks on a menu item the drop down menu can call a non-static "executeFunction(funct, args)" on its parent. However, when I try to compile, the static function setting up the DataProvider it can't find the non-static functions being passed. If the compiler would just trust me the code would work fine!
The simple solution is to just pass strings and use a switch statement to call functions based on that, but that's big, ugly, inelegant, and difficult to maintain, especially if something inherits from this class.
The simpler solution is to just make the DataProvider non-static, but I'm wondering if anyone else has a good way of dealing with this? Making the static function able to see its non-static brethren?
Thanks.
OK, the basic reason for making things static is if you want to make it independant of an instance, for example the Math functions in as3 (you call Math.min() as opposed to var math = new Math(); math.min()...) this is useful for reference, repetetive calculation, simple actions (add 10 to x value) etc.
the problem with combining static and non static functionality is that when calling a static function, there is a possibility that the class has no instance at that point, or (in this case) that there is any reference to the function that would make sense in compilation (if a seperate class called the function, how would it reference openDetailsMenu?).
what you need to do is either go through getting function by name (object"functionname" works for example), make annonymous functions in your array or alternatively add a callback method to your static function something similar to this:
public static function doAction(object:Menu, event:String){
if(event == "details") object.openDetailsMenu() ;
}
all in all you are just adding layers of complexity that isnt really going to help. if you just add a class function and get them all to do the same action it is not taking more space or effort than if you are calling to a static function. you need to think about how and why the function is going to be used in (or out of) the class.
you could just store a static reference to the instance, in this case _instance. ( Kind of like a ghetto singleton ) just be careful not to call the static method before the class has been instantiated.
/// in your constructor define a static reference handle to the instance
public function ClassName(){
_instance = this;
}
public static function doSomethingStatic(){
var varValue = ClassName._instance.someInstanceVariable;
}
For example, suppose I this:
class Gundam00 extends Gundam implements MobileSuit {
...
public void fight(final List<MobileSuit> mobiruSuitso, final List<Gundam> theOtherDudes, final List<Person> casualities) {
....
}
}
Suppose theOtherDudes and casualities parameters are optional. How can I make this method as clean as possible? I thought about having booleans indicating if they're null, and then checking them as needed.
I could also have different versions of the method for each combination of parameters but there would be a lot of code duplication I think.
Any suggestions?
I find that past 2-3 arguments, the ability to remember what all the arguments to a function are suffers. And comprehensibility along with it.
Passing named arguments can help. Languages with a convenient hash-like literal syntax make this really easy. Take JavaScript:
g = new Gundam00();
g.fight({opponent: enemy, casualties: 'numerous'});
You can also take advantage of variable length argument features to work this in (treat odd arguments as names, even arguments as the actual parameters).
g.fight('opponent',enemy,'casualties', 'numerous');
And some languages actually support named arguments straight-out (see: http://en.wikipedia.org/wiki/Named_parameter#Use_in_programming_languages ).
Finally, you might want to consider adding other methods for this using what some call a Fluent Interface (http://en.wikipedia.org/wiki/Fluent_interface ). Basically, you've got method call which return the object itself, so you can chain calls together:
g.opponent(enemy).casualties('numerous').fight();
This might be the easiest option if you're working in a manifestly/statically-typed class-focused language.
Update
Responding to Setsuna's comment... in that last example, if you've got the luxury, you can make methods like opponent and casualties simple setters that don't affect any internal state or computation in any other way than setting a parameter for which they're named. They simply set internal properties up, and then all of the real work happens inside action methods like fight.
If you can't do that (or if you don't like writing methods whose operations are sub-atomic), you could stake out a half-way spot between this idea with the hash-like literal idea, and create your own collection class specifically for invoking named arguments:
n = new NArgs();
g.fight(n.arg('opponent',enemy).arg('casualties','numerous').arg('motion','slow'));
A little more unwieldy, but it separates out the named arguments problem and lets you keep your methods a bit more atomic, and NArgs is probably something you could implement pretty easily just wrapping some methods around one type of Collection (HashTable?) or another that's available in your language.
Add the methods. Overloading methods is generally an antipattern and a refactoring opportunity for someone else.
http://www.codinghorror.com/blog/2007/03/curlys-law-do-one-thing.html
I thought about having booleans indicating if they're null, and then checking them inside and reacting accordingly.
Or ... you could just check if they're null.
if(theOtherDudes == null)
...
If there is only one "main method" in your class, then you can implement the optional arguments as getter/setter functions. Example:
public void setOtherDudes(final List<Gundam> theOtherDudes) {} // for input arguments
public List<Person> getCasualities() {} // for output arguments
And then, in your documentation, mention that if the caller has any optional input arguments it has to be passed in before calling fight(), and the optional output values will be available when fight() has been called.
This is worthwhile if there are dozens of optional arguments. Otherwise, I suggest overloading the method as the simplest way.