This seems to be a simple matter and maybe it's solved already, but I'm not sure how to do it. I'd like to keep arbitrary unresolved expressions for later evaluation. Note that I still don't know which expressions are already defined.
For example, suppose I have the expression...
{{source.path}}/mainmenu{{ext}}"
...and the context defines ext as .js, but source.path is still undefined. What I get is/mainmenu.js", but I'd like to get {{source.path}}/mainmenu.js" instead so that I can evaluate {{source.path}} at a later time. HandlebarsConfiguration.UnresolvedBindingFormatter seemed promising, but it doesn't handle the complete original expression. HandlebarsConfiguration.ExpressionNameResolver also didn't help.
So, is it possible to do this at all? Thanks in advance for any help.
In code that I help maintain, I have found multiple examples of code that looks like this:
Description := IfThen(Assigned(Widget), Widget.Description, 'No Widget');
I expected this to crash when Widget was nil, but when I tested it, it worked perfectly.
If I recompile it with "Code inlining control" turned off in Project - Options - Compiler, I do get an Access Violation.
It seems that, because IfThen is marked as inline, the compiler is normally not evaluating Widget.Description if Widget is nil.
Is there any reason that the code should be "fixed", as it doesn't seem to be broken? They don't want the code changed unnecessarily.
Is it likely to bite them?
I have tested it with Delphi XE2 and XE6.
Personally, I hate to rely on a behavior that isn't contractual.
The inline directive is a suggestion to the compiler.
If I understand correctly what I read, your code would also crash if you build using runtime packages.
inlining never occurs across package boundaries
Like Uli Gerhardt commented, it could be considered a bug that it works in the first place. Since the behavior isn't contractual, it can change at any time.
If I was to make any recommendation, I would flag that as a low priority "fix". I'm pretty sure some would argue that if the code works, it doesn't need fixing, there is no bug. At that point, it becomes more of a philosophical question (If a tree falls in a forest and no one is around to hear it, does it make a sound?)
Is there any reason that the code should be "fixed", as it doesn't seem to be broken?
That's really a question that only you can answer. However, to answer it then you need to understand fully the implications of reliance on this behaviour. There are two main issues that I perceive:
Inlining of functions is not guaranteed. The compiler may choose not to inline, and in the case of runtime packages or DLLs, a function in another package cannot be inlined.
Skipping evaluation of an argument only occurs when the compiler is sure that there are no side effects associated with evaluation of the argument. For instance, if the argument involved a function call, the compiler will ensure that it is always evaluated.
To expand on point 2, consider the statement in your question:
Description := IfThen(Assigned(Widget), Widget.Description, 'No Widget');
Now, if Widget.Description is a field, or is a property with a getter that reads a field, then the compiler decides that evaluation has no side effects. This evaluation can safely be skipped.
On the other hand, if Widget.Description is a function, or property with a getter function, then the compiler determines that there may be side effects. And so it ensures that Widget.Description is evaluated exactly once.
So, armed with this knowledge, here are a couple of ways for your code to fail:
You move to runtime packages, or the compiler decides not to inline the function.
You change the Description property getter from a field getter to a function getter.
If it were me, I would not like to rely on this behaviour. But as I said right at the top, ultimately it is your decision.
Finally, the behaviour has been changed from XE7. All arguments to inline functions are evaluated exactly once. This is in keeping with other languages and means that observable behaviour is no longer affected by inlining decisions. I would regard the change in XE7 as a bug fix.
It already has been fixed - in XE7 and confirmed that this was supposed to be wrong behavior.
See https://quality.embarcadero.com/browse/RSP-11531
I'm using the closure compiler to minify and speed up my code but I'm running into some issues with JSLint when I try to export my functions.
Basically, I have an object, foo{} with a function, foo.bar() that gets called via an external file as. In order for this function to be called externally I need to add some declarations to my script before it gets compiled:
window['foo'] = foo;
window['foo']['bar'] = foo.bar;
This works great, but—as ever—JSLint thinks I'm mental for even attempting this. I've managed to suppress the dot notation error by declaring, /*jslint sub: true */ just before these two lines but I still get the following error:
"window['foo']['bar'] = foo.bar;" - Weird assignment
It's not wrong, it is a weird assignment out of context, but I need it in there in order for my code to work.
The way I see it, I have three possible options:
Tell JSLint not to bother even looking at them two lines.
Suppress the Weird assignment error.
Find another way to make my code work with closure compiler.
The problem is, I have no idea how to go about doing any of them.
You can export names using goog.exportSymbol instead of bracket notation: https://github.com/google/closure-library/blob/master/closure/goog/base.js#L1532
The Closure Compiler understands what goog.exportSymbol is so it'll remove the explicit exportSymbol call and add foo and bar directly to the window for you.
CLHS says
An attempt to print a circular structure with *print-circle* set
to nil may lead to looping behavior and failure to terminate.
And then there's this:
Why does this Lisp macro as a whole work, even though each piece doesn't work?
Apparently, having *print-circle* set to nil leads to surprises. Why is *print-circle* set to nil by default on many systems? What can go wrong if I set it to t globally right from the beginning of my code?
If you set *print-circle* to true, then all your output functions have to do cycle checking. That means they may slow down and take more memory.
If you don't actually use circular structures (and I'm not a Lisp pro, but I tend to avoid them like the plague), I wouldn't turn cycle checking on in production code.
I have to ask this, because: The only thing I recognize is, that if the assertion fails, the app crashes. Is that the reason why to use NSAssert? Or what else is the benefit of it? And is it right to put an NSAssert just above any assumption I make in code, like a function that should never receive a -1 as param but may a -0.9 or -1.1?
Assert is to make sure a value is what its supposed to be. If an assertion fails that means something went wrong and so the app quits. One reason to use assert would be if you have some function that will not behave or will create very bad side effects if one of the parameters passed to it is not exactly some value (or a range of values) you can put an assert to make sure that value is what you expect it to be, and if it's not then something is really wrong, and so the app quits. Assert can be very useful for debugging/unit testing, and also when you provide frameworks to stop the users from doing "evil" things.
I can't really speak to NSAssert, but I imagine that it works similarly to C's assert().
assert() is used to enforce a semantic contract in your code. What does that mean, you ask?
Well, it's like you said: if you have a function that should never receive a -1, you can have assert() enforce that:
void gimme_positive_ints(int i) {
assert(i > 0);
}
And now you'll see something like this in the error log (or STDERR):
Assertion i > 0 failed: file example.c, line 2
So not only does it safe-guard against potentially bad inputs but it logs them in a useful, standard way.
Oh, and at least in C assert() was a macro, so you could redefine assert() as a no-op in your release code. I don't know if that's the case with NSAssert (or even assert() any more), but it was pretty useful to compile out those checks.
NSAssert gives you more than just crashing the app. It tells you the class, method, and the line where the assertion occurred. All the assertions can also be easily deactivated using NS_BLOCK_ASSERTIONS. Thus making it more suitable for debugging. On the other hand, throwing an NSException only crashes the app. It also does not tell about the location of the exception, nor can it be disabled so simply. See the difference in the images below.
The app crashes because an assertion also raises an exception, as the NSAssert documentation states:
When invoked, an assertion handler prints an error message that
includes the method and class names (or the function name). It then
raises an NSInternalInconsistencyException exception.
NSAssert:
NSException:
Apart from what everyone said above, the default behaviour of NSAssert() (unlike C’s assert()) is to throw an exception, which you can catch and handle. For instance, Xcode does this.
Just to clarify, as somebody mentioned but not fully explained, the reason for having and using asserts instead of just creating custom code (doing ifs and raising an exception for bad data, for instance) is that asserts SHOULD be disabled for production applications.
While developing and debugging, asserts are enabled for you to catch errors. The program will halt when an assert is evaluated as false.
But, when compiling for production, the compiler omits the assertion code and actually MAKE YOUR PROGRAM RUN FASTER. By then, hopefully, you have fixed all the bugs.
In case your program still has bugs while in production (when assertions are disabled and the program "skips over" the assertions), your program will probably end up crashing at some other point.
From NSAssert's help: "Assertions are disabled if the preprocessor macro NS_BLOCK_ASSERTIONS is defined."
So, just put the macro in your distribution target [only].
NSAssert (and its stdlib equivalent assert) are to detect programming errors during development. You should never have an assertion that fails in a production (released) application. So you might assert that you never pass a negative number to a method that requires a positive argument. If the assertion ever fails during testing, you have a bug. If, however, the value that's passed is entered by the user, you need to do proper validation of the input rather than relying on the assertion in production (you can set a #define for release builds that disables NSAssert*.
Assertions are commonly used to enforce the intended use of a particular method or piece of logic. Let's say you were writing a method which calculates the sum of two greater than zero integers. In order to make sure the method was always used as intended you would probably put an assert which tests that condition.
Short answer: They enforce that your code is used only as intended.
It's worthwhile to point out that aside from run time checking, assert programming is a important facility used when you design your code by contract.
More info on the subject of assertion and design by contract can be found below:
Assertion (software development)
Design by contract
Programming With Assertions
Design by Contract, by Example [Paperback]
To fully answer his question, the point of any type of assert is to aid debugging. It is more valuable to catch errors at their source, then to catch them in the debugger when they cause crashes.
For example, you may pass a value to a function expects values in a certain range. The function may store the value for later use, and on later use the application crashes. The call stack seen in this scenario would not show the source of the bad value. It's better to catch the bad value as it comes in to find out who's passing the bad value and why.
NSAssert make app crash when it match with the condition. If not match with the condition the next statements will execute. Look for the EX below:
I just create an app to test what is the task of NSAssert is:
- (void)viewDidLoad {
[super viewDidLoad];
// Do any additional setup after loading the view, typically from a nib.
[self testingFunction:2];
}
-(void)testingFunction: (int)anNum{
// if anNum < 2 -> the app will crash
// and the NSLog statement will not execute
// that mean you cannot see the string: "This statement will execute when anNum < 2"
// into the log console window of Xcode
NSAssert(anNum >= 2, #"number you enter less than 2");
// If anNum >= 2 -> the app will not crash and the below
// statement will execute
NSLog(#"This statement will execute when anNum < 2");
}
into my code the app will not crash.And the test case is:
anNum >= 2 -> The app will not crash and you can see the log string:"This statement will execute when anNum < 2" into the outPut log console window
anNum < 2 -> The app will crash and you can not see the log string:"This statement will execute when anNum < 2"