I've created an interprter for a simple language. It is AST based (to be more exact, an irregular heterogeneous AST) with visitors executing and evaluating nodes. However I've noticed that it is extremely slow compared to "real" interpreters. For testing I've ran this code:
i = 3
j = 3
has = false
while i < 10000
j = 3
has = false
while j <= i / 2
if i % j == 0 then
has = true
end
j = j+2
end
if has == false then
puts i
end
i = i+2
end
In both ruby and my interpreter (just finding primes primitively). Ruby finished under 0.63 second, and my interpreter was over 15 seconds.
I develop the interpreter in C++ and in Visual Studio, so I've used the profiler to see what takes the most time: the evaluation methods.
50% of the execution time was to call the abstract evaluation method, which then casts the passed expression and calls the proper eval method. Something like this:
Value * eval (Exp * exp)
{
switch (exp->type)
{
case EXP_ADDITION:
eval ((AdditionExp*) exp);
break;
...
}
}
I could put the eval methods into the Exp nodes themselves, but I want to keep the nodes clean (Terence Parr saied something about reusability in his book).
Also at evaluation I always reconstruct the Value object, which stores the result of the evaluated expression. Actually Value is abstract, and it has derived value classes for different types (That's why I work with pointers, to avoid object slicing at returning). I think this could be another reason of slowness.
How could I make my interpreter as optimized as possible? Should I create bytecodes out of the AST and then interpret bytecodes instead? (As far as I know, they could be much faster)
Here is the source if it helps understanding my problem: src
Note: I haven't done any error handling yet, so an illegal statement or an error will simply freeze the program. (Also sorry for the stupid "error messages" :))
The syntax is pretty simple, the currently executed file is in OTZ1core/testfiles/test.txt (which is the prime finder).
I appreciate any help I can get, I'm really beginner at compilers and interpreters.
One possibility for a speed-up would be to use a function table instead of the switch with dynamic retyping. Your call to the typed-eval is going through at least one, and possibly several, levels of indirection. If you distinguish the typed functions instead by name and give them identical signatures, then pointers to the various functions can be packed into an array and indexed by the type member.
value (*evaltab[])(Exp *) = { // the order of functions must match
Exp_Add, // the order type values
//...
};
Then the whole switch becomes:
evaltab[exp->type](exp);
1 indirection, 1 function call. Fast.
I read a coding style suggestion about comparing bools that said to write
if (myBoolValue == true) { ... }
if (myBoolValue == false) { ... }
instead of writing
if (myBoolValue) { ... }
if (!myBoolValue) { ... }
since it increases readability of the code even though they are equivalent statements. I am well aware that it is not a usual coding practice but I will agree that it may increase readability in some cases.
My question is if there is any difference between the two in regards of optimization of code execution or if (a well implemented) compiler translate them to the same thing?
The productions are not the same in all languages.
For instance, they may produce different results for some "non-boolean" values of "myBoolValue" in both JavaScript and C.
// JavaScript
[] == true // false
[] ? true : false // true
// C
#define true 1
#define false 0
int x = -1;
x == true // 0 (false)
x ? true : false // true (1)
To see what a specific compiler does for a specific programming language (there is both what it is allowed to do and then what it will do), check the generated assembly/machine/byte code.
(Anyway, I prefer and use the latter form exclusively.)
Apart from cases like JavaScript and I guess Java with autoboxing and Boolean, where the semantics can differ between the two forms, it really depends on compiler optimization.
I never understand this recommendation. If myBool is a boolean, so is (myBool == true), so why doesn't it require ((myBool == true) == true), and so on forever? Where does this stop? Surely the answer is not to start in the first place?
Surely the more there is to read, the less readable it is?
In C (and many other languages) there's no boolean type, all are numeric expression, so myBoolValue may be equal to 2 or any value different from 0 and 1, and comparing them numerically won't give you the correct result.
#define true 1
#define false 0
if (5 == true) //...
To improve readability, name the variable meaningful (such as isNonZero, isInGoodState, isClosed, LoopEnded...) rather than comparing them with true
Another way is comparing them with 0 or false when you need the true case (myBoolValue != false). Also, using bool type (if it's available) may solve some problems.
The behavior depends on the language and on the type of the value being tested. The two forms are not equivalent in all cases.
In C, conditions are commonly represented using integers, with 0 being treated as false and any non-zero value being treated as true. Comparing such a value for equality to true is invalid; for example, if the value happens to be 2 or -1, it's logically true but not equal to true. As of C99, C does have a built-in boolean type called _Bool, aliased as bool if you have #include <stdbool.h>. But it's still common to use other types, particularly int to represent conditions, and even to redefine bool, false, and true in various ways.
C++ has has a built-in bool type, with false and true literals, since very early in its development, but it still retains much of its C ancestry. Other C-inspired languages are likely to have similar issues.
In languages that have a built-in boolean type and don't allow values of other types to be used directly as conditions, if (x) and if (x == true) are more likely to be equivalent. And if they're semantically equivalent, I would expect any decent optimizing compiler to generate the same code for either form, or very nearly so.
But I strongly disagree with the style advice. An explicit comparison to false or true, even in cases where it's equivalent to testing the value directly, do not IMHO aid readability.
If you have a variable whose value denotes a condition, it's more important to give it a name that indicates that. Programming language code shouldn't necessarily follow English grammar, but something like
if (file_is_open)
is not made more readable by changing it to
if (file_is_open == true)
As a C programmer, I like to write my conditions more explicitly than a lot of other programmers do; I'll write if (ptr != NULL) rather than if (ptr), and if (c != '\0') rather than if (c). But if a variable is already a condition, I see no point in adding a superfluous comparison.
And value == true is itself a condition; if if (value == true) is supposed to be more readable than if (value), why wouldn't if ((value == true) == true) be even better?
There's a certain over-verbosity that I have to engage in when writing certain Boolean expressions, at least with all the languages I've used, and I was wondering if there were any languages that let you write more concisely?
The way it goes is like this:
I want to find out if I have a Thing that can be either A, B, C, or D.
And I'd like to see if Thing is an A or a B.
The logical way for me to express this is
//1: true if Thing is an A or a B
Thing == (A || B)
Yet all the languages I know expect it to be written as
//2: true if Thing is an A or a B
Thing == A || Thing == B
Are there any languages that support 1? It doesn't seem problematic to me, unless Thing is a Boolean.
Yes. Icon does.
As a simple example, here is how to get the sum of all numbers less than 1000 that are divisble by three or five (the first problem of Project Euler).
procedure main ()
local result
local n
result := 0
every n := 1 to 999 do
if n % (3 | 5) == 0 then
result +:= n
write (result)
end
Note the n % (3 | 5) == 0 expression. I'm a bit fuzzy on the precise semantics, but in Icon, the concept of booleans is not like other languages. Every expression is a generator and it may pass (generating a value) or fail. When used in an if expression, a generator will continue to iterate until it passes or exhausts itself. In this case, n % (3 | 5) == 0 is a generator which uses another generator (3 | 5) to test if n is divisible by 3 or 5. (To be entirely technical, this isn't even syntactic sugar.)
Likewise, in Python (which was influenced by Icon) you can use the in statement to test for equality on multiple elements. It's a little weaker than Icon though (as in, you could not translate the modulo comparison above directly). In your case, you would write Thing in (A, B), which translates exactly to what you want.
There are other ways to express that condition without trying to add any magic to the conditional operators.
In Ruby, for example:
$> thing = "A"
=> "A"
$> ["A","B"].include? thing
=> true
I know you are looking for answers that have the functionality built into the language, but here are two other means that I find work better as they solve more problems and have been in use for many decades.
Have you considered using a preprocessor?
Also languages like Lisp have macros which is part of the language.
This question already has answers here:
Closed 12 years ago.
Possible Duplicate:
Why do most programming languages only have binary equality comparison operators?
I have had a simple question for a fairly long time--since I started learning programming languages.
I'd like to write like "if x is either 1 or 2 => TRUE (otherwise FALSE)."
But when I write it in a programming language, say in C,
( x == 1 || x == 2 )
it really works but looks awkward and hard to read. I guess it should be possible to simplify such an or operation, and so if you have any idea, please tell me. Thanks, Nathan
Python allows test for membership in a sequence:
if x in (1, 2):
An extension version in C#
step 1: create an extension method
public static class ObjectExtensions
{
public static bool Either(this object value, params object[] array)
{
return array.Any(p => Equals(value, p));
}
}
step 2: use the extension method
if (x.Either(1,2,3,4,5,6))
{
}
else
{
}
While there are a number of quite interesting answers in this thread, I would like to point out that they may have performance implications if you're doing this kind of logic inside of a loop depending on the language. A far as for the computer to understand, the if (x == 1 || x == 2) is by far the easiest to understand and optimize when it's compiled into machine code.
When I started programming it seemed weird to me as well that instead of something like:
(1 < x < 10)
I had to write:
(1 < x && x < 10)
But this is how most programming languages work, and after a while you will get used to it.
So I believe it is perfectly fine to write
( x == 1 || x == 2 )
Writing it this way also has the advantage that other programmers will understand easily what you wrote. Using a function to encapsulate it might just make things more complicated because the other programmers would need to find that function and see what it does.
Only more recent programming languages like Python, Ruby etc. allow you to write it in a simpler, nicer way. That is mostly because these programming languages are designed to increase the programmers productivity, while the older programming languages' main goal was application performance and not so much programmer productivity.
It's Natural, but Language-Dependent
Your approach would indeed seem more natural but that really depends on the language you use for the implementation.
Rationale for the Mess
C being a systems programming language, and fairly close to the hardware (funny though, as we used to consider a "high-level" language, as opposed to writing machine code), it's not exactly expressive.
Modern higher-level languages (again, arguable, lisp is not that modern, historically speaking, but would allow you to do that nicely) allow you to do such things by using built-in constructs or library support (for instances, using Ranges, Tuples or equivalents in languages like Python, Ruby, Groovy, ML-languages, Haskell...).
Possible Solutions
Option 1
One option for you would be to implement a function or subroutine taking an array of values and checking them.
Here's a basic prototype, and I leave the implementation as an exercise to you:
/* returns non-zero value if check is in values */
int is_in(int check, int *values, int size);
However, as you will quickly see, this is very basic and not very flexible:
it works only on integers,
it works only to compare identical values.
Option 2
One step higher on the complexity ladder (in terms of languages), an alternative would be to use pre-processor macros in C (or C++) to achieve a similar behavior, but beware of side effects.
Other Options
A next step could be to pass a function pointer as an extra parameter to define the behavior at call-point, define several variants and aliases for this, and build yourself a small library of comparators.
The next step then would be to implement a similar thing in C++ using templates to do this on different types with a single implementation.
And then keep going from there to higher-level languages.
Pick the Right Language (or learn to let go!)
Typically, languages favoring functional programming will have built-in support for this sort of thing, for obvious reasons.
Or just learn to accept that some languages can do things that others cannot, and that depending on the job and environment, that's just the way it is. It mostly is syntactic sugar, and there's not much you can do. Also, some languages will address their shortcomings over time by updating their specifications, while others will just stall.
Maybe a library implements such a thing already and that I am not aware of.
that was a lot of interesting alternatives. I am surprised nobody mentioned switch...case - so here goes:
switch(x) {
case 1:
case 2:
// do your work
break;
default:
// the else part
}
it is more readable than having a
bunch of x == 1 || x == 2 || ...
more optimal than having a
array/set/list for doing a
membership check
I doubt I'd ever do this, but to answer your question, here's one way to achieve it in C# involving a little generic type inference and some abuse of operator overloading. You could write code like this:
if (x == Any.Of(1, 2)) {
Console.WriteLine("In the set.");
}
Where the Any class is defined as:
public static class Any {
public static Any2<T> Of<T>(T item1, T item2) {
return new Any2<T>(item1, item2);
}
public struct Any2<T> {
T item1;
T item2;
public Any2(T item1, T item2) {
this.item1 = item1;
this.item2 = item2;
}
public static bool operator ==(T item, Any2<T> set) {
return item.Equals(set.item1) || item.Equals(set.item2);
}
// Defining the operator== requires these three methods to be defined as well:
public static bool operator !=(T item, Any2<T> set) {
return !(item == set);
}
public override bool Equals(object obj) { throw new NotImplementedException(); }
public override int GetHashCode() { throw new NotImplementedException(); }
}
}
You could conceivably have a number of overloads of the Any.Of method to work with 3, 4, or even more arguments. Other operators could be provided as well, and a companion All class could do something very similar but with && in place of ||.
Looking at the disassembly, a fair bit of boxing happens because of the need to call Equals, so this ends up being slower than the obvious (x == 1) || (x == 2) construct. However, if you change all the <T>'s to int and replace the Equals with ==, you get something which appears to inline nicely to be about the same speed as (x == 1) || (x == 2).
Err, what's wrong with it? Oh well, if you really use it a lot and hate the looks do something like this in c#:
#region minimizethisandneveropen
public bool either(value,x,y){
return (value == x || value == y);
}
#endregion
and in places where you use it:
if(either(value,1,2))
//yaddayadda
Or something like that in another language :).
In php you can use
$ret = in_array($x, array(1, 2));
As far as I know, there is no built-in way of doing this in C. You could add your own inline function for scanning an array of ints for values equal to x....
Like so:
inline int contains(int[] set, int n, int x)
{
int i;
for(i=0; i<n; i++)
if(set[i] == x)
return 1;
return 0;
}
// To implement the check, you declare the set
int mySet[2] = {1,2};
// And evaluate like this:
contains(mySet,2,x) // returns non-zero if 'x' is contained in 'mySet'
In T-SQL
where x in (1,2)
In COBOL (it's been a long time since I've even glanced briefly at COBOL, so I may have a detail or two wrong here):
IF X EQUALS 1 OR 2
...
So the syntax is definitely possible. The question then boils down to "why is it not used more often?"
Well, the thing is, parsing expressions like that is a bit of a bitch. Not when standing alone like that, mind, but more when in compound expressions. The syntax starts to become opaque (from the compiler implementer's perspective) and the semantics downright hairy. IIRC, a lot of COBOL compilers will even warn you if you use syntax like that because of the potential problems.
In .Net you can use Linq:
int[] wanted = new int{1, 2};
// you can use Any to return true for the first item in the list that passes
bool result = wanted.Any( i => i == x );
// or use Contains
bool result = wanted.Contains( x );
Although personally I think the basic || is simple enough:
bool result = ( x == 1 || x == 2 );
Thanks Ignacio! I translate it into Ruby:
[ 1, 2 ].include?( x )
and it also works, but I'm not sure whether it'd look clear & normal. If you know about Ruby, please advise. Also if anybody knows how to write this in C, please tell me. Thanks. -Nathan
Perl 5 with Perl6::Junction:
use Perl6::Junction 'any';
say 'yes' if 2 == any(qw/1 2 3/);
Perl 6:
say 'yes' if 2 == 1|2|3;
This version is so readable and concise I’d use it instead of the || operator.
Pascal has a (limited) notion of sets, so you could do:
if x in [1, 2] then
(haven't touched a Pascal compiler in decades so the syntax may be off)
A try with only one non-bitwise boolean operator (not advised, not tested):
if( (x&3) ^ x ^ ((x>>1)&1) ^ (x&1) ^ 1 == 0 )
The (x&3) ^ x part should be equal to 0, this ensures that x is between 0 and 3. Other operands will only have the last bit set.
The ((x>>1)&1) ^ (x&1) ^ 1 part ensures last and second to last bits are different. This will apply to 1 and 2, but not 0 and 3.
You say the notation (x==1 || x==2) is "awkward and hard to read". I beg to differ. It's different than natural language, but is very clear and easy to understand. You just need to think like a computer.
Also, the notations mentioned in this thread like x in (1,2) are semantically different then what you are really asking, they ask if x is member of the set (1,2), which is not what you are asking. What you are asking is if x equals to 1 or to 2 which is logically (and semantically) equivalent to if x equals to 1 or x equals to 2 which translates to (x==1 || x==2).
In java:
List list = Arrays.asList(new Integer[]{1,2});
Set set = new HashSet(list);
set.contains(1)
I have a macro that I use a lot that's somewhat close to what you want.
#define ISBETWEEN(Var, Low, High) ((Var) >= (Low) && (Var) <= (High))
ISBETWEEN(x, 1, 2) will return true if x is 1 or 2.
Neither C, C++, VB.net, C#.net, nor any other such language I know of has an efficient way to test for something being one of several choices. Although (x==1 || x==2) is often the most natural way to code such a construct, that approach sometimes requires the creation of an extra temporary variable:
tempvar = somefunction(); // tempvar only needed for 'if' test:
if (tempvar == 1 || tempvar == 2)
...
Certainly an optimizer should be able to effectively get rid of the temporary variable (shove it in a register for the brief time it's used) but I still think that code is ugly. Further, on some embedded processors, the most compact and possibly fastest way to write (x == const1 || x==const2 || x==const3) is:
movf _x,w ; Load variable X into accumulator
xorlw const1 ; XOR with const1
btfss STATUS,ZERO ; Skip next instruction if zero
xorlw const1 ^ const2 ; XOR with (const1 ^ const2)
btfss STATUS,ZERO ; Skip next instruction if zero
xorlw const2 ^ const3 ; XOR with (const2 ^ const3)
btfss STATUS,ZERO ; Skip next instruction if zero
goto NOPE
That approach require two more instructions for each constant; all instructions will execute. Early-exit tests will save time if the branch is taken, and waste time otherwise. Coding using a literal interpretation of the separate comparisons would require four instructions for each constant.
If a language had an "if variable is one of several constants" construct, I would expect a compiler to use the above code pattern. Too bad no such construct exists in common languages.
(note: Pascal does have such a construct, but run-time implementations are often very wasteful of both time and code space).
return x === 1 || x === 2 in javascript
If I have a Boolean expression to check
(A && B)
If A is found to be false will the language bother to check B? Does this vary from language to language?
The reason I ask is that I'm wondering if it's the case that B is checked even if A is false then wouldn't
if (A) {
if(B) {
} else {
// code x
}
} else {
// code x
}
be marginally quicker than
if (A && B) {
} else {
// code x
}
This depends on the language. Most languages will implement A && B as a short-circuit operator, meaning that if A evaluates to false, B will never be evaluated. There's a detailed list on Wikipedia.
Almost every language implements something called short-circuit evaluation, which means that yes, (A && B) will not evaluate B if A is false. This also takes effect if you write:
if (A || B) {
...
}
and A is true. This is worth remembering if B may take a long time to load, but generally it's not something to worry about.
As a bit of history, in my mind this is a bit of a sore part of LISP because code like this:
(if (and (= x 5) (my-expensive-query y)) "Yes" "No")
is not made of functions, but rather so-called "special forms" (that is, "and" could not be defun'd here).
This would depend 100% on how the language compiles said code. Anything is possible :)
Is there a specific language you're wondering about?
In short, no. A double branch involves various forms of branch prediction. If A and B are simple to evaluate, it may be faster to do if (A && B) in a non-short-circuit way than if (A) if (B). In addition, you've duplicated code in the second form. This is virtually always (exception to every rule .. I guess) bad and far worse than any gain.
Secondly, this is the kind of micro-optimization that you give to the language interpreter, JIT or compiler.
Many languages (including almost all curly-brace languages, like C/C++/Java/C#) offer short-circuit boolean evaluation. In those languages, if A is false then B won't be evaluated. You'll need to see (or ask) whether this is true for your specific language, or whether there's a way to do it (VB has AndAlso, for example).
If you find your language doesn't support it, you'll also need to consider whether the cost of evaluating B is worth having to maintain two identical pieces of code -- and the potential doubling in cache footprint (not to mention all the extra branching) that'd come from doing that duplication every time.
As others have said, it depends on the language and/or compiler. For me, I don't care how fast or slow it might be to short-circuit or not, the need to duplicate code is a deal-killer.
If A and B are actually calls that have side-effects (i.e. they do more than simply return a value suitable for comparison), then I would argue that those calls should be made into variable assignments that are then used in your comparison. It doesn't matter whether or not you always require those side-effects or only require them conditionally, the code will be more readable if you don't depend on whether or not short-circuit exists.
That last bit about readability is based on my feeling that reducing the need to refer to external documentation improves readability. Reading a book with a bunch of new words that require dictionary look-ups is much more challenging than reading that same book when you already have the necessary vocabulary. In this case, short-circuit is invisible, so anybody that needs to look it up won't even know that they need to look it up.