Enumerate possible set values in Delphi - delphi

I have a calculation algorithm in Delphi with a number of different options, and I need to try every combination of options to find an optimal solution.
TMyOption = (option1, option2, option3, option4);
TMyOptions = set of TMyOption;
I wondered about using an Integer loop to enumerate them:
for EnumerationInteger := 0 to 15 do begin
Options := TMyOptions(EnumerationInteger);
end;
This does not compile. What I was wondering was if there was any fairly simple method to convert from Integer to Set (most questions on the Web try to go the other way, from Set to Integer), and if so what is it?
Another possibility is to just use the Integer as a bit-field:
C_Option1 = 1;
C_Option2 = 2;
C_Option3 = 4;
C_Option4 = 8;
and then test membership with a bitwise and:
if (Options and C_Option2) > 0 then begin
...
end;
I've tried this, and it works, but it feels like working with sets would be more natural and use the type system better (even though I'm going outside the said type system to enumerate the sets).
Is there a better/safer way to enumerate all possible set combinations than enumerating the underlying integer representation?
Notes:
I know that the integer values of a set are not guaranteed in theory (though I suspect they are in practice if you don't play with the enumeration numbering).
There could be more than four options (yes, I know that it grows exponentially and if there are too many options the algorithm could take forever).

I know this question is quite old, but this is my preference since it's simple and natural to me :
function NumericToMyOptions(n: integer): TMyOptions;
var
Op: TMyOption;
begin
Result:= [];
for Op:= Low(TMyOption) to High(TMyOption) do
if n and (1 shl ord(Op)) > 0 then Include(Result, Op);
end;

Try
var EnumerationByte: Byte;
...
for EnumerationByte := 0 to 15 do begin
Options := TMyOptions(EnumerationByte);
end;

Your code does not compile because your enumeration (TMyOption) have less than 8 values, and Delphi utilize the minimum possible size (in bytes) for sets. Thus, a byte variable will work for you.
If you have a set with more than 8 but less than 16 possible elements, a Word will work (and not an integer).
For more than 16 but less than 32 a DWord variable and typecast.
For more than 32 possible elements, I think a better approach is to use an array of bytes or something like that.

500 - Internal Server Error's answer is probably the most simple.
Another approach that would less likely to break with changes to the number of options would be to declare an array of boolean, and switch them on/off. This is slower than working with pure integers though. The main advantage, you won't need to change the integer type you use, and you can use it if you have more than 32 options.
procedure DoSomething
var BoolFlags : Array[TOption] of Boolean;
I: TOption;
function GetNextFlagSet(var Bools : Array of Boolean) : Boolean;
var idx, I : Integer;
begin
idx := 0;
while Bools[idx] and (idx <= High(Bools)) do Inc(idx);
Result := idx <= High(Bools);
if Result then
for I := 0 to idx do
Bools[I] := not Bools[I];
end;
begin
for I := Low(BoolFlags) to High(BoolFlags) do BoolFlags[i] := False;
repeat
if BoolFlags[Option1] then
[...]
until not GetNextFlagSet(BoolFlags);
end;

Casting from an Integer to a Set is not possible, but Tondrej once wrote a blog article on SetToString and StringToSet that exposes what you want in the SetOrdValue method:
uses
TypInfo;
procedure SetOrdValue(Info: PTypeInfo; var SetParam; Value: Integer);
begin
case GetTypeData(Info)^.OrdType of
otSByte, otUByte:
Byte(SetParam) := Value;
otSWord, otUWord:
Word(SetParam) := Value;
otSLong, otULong:
Integer(SetParam) := Value;
end;
end;
Your code then would become this:
for EnumerationInteger := 0 to 15 do begin
SetOrdValue(TypeInfo(TMyOptions), Options, EnumerationInteger);
end;
--jeroen

The problem is that you are trying to cast to the set type instead of the enumerated type. You can cast between integer and enumerated because both are ordinal types, but you can't cast to a set because they use bitfiels as you already noted. If you use:
for EnumerationInteger := 0 to 15 do begin
Option := TMyOption(EnumerationInteger);
end;
it would work, although is not what you want.
I had this same problem a few months ago and came to the conclusion that you can't enumerate the contents of a set in Delphi (at least in Delphi 7) because the language doesn't define such operation on a set.
Edit: It seems that you can even in D7, see coments to this answer.

Related

Is there any difference between Exit(1) or Result := 1; Exit in Delphi?

In Delphi, you can Exit a function and give it a return value which is very similar to the return keyword in C/C++/Java/C# etc.
Exit(1);
But, I usually write something like this:
Result := 1;
Exit;
Any reasons to choose one over another?
These two variants are semantically identical and you are free to choose between them.
If you need code to compile on older versions of the compiler which do not support the Exit(<value>) syntax then clearly you would have to avoid that variant.
Personally I avoid Exit(<value>) simply because I don't like there being two different ways to assign a return value. I hate functions like this:
function FindValue(Value: Integer): Integer;
var
i: Integer;
begin
for i := 0 to Count - 1 do
if Items[i] = Value then
Exit(i);
Result := -1;
end;
Here we mix the two different approaches. Obviously we could opt to use Exit(<value>) at all times, but then I could not write that function like so:
function FindValue(Value: Integer): Integer;
begin
for Result := 0 to Count - 1 do
if Items[Result] = Value then
Exit;
Result := -1;
end;
In my opinion, this is one of those times where an enhancement has been added that introduces programmer choice but yields very little benefit. I would have preferred for Exit(<value>) not to have been introduced. However, as I said, these are my personal opinions and I'm sure other people have different views.

Delphi - Calculate Minimum Maximum and Average in items in a list

I am building a single application to Calculate Min Max and Avg of Values in a List.
It is actually Temperatures. So I think I am Almost correct but there are 2 Errors.
var
Count, Average, Sum,i, Max, Min, K : Integer;
Temperatures : Array of Integer;
NoItems : Double;
begin
Count := 0;
Sum := 0;
Max := 0;
Min := 0;
Average := 0;
Count := lstTemp.Items.Count;
{Calculate Sum of Values in the list}
for i := 0 to Count - 1 do
Sum := Sum + StrToInt(lstTemp.Items[i]);
{Calculate Min and Max}
SetLength(Temperatures,Count);
for K:=0 to Count-1 do
Temperatures[K] := lstTemp.Items[K];
if (Temperatures[K] > Max) then
Max := Temperatures[K];
if (Temperatures[K] < Min) then
Min := Temperatures[K];
{Calculate Average}
Average := Sum / Count;
edtAvg.Text:=IntToStr(Average); //Display Average
edtAvg.Text:=IntToStr(Min); //Display Minimum Temp.
edtAvg.Text:=IntToStr(Max); //Display Maximum Temp.
end;
So the 2 Errors are
Error: Incompatible types: got "AnsiString" expected "LongInt"
This is for Average := Sum / Count;
Error: Incompatible types: got "Set Of Byte" expected "Double"
This Error is for Temperatures[K] := lstTemp.Items[K];
Any Ideas how to solve this?
Sum and Count are both Integers so I dont know why it shouldnt work!
Thanks
There is a number of problems. First, when you write
for K:=0 to Count-1 do
Temperatures[K] := lstTemp.Items[K];
if (Temperatures[K] > Max) then
Max := Temperatures[K];
if (Temperatures[K] < Min) then
Min := Temperatures[K];
you actually do
for K:=0 to Count-1 do
Temperatures[K] := lstTemp.Items[K];
if (Temperatures[K] > Max) then
Max := Temperatures[K];
if (Temperatures[K] < Min) then
Min := Temperatures[K];
which is nonsense. You want all these lines to be part of the for loop:
for K:=0 to Count-1 do
begin
Temperatures[K] := lstTemp.Items[K];
if (Temperatures[K] > Max) then
Max := Temperatures[K];
if (Temperatures[K] < Min) then
Min := Temperatures[K];
end;
Second, in order for this algorithm to work, the initial value of Min (Max) needs to be larger (smaller) than the values in the list. This might work for Max := 0, but probably not for Min := 0. You need to set Min to a very large value before you run the loop, obviously. The best value you can use is the highest-possible signed 32-bit integer value, that is, 2^31 - 1, which is the value of the MaxInt constant.
Third,
Temperatures[K] := lstTemp.Items[K];
is probably wrong. Temperatures is an array of integers, while lstTemp.Items[K] is a string (at least according to StrToInt(lstTemp.Items[i])), so you need
Temperatures[K] := StrToInt(lstTemp.Items[K]);
Fourth, you declare Average as an integer, but it needs to be a floating-point number (obviously), like real or double.
Fifth,
edtAvg.Text:=IntToStr(Average); //Display Average
edtAvg.Text:=IntToStr(Min); //Display Minimum Temp.
edtAvg.Text:=IntToStr(Max); //Display Maximum Temp.
is not techncally incorrect, but will most likely not do what you want.
Sixth, although not an error, there is no need for you to initialise Count and Average to 0. Finally, you only need a single for loop.
There is (at least in Delphi 2010 - unit Math) one function that will calculate the mean and standard deviation in one step and functions that return the minimum and maximum values in an array. BTW, Mean is the arithmetic average of all the values and is the correct term. (I copied an example that I am working on and modified to your example - it compiles at least):
type
a = array of double;
var
Temperatures : a;
Average,stddev3, Max, Min : extended;
// Compiler insists on extended for these properties
begin
Max := Math.MaxValue(Temperatures);
Min := Math.MinValue(Temperatures);
Math.MeanAndStdDev(Temperatures ,Average,stddev3);
end;
For the maximum value in an array use (it takes an array of double and returns double):
function MaxValue(const Data: array of Double): Double;
For the minimum value use the corresponding:
function MinValue(const Data: array of Double): Double;
I agree that average cannot be an integer, but there are 2 similar functions for integer arrays:
function MinIntValue(const Data: array of Integer): Integer; and
function MaxIntValue(const Data: array of Integer): Integer;
0909EM's reply was very well done, but I have a few disagreements. First, I don't believe there's a need to set any sentinel value at all; simply use the first temperature value. Second, if we put a Begin and End around every single line If statement we'd approach COBOL-like levels of English verbosity. As it is, it's a crying shame this simple problem takes so much code. Third, I would not use StrToIntDef. Remember these lines from the Zen Of Python (I don't care if you don't know Python; everyone should memorize it, at least until we get an I Ching of Intersimone):
Errors should never pass silently.
Unless explicitly silenced.
If a user passes incorrect data into the temperature stats procedure, StrToIntDef is going to silently convert these values to zeroes, an unexpected and undesired behavior. The caller is going to get back answers that they assume are ok (because of no errors), yet will have incorrect values (especially the average). It is a far better thing to let the procedure blow up so testing will reveal the incorrect input.
I'd also replace the For loops with For...in. I banged this together:
program temps;
{$APPTYPE CONSOLE}
{$R *.res}
uses
System.SysUtils, System.Classes, Generics.Collections, Math;
Var
someTemps : TStringList;
Procedure TempStats(temperatures : TStringList);
Var
temps : TList<Real>;
minTemp, maxTemp, sumTemps : Real;
numTemps : Integer;
tempStr : String;
temp : Real;
avgTemp : Real;
Begin
numTemps := temperatures.Count;
If numTemps > 0 then
Begin
temps := TList<Real>.Create;
For tempStr in temperatures Do
temps.Add(StrToFloat(tempStr));
minTemp := temps[0];
maxTemp := temps[0];
sumTemps := 0;
For temp in temps Do
Begin
minTemp := Min(minTemp, temp);
maxTemp := Max(maxTemp, temp);
sumTemps := sumTemps + temp;
End;
avgTemp := sumTemps / numTemps;
WriteLn(avgTemp:0:2);
WriteLn(minTemp:0:2);
WriteLn(maxTemp:0:2);
temps.Free;
End
Else
WriteLn('No temperatures passed.');
End;
Begin
someTemps := TStringList.Create;
someTemps.AddStrings(TArray<String>.Create('72', '93', '84', '76', '82'));
TempStats(someTemps);
ReadLn;
someTemps.Clear;
TempStats(someTemps);
someTemps.Free;
ReadLn;
end.
Firstly, Consider using StrToIntDef (String To Integer with a Default value) instead of StrToInt (String to Integer) this will yield the following...
value := StrToIntDef('Abcdef', 0); // value will be zero
vs
value := StrToInt('Abcdef'); // exception
But the question is do you want integers or floating point values for your temperatures? (eg 1 or 1.6?) If you want floating point values, maybe use StrToFloatDef...
Second, I've seen lots of grads that use Delphi make this mistake, try to always use begin and end, it'll help... because it makes it really clear what you are doing inside a if/for/while and what you intend to do outside..
for i := 0 to lstTemp.Items.Count - 1 do
begin
// Sum all the items in the list
Sum := Sum + StrToIntDef(lstTemp.Items[i], 0);
end;
Next up your array is a bit pointless, the SetLength and adding items bit is OK, but its not very functional, when you could just use the items in the list. All you need to do is hang onto the max and min values.
Then your last problem is that Average isn't going to be a whole integer, its going to have a fractional part. Eg. 5 divided by 2 is 2.5, not 2 and not 3. You could use trunc to return just the integer part, or change Average so that its a floating point number...
for K:=0 to lstTemp.Items.Count-1 do
begin
if (StrToIntDef(lstTemp.Items[K], 0) > Max) then
begin
Max := StrToIntDef(lstTemp.Items[K], 0);
end;
if (StrToIntDef(lstTemp.Items[K], 1000) < Min) then // note, really high number
begin
Min := StrToIntDef(lstTemp.Items[K], 1000);
end;
end;
{Calculate Average}
Average := Trunc(Sum / Count); // do you really want to trunc this? I suspect not.
if Min = 1000 then // just incase
begin
Min := 0;
end;
The final problem you will face is that your always setting the text of the same text box...
edtAvg.Text:=IntToStr(Average); //Display Average
edtMin.Text:=IntToStr(Min); //Display Minimum Temp. (I assume this is supposed to be edtMin)
edtMax.Text:=IntToStr(Max); //Display Maximum Temp. (I assume this is supposed to be edtMax)
I suppose the final improvement I'd make is noticing that you only need one for loop...
for K:=0 to lstTemp.Items.Count-1 do
begin
// Sum all the items in the list
Sum := Sum + StrToIntDef(lstTemp.Items[K], 0);
if (StrToIntDef(lstTemp.Items[K], Low(Integer)) > Max) then // A really low value
begin
Max := StrToIntDef(lstTemp.Items[K], Low(Integer));
end;
if (StrToIntDef(lstTemp.Items[K], High(Integer)) < Min) then // A really high value
begin
Min := StrToIntDef(lstTemp.Items[K], High(Integer));
end;
end;
The most important idea on how to solve this is to read your error messages properly. On a previous question you commented: "the error was saying it is an overloaded function or something". That attitude is not going to help you understand the problem. You need to read the error messages properly.
In this question you give the following description of your errors:
So the 2 Errors are Error: Incompatible types: got "AnsiString" expected "LongInt" This is for Average := Sum / Count; Error: Incompatible types: got "Set Of Byte" expected "Double" This Error is for Temperatures[K] := lstTemp.Items[K];
However, the description does not correspond to the errors you should be seeing based on the code provided.
It looks like you didn't read your errors, and just blindly started making changes in the hopes you would accidentally do something right. Because you didn't read the errors, you didn't notice that they changed. So when you came to us looking for help, you provided old errors with new code or vice-versa.
If you had actually read your error messages properly, you might have been able to solve the problem yourself. At the least, you would have been able to ask a better question with a description that actually matched the code.
Average := Sum / Count;
Average, Sum and Count are all declared as Integer. The error message you should be getting is: "Incompatible types: Integer and Extended".
If you read the error message, it should give you a clue to read up on Integer and Extended.
The problem here is that, in maths, division produces a Rational number. And correspondingly the result of a division operation in a program is not an Integer. So you need to declare Average as either Double or Extended.
Temperatures[K] := lstTemp.Items[K];
Temperatures is declared as an array of Integer. You haven't shown the declaration of lstTemp, but based on other code it's one of the standard Delphi Controls that has Items declared as TStrings. So the error message you should be getting is: "Incompatible types: Integer and string".
If you read the error message, it should give you a clue to do the same thing you did 5 lines earlier.
The reason for this error is that Delphi is a "strongly typed" language. The compiler tries to prevent you from making certain kinds of mistakes because it is much better catch them early. Imagine what might happen if one of the values in lstTemp were 'Hello'. That cannot be converted to an Integer; and would cause a "run-time" error in your program.
To fix this problem you need to tell the compiler: "I know the value is a string and could be any string, but I want you to convert it to an Integer". You do this by calling the StrToInt function. NOTE: You will still get a run time error if an invalid string is passed to the function, but by being forced to explicitly do the conversion, you can think about whether you want to do some pre-validation of your input data.
You asked about the errors reported by the compiler. That's just one kind of error you'll face when programming - and usually the easiest to resolve. You'll also encounter logic errors: where your program compiles successfully, but doesn't behave correctly. Andreas's excellent answer has covered those already, so I'll not repeat them.
However, I will give you some valuable advice. Once you've gotten over the hurdle of resolving compiler errors, and are able to do so easily - you need to as quickly as possible:
Get into the habit of testing your code thoroughly.
Learn how to use the integrated debugger.
Learn about its limitations.
Learn other debugging techniques: logging, profiling, pre- and post-condition checking.
Finally, as a response to alcalde's rant about there not being any simple functions to get Min, Max, Sum or Avg: I offer another possible implementation.
Basically the rant was about the fact that he'd far rather write something along the lines of:
begin
if (lstTemp.Count > 0) then
begin
edtMin.Text := lstTemp.Min;
edtMax.Text := lstTemp.Max;
edtAvg.Text := lstTemp.Average;
end
else
begin
ShowMessage('List is empty');
end;
end;
Obviously the above code won't compile, but with a little work we can achieve something similar.
He's perfectly right on two counts: (1) that this implementation would be cleaner, much easier to maintain and with less chance of errors. (2) Delphi doesn't provide a way to simply do that.
In fact, if you follow a top-down design approach, this might be your initial pseudo-code. You should be taught about top-down design, if not demand your money back. :)
The whole point behind the top-down-design approach is that you're looking for an ideal implementation. You're not worrying about what is/isn't there. If the current library and tools don't provide a Min function, you can write your own.
You are a programmer, you have the power!
I sometimes like to call this "wishful thinking programming". You're wishing if other things were in place, I could implement the functionality much more easily like "this". Then you go about making your wish come true.
Without further ado, here's the implementation. You will need to use the Math unit.
type
{ We will call existing functions that take TDoubleArray as input }
TDoubleArray = array of Double;
TStringsHelper = class(TStrings)
{ A useful class to help us convert TStrings into TDoubleArray }
public
class function Using(AStrings: TStrings): TStringsHelper;
function AsDoubleArray: TDoubleArray;
end;
{ TStringsHelper }
function TStringsHelper.AsDoubleArray: TDoubleArray;
var
LoopI: Integer;
begin
SetLength(Result, Count);
for LoopI := 0 to Count - 1 do
begin
Result[LoopI] := StrToFloat(Strings[LoopI]);
end;
end;
class function TStringsHelper.Using(AStrings: TStrings): TStringsHelper;
begin
Result := TStringsHelper(AStrings);
end;
var
LTemperatures: TDoubleArray;
begin
{ This code is almost the same as our "ideal" implementation }
if (lstTemp.Items.Count > 0) then
begin
LTemperatures := TStringsHelper.Using(lstTemp.Items).AsDoubleArray;
edtMin.Text := FloatToStr(MinValue(LTemperatures));
edtMin.Text := FloatToStr(MaxValue(LTemperatures));
edtMin.Text := FloatToStr(Mean(LTemperatures));
end
else
begin
ShowMessage('List is empty');
end;
end;
What values are in lstTemp.Items[i]?
I suppose the values are integers (without floating points), because you are using IntToStr.
Average cannot be an integer. Integer is a number (4 bytes) without a floating point. A simple numbers, such as 2,3,50,1500, -100
Assume that Sum = 100, and the Count = 3.
What Average will be?
So, you have to use float variable type, Double for example.
I hope it helps...

How can I search faster for name/value pairs in a Delphi TStringList?

I implemented language translation in an application by putting all strings at runtime in a TStringList with:
procedure PopulateStringList;
begin
EnglishStringList.Append('CAN_T_FIND_FILE=It is not possible to find the file');
EnglishStringList.Append('DUMMY=Just a dummy record');
// total of 2000 record appended in the same way
EnglishStringList.Sorted := True; // Updated comment: this is USELESS!
end;
Then I get the translation using:
function GetTranslation(ResStr:String):String;
var
iIndex : Integer;
begin
iIndex := -1;
iIndex := EnglishStringList.IndexOfName(ResStr);
if iIndex >= 0 then
Result := EnglishStringList.ValueFromIndex[iIndex] else
Result := ResStr + ' (Translation N/A)';
end;
Anyway with this approach it takes about 30 microseconds to locate a record, is there a better way to achieve the same result?
UPDATE: For future reference I write here the new implementation that uses TDictionary as suggested (works with Delphi 2009 and newer):
procedure PopulateStringList;
begin
EnglishDictionary := TDictionary<String, String>.Create;
EnglishDictionary.Add('CAN_T_FIND_FILE','It is not possible to find the file');
EnglishDictionary.Add('DUMMY','Just a dummy record');
// total of 2000 record appended in the same way
end;
function GetTranslation(ResStr:String):String;
var
ValueFound: Boolean;
begin
ValueFound:= EnglishDictionary.TryGetValue(ResStr, Result);
if not ValueFound then Result := Result + '(Trans N/A)';
end;
The new GetTranslation function performs 1000 times faster (on my 2000 sample records) then the first version.
THashedStringList should be better, I think.
In Delphi 2009 or later I would use TDictionary< string,string > in Generics.Collections.
Also note that there are free tools such as http://dxgettext.po.dk/ for translating applications.
If THashedStringList works for you, that's great. Its biggest weakness is that every time you change the contents of the list, the Hash table is rebuilt. So it will work for you as long as your list remains small or doesn't change very often.
For more info on this, see: THashedStringList weakness, which gives a few alternatives.
If you have a big list that may be updated, you might want to try GpStringHash by gabr, that doesn't have to recompute the whole table at every change.
I think that you don't use the EnglishStringList(TStringList) correctly. This is a sorted list, you add elements (strings), you sort it, but when you search, you do this by a partial string (only the name, with IndexOfName).
If you use IndexOfName in a sorted list, the TStringList can't use Dicotomic search. It use sequential search.
(this is the implementation of IndexOfName)
for Result := 0 to GetCount - 1 do
begin
S := Get(Result);
P := AnsiPos('=', S);
if (P <> 0) and (CompareStrings(Copy(S, 1, P - 1), Name) = 0) then Exit;
end;
I think that this is the reason of poor performance.
The alternative is use 2 TStringList:
* The first (sorted) only containts the "Name" and a pointer to the second list that contain the value; You can implement this pointer to the second list using the "pointer" of Object property.
* The second (not sorted) list containt the values.
When you search, you do it at first list; In this case you can use the Find method. when you find the name, the pointer (implemented with Object property) give you the position on second list with the value.
In this case, Find method on Sorted List is more efficient that HashList (that must execute a funcion to get the position of a value).
Regards.
Pd:Excuse-me for mistakes with english.
You can also use a CLASS HELPER to re-program the "IndexOfName" function:
TYPE
TStringsHelper = CLASS HELPER FOR TStrings
FUNCTION IndexOfName(CONST Name : STRING) : INTEGER;
END;
FUNCTION TStringsHelper.IndexOfName(CONST Name : STRING) : INTEGER;
VAR
SL : TStringList ABSOLUTE Self;
S,T : STRING;
I : INTEGER;
BEGIN
IF (Self IS TStringList) AND SL.Sorted THEN BEGIN
S:=Name+NameValueSeparator;
IF SL.Find(S,I) THEN
Result:=I
ELSE IF (I<0) OR (I>=Count) THEN
Result:=-1
ELSE BEGIN
T:=SL[I];
IF CompareStrings(COPY(T,1,LENGTH(S)),S)=0 THEN Result:=I ELSE Result:=-1
END;
EXIT
END;
Result:=INHERITED IndexOfName(Name)
END;
(or implement it in a descendant TStrings class if you dislike CLASS HELPERs or don't have them in your Delphi version).
This will use a binary search on a sorted TStringList and a sequential search on other TStrings classes.

Are there any advantages in using const parameters with an ordinal type?

I know marking string parameters as const can make a huge performance difference, but what about ordinal types? Do I gain anything by making them const?
I've always used const parameters when handling strings, but never for Integer, Pointer, class instances, etc.
When using const I often have to create additional temporary variables, which replace the now write-protected parameters, so I'm wondering: Do I gain anything from marking ordinal parameters as const?
You need to understand the reason, to avoid "cargo-cult programming." Marking strings as const makes a performance difference because you no longer need to use an interlocked increment and decrement of the refcount on the string, an operation that actually becomes more expensive, not less, as time goes by because more cores means more work that has to be done to keep atomic operations in sync. This is safe to do since the compiler enforces the "this variable will not be changed" constraint.
For ordinals, which are usually 4 bytes or less, there's no performance gain to be had. Using const as optimization only works when you're using value types that are larger than 4 bytes, such as arrays or records, or reference-counted types such as strings and interfaces.
However, there's another important advantage: code readability. If you pass something as const and it makes no difference whatsoever to the compiler, it can still make a difference to you, since you can read the code and see that the intention of it was to have this not be modified. That can be significant if you haven't seen the code before (someone else wrote it) or if you're coming back to it after a long time and don't remember exactly what you were thinking when you originally wrote it.
You can't accidentally treat them like var parameters and have your code compile. So it makes your intentions clear.
Declaring ordinal types const makes no difference because they are copied anyway (call-by-value), so any changes to the variable do not affect the original variable.
procedure Foo (Val : Integer)
begin
Val := 2;
end;
...
SomeVar := 3;
Foo (SomeVar);
Assert (SomeVar = 3);
IMHO declaring ordinal types const makes no sense and as you say requires you to introduce local variables often.
It depends on how complex is your routine and how it is used. If it is used many places and required that the value stay the same, declare it as "const" to make it cleared and safe. For string type, there was a bug (for Delphi 7 as I stump on it) that causes memory corruption if declare as "const". Below is sample codes
type
TFoo = class
private
FStr: string;
public
procedure DoFoo(const AStr: string);
begin
FStr := AStr; //the trouble code 1
......
end;
procedure DoFoo2;
begin
.....
DoFoo(FStr); //the trouble code 2
end;
end;
There's a huge speed improvement using Const with strings:
function test(k: string): string;
begin
Result := k;
end;
function test2(Const k: string): string;
begin
Result := k;
end;
function test3(Var k: string): string;
begin
Result := k;
end;
procedure TForm1.Button1Click(Sender: TObject);
Var a: Integer;
s,b: string;
x: Int64;
begin
s := 'jkdfjklf lkjj3i2ej39ijkl jkl2eje23 io3je32 e832 eu283 89389e3jio3 j938j 839 d983j9';
PerfTimerInit;
for a := 1 to 10000000 do
b := test(s);
x := PerfTimerStopMS;
Memo1.Lines.Add('default: '+x.ToString);
PerfTimerInit;
for a := 1 to 10000000 do
b := test2(s);
x := PerfTimerStopMS;
Memo1.Lines.Add('const: '+x.ToString);
PerfTimerInit;
for a := 1 to 10000000 do
b := test3(s);
x := PerfTimerStopMS;
Memo1.Lines.Add('var: '+x.ToString);
end;
default: 443 const: 320 var: 325
default: 444 const: 303 var: 310
default: 444 const: 302 var: 305
Same with Integers:
default: 142 const: 13 var: 14
Interestingly though, in 64-bit there seems to be almost no difference with strings (default mode is only a bit slower than Const):
default: 352 const: 313 var: 314

How to use TComparer to sort a list of records?

I have a record that holds data about a file:
TYPE
RFile= record
public
FileName : string;
Resolution : Integer;
FileSize : Cardinal;
Rating : Byte;
end;
PFile= ^RFile;
And I keep a list of these files/records in a TList<>
TFileList= class(TList<PFile>)
procedure SortByFilename;
procedure SortByRating;
procedure SortByResolution;
procedure SortBySize;
end;
I have methods like SortByFilename, SortBySize, etc in which I sort the list.
I do "classic" sorting.
Now I want to upgrade to the new-and-cool System.Generics.Defaults.TComparer.
From what I understand I need to assign a comparer to my TFileList, like
TIntStringComparer = class(TComparer<String>)
public
function Compare(const Left, Right: String): Integer; override;
end;
How do I do this?
How do I deal with one comparer for each data field (filename, filesize, resolution)?
Update:
This code compiles but I have an EIntegerOverflow because the FileSize is a cardinal while I return an integer (diff between two cardinals).
Sort(TComparer<PFile>.Construct(
function(CONST A,B: PFile): integer
begin
Result:= A.FileSize - B.FileSize;
end
));
When you write a comparer for a numeric type, you should never use subtraction, even if your data type is signed.
Indeed, try to compare a = 100 and b = -2147483640 as Integers; clearly a > b, but subtraction will yield the wrong result.
Instead, you should always do something similar to
if a = b then
Result := 0
else if a < b then
Result := -1
else
Result := 1;
But Delphi's RTL already contains functions for this: there are several CompareValue overloads in the Math unit (for different types of integers and floats -- but, unfortunately, not for Cardinals).
Thus, although your snippet will work "most of the time" if you do
Result := Integer(A.FileSize) - Integer(B.FileSize)
this is not good enough: For one thing, not every Cardinal will fit in a Integer. Also, as noted above, subtraction is not the way to go.
In your case, you can simply use the if thing above directly, or you could create a new CompareValue overload for Cardinals. Or, you could do
Result := CompareValue(Int64(A.FileSize), Int64(B.FileSize)).
(Also, as others have stated in comments, you should reconsider if it is wise to use a Cardinal to store a file size in the first place. If you upgrade this to an Int64 or UInt64 you can write simply
Result := CompareValue(A.FileSize, B.FileSize).)

Resources