say I have an enum

    enum X
    {
     A,
     B,
     C
    }

and I have

    void someFunc( X e )
    {
     //.. some code
    }

I want to print the value of 'e' but I don't want to get a number!! I want to get a string that represents it. i.e. A or B or C


    void someFunc( X e )
    {
       toString(e);
       e.string;
       e.value;
       //or something like that ..
    }

Is there any such thing in D?

Hasan Aljudy wrote:
> say I have an enum
> 
>     enum X
>     {
>      A,
>      B,
>      C
>     }
> 
> and I have
> 
>     void someFunc( X e )
>     {
>      //.. some code
>     }
> 
> I want to print the value of 'e' but I don't want to get a number!! I want to get a string that represents it. i.e. A or B or C
> 
> 
>     void someFunc( X e )
>     {
>        toString(e);
>        e.string;
>        e.value;
>        //or something like that ..
>     }
> 
> Is there any such thing in D?
> 
Hello,

AFAIK, there is now way to do what you want directly. Simplest solution is to create helper map that would have X as a key and char[] as a value and then to define function toString(X x) the would return string representation.


-- 
Victor (aka nail) Nakoryakov
nail-mail<at>mail<dot>ru

Krasnoznamensk, Moscow, Russia

Hasan Aljudy wrote:
> say I have an enum
> 
>     enum X
>     {
>      A,
>      B,
>      C
>     }
> 
> and I have
> 
>     void someFunc( X e )
>     {
>      //.. some code
>     }
> 
> I want to print the value of 'e' but I don't want to get a number!! I want to get a string that represents it. i.e. A or B or C
> 
> Is there any such thing in D?
> 

Sadly, no. I once suggested that enums be a little smarter, more like .NET enums or, even better in my opinion, Java 5 enums. Being able to obtain a string representation from an enum value, parsing an enum value from a string, and (in Java versions) attach functionality or custom strings to enum values, is very powerful.

If I remember correctly, the idea was rejected out right. Maybe there are performance issues?

Jason

I've done that "the hard way" in C.
Here's an example in D:

/////
import std.stdio;
import std.string;

enum X {
  Apple,
  Bat,
  Car,
}

char [][] X_names = [
   X.Apple : "Apple",
   X.Bat   : "Bat",
   X.Car   : "Car",
];

char [] get_X_name(X e)
{
   if ((e >= X.min) && (cast(int)e < X_names.length) &&
       (X_names[e] !is null)) {
      return X_names[e];
   }
   return ("invalid");
}

X get_X_id(char [] name)
{
   for (int idx = 0; idx < X_names.length; idx++) {
      if ((X_names[idx] !is null) && (icmp(X_names[idx], name) == 0))
         return cast(X)idx;
   }
   return cast(X)-1;
}

void main(char [][] args)
{
   for (int i = -1; i < 4; i++)
   {
      writef("%d = '%s'\n", i, get_X_name(cast(X)i));
   }

   char [] name = "bat";
   writef("id for '%s' is %d\n", name, cast(int)get_X_id(name));
}
////

Running this produces the output:
-1 = 'invalid'
0 = 'Apple'
1 = 'Bat'
2 = 'Car'
3 = 'invalid'
id for 'bat' is 1

Ben

Hasan Aljudy wrote:
> say I have an enum
> 
>     enum X
>     {
>      A,
>      B,
>      C
>     }
> 
> and I have
> 
>     void someFunc( X e )
>     {
>      //.. some code
>     }
> 
> I want to print the value of 'e' but I don't want to get a number!! I want to get a string that represents it. i.e. A or B or C
> 
> 
>     void someFunc( X e )
>     {
>        toString(e);
>        e.string;
>        e.value;
>        //or something like that ..
>     }
> 
> Is there any such thing in D?
> 

You have the right idea, although I think it would be better to name the X->char[] function 'toString' and the char[]->X function 'toX' and the map with all capitals. Convention, you see, and clarity through expressions like:
# X var = X.Apple;
# char[] str = var.toString;
# X another = str.toX;

In fact, if you don't want the char[]->X conversion at all, then the map can be a static variable of 'toString(X)', to prevent namespace pollution.

One other suggestion: note the rewritten code below:

Ben Gardner wrote:
> I've done that "the hard way" in C.
> Here's an example in D:
> 
> /////
> import std.stdio;
> import std.string;
> 
> enum X {
>   Apple,
>   Bat,
>   Car,
> }
> 
> char [][] X_names = [
>    X.Apple : "Apple",
>    X.Bat   : "Bat",
>    X.Car   : "Car",
> ];
> 
> char [] get_X_name(X e)
> {
>    if ((e >= X.min) && (cast(int)e < X_names.length) &&
>        (X_names[e] !is null)) {
>       return X_names[e];
>    }
>    return ("invalid");
> }

char[] toString (X value) {
  if (auto foo = value in X_NAMES)
    return *foo;
  return "invalid";
}

> X get_X_id(char [] name)
> {
>    for (int idx = 0; idx < X_names.length; idx++) {
>       if ((X_names[idx] !is null) && (icmp(X_names[idx], name) == 0))
>          return cast(X)idx;
>    }
>    return cast(X)-1;
> }

X toX (char[] value) {
  foreach (id, name; X_NAMES) {
    if (icmp(name, value) == 0)
      return id;
  }
  return cast(X)-1;
}

-- Chris Nicholson-Sauls

Ben Gardner wrote:
> I've done that "the hard way" in C.
> Here's an example in D:
> 
> /////
> import std.stdio;
> import std.string;
> 
> enum X {
>   Apple,
>   Bat,
>   Car,
> }
> 
> char [][] X_names = [
>    X.Apple : "Apple",
>    X.Bat   : "Bat",
>    X.Car   : "Car",
> ];
> 
> char [] get_X_name(X e)
> {
>    if ((e >= X.min) && (cast(int)e < X_names.length) &&
>        (X_names[e] !is null)) {
>       return X_names[e];
>    }
>    return ("invalid");
> }
> 
> X get_X_id(char [] name)
> {
>    for (int idx = 0; idx < X_names.length; idx++) {
>       if ((X_names[idx] !is null) && (icmp(X_names[idx], name) == 0))
>          return cast(X)idx;
>    }
>    return cast(X)-1;
> }
> 
> void main(char [][] args)
> {
>    for (int i = -1; i < 4; i++)
>    {
>       writef("%d = '%s'\n", i, get_X_name(cast(X)i));
>    }
> 
>    char [] name = "bat";
>    writef("id for '%s' is %d\n", name, cast(int)get_X_id(name));
> }
> ////
> 
> Running this produces the output:
> -1 = 'invalid'
> 0 = 'Apple'
> 1 = 'Bat'
> 2 = 'Car'
> 3 = 'invalid'
> id for 'bat' is 1
> 
> Ben
> 
> Hasan Aljudy wrote:
> 
>>say I have an enum
>>
>>    enum X
>>    {
>>     A,
>>     B,
>>     C
>>    }
>>
>>and I have
>>
>>    void someFunc( X e )
>>    {
>>     //.. some code
>>    }
>>
>>I want to print the value of 'e' but I don't want to get a number!! I
>>want to get a string that represents it. i.e. A or B or C
>>
>>
>>    void someFunc( X e )
>>    {
>>       toString(e);
>>       e.string;
>>       e.value;
>>       //or something like that ..
>>    }
>>
>>Is there any such thing in D?
>>


I'll propose that a new property be added, somewhat like the .mangleof property. Instead, a .nameof property would simply return the lexical token for the named entity. Doesn't matter whether it refers to a struct, class, some attribute thereof, enum types or members, whatever ... the x.nameof should just return a char[] of the respective name.

Thoughts?

kris wrote:
> 
> I'll propose that a new property be added, somewhat like the .mangleof property. Instead, a .nameof property would simply return the lexical token for the named entity. Doesn't matter whether it refers to a struct, class, some attribute thereof, enum types or members, whatever ... the x.nameof should just return a char[] of the respective name.
> 
> Thoughts?

This would be easy implement if the enum value is known at compile time
(ie, X.Apple.nameof).

But to do this for an unknown enum value would require that a complete string table be defined for every enum.

void foo(X e)
{
   writef("the enum name is %s\n", e.nameof);
}

I suppose that the compiler would be smart enough to drop the string table if it is never used, so there is no harm in defining the table for all enums.

Ben

kris wrote:
> Ben Gardner wrote:
> 
>> I've done that "the hard way" in C.
>> Here's an example in D:
>>
>> /////
>> import std.stdio;
>> import std.string;
>>
>> enum X {
>>   Apple,
>>   Bat,
>>   Car,
>> }
>>
>> char [][] X_names = [
>>    X.Apple : "Apple",
>>    X.Bat   : "Bat",
>>    X.Car   : "Car",
>> ];
>>
>> char [] get_X_name(X e)
>> {
>>    if ((e >= X.min) && (cast(int)e < X_names.length) &&
>>        (X_names[e] !is null)) {
>>       return X_names[e];
>>    }
>>    return ("invalid");
>> }
>>
>> X get_X_id(char [] name)
>> {
>>    for (int idx = 0; idx < X_names.length; idx++) {
>>       if ((X_names[idx] !is null) && (icmp(X_names[idx], name) == 0))
>>          return cast(X)idx;
>>    }
>>    return cast(X)-1;
>> }
>>
>> void main(char [][] args)
>> {
>>    for (int i = -1; i < 4; i++)
>>    {
>>       writef("%d = '%s'\n", i, get_X_name(cast(X)i));
>>    }
>>
>>    char [] name = "bat";
>>    writef("id for '%s' is %d\n", name, cast(int)get_X_id(name));
>> }
>> ////
>>
>> Running this produces the output:
>> -1 = 'invalid'
>> 0 = 'Apple'
>> 1 = 'Bat'
>> 2 = 'Car'
>> 3 = 'invalid'
>> id for 'bat' is 1
>>
>> Ben
>>
>> Hasan Aljudy wrote:
>>
>>> say I have an enum
>>>
>>>    enum X
>>>    {
>>>     A,
>>>     B,
>>>     C
>>>    }
>>>
>>> and I have
>>>
>>>    void someFunc( X e )
>>>    {
>>>     //.. some code
>>>    }
>>>
>>> I want to print the value of 'e' but I don't want to get a number!! I
>>> want to get a string that represents it. i.e. A or B or C
>>>
>>>
>>>    void someFunc( X e )
>>>    {
>>>       toString(e);
>>>       e.string;
>>>       e.value;
>>>       //or something like that ..
>>>    }
>>>
>>> Is there any such thing in D?
>>>
> 
> 
> I'll propose that a new property be added, somewhat like the .mangleof property. Instead, a .nameof property would simply return the lexical token for the named entity. Doesn't matter whether it refers to a struct, class, some attribute thereof, enum types or members, whatever ... the x.nameof should just return a char[] of the respective name.
> 
> Thoughts?

What if it's an alias?  Return the alias identifier's name or the thing which it aliases?

Template parameters?  Return the name of the template parameter or the actual object passed in?

-- 
Regards,
James Dunne

Ben Gardner wrote:
> kris wrote:
> 
>>I'll propose that a new property be added, somewhat like the .mangleof
>>property. Instead, a .nameof property would simply return the lexical
>>token for the named entity. Doesn't matter whether it refers to a
>>struct, class, some attribute thereof, enum types or members, whatever
>>... the x.nameof should just return a char[] of the respective name.
>>
>>Thoughts?
> 
> 
> This would be easy implement if the enum value is known at compile time
> (ie, X.Apple.nameof).
> 
> But to do this for an unknown enum value would require that a complete
> string table be defined for every enum.
> 
> void foo(X e)
> {
>    writef("the enum name is %s\n", e.nameof);
> }
> 
> I suppose that the compiler would be smart enough to drop the string
> table if it is never used, so there is no harm in defining the table for
> all enums.
> 
> Ben

Yeah, I think it's actually very easy to implement, I don't see why dmd doesn't do it.

for every enum X, the parser can very easily identify EnumMembers and generate a table along the lines of:

    char[] [X] XMemberToStringTable;

    static this()
    {
        XMemberToStringTable[X.A] = "A";
        XMemberToStringTable[X.B] = "B";
        XMemberToStringTable[X.C] = "C";
    }

    char[] XtoString( X a )
    {
        return XMemberToStringTable[a];
    }

not hard at all.

kris wrote:
>
> I'll propose that a new property be added, somewhat like the .mangleof property. Instead, a .nameof property would simply return the lexical token for the named entity. Doesn't matter whether it refers to a struct, class, some attribute thereof, enum types or members, whatever ... the x.nameof should just return a char[] of the respective name.
> 
> Thoughts?

This would be great.


Sean