www.digitalmars.com         C & C++   DMDScript  

digitalmars.D - Wouldn't it be nice (case range statements)

reply "John Colvin" <john.loughran.colvin gmail.com> writes:
if code like this worked: http://dpaste.dzfl.pl/7ea4eb03f02e

A few reasons why it doesn't:

You have to duplicate the case keyword when declaring case 
ranges. Why?

Case ranges are inclusive at both ends of the range, unlike in 
foreach. Again, why?

exponential notation (e.g. `2e9`) returns a double, not a long.


The exponential notation isn't really a problem, declaring some 
enums `enum i2e9 = cast(long)2e9;` deals with it fine. The case 
ranges are a wart though.

Solution:

Allow the second `case` keyword to be removed, which would then 
have the same semantics as the range in foreach.

E.g.
case 0 .. 4: // matches 0,1,2,3
case 0: .. case 4: // matches 0,1,2,3,4

No breakage, greater consistency, neater code, good stuff. At 
least as good as pascal.

Even better, the .. operator would become general (overloadable, 
too) and the case range would just be a special case of it.
Oct 14 2014
next sibling parent reply "Adam D. Ruppe" <destructionator gmail.com> writes:
On Tuesday, 14 October 2014 at 21:29:59 UTC, John Colvin wrote:

 You have to duplicate the case keyword when declaring case 
 ranges. Why?

 Case ranges are inclusive at both ends of the range, unlike in 
 foreach. Again, why?


It comes from writing:

switch(foo) {
   case 1:
   case 2:
   case 3:
   case 4:
      // code
}

Then just replacing the case 2 and case 3 with .. collapsing the 
repetition to just the beginning and the end. If you write it as:

switch(foo) {
   case 1:
   ..
   case 4:
      // code
}

vertically, that is, I think the rationale becomes a lot more 
clear. I like this a lot, it works brilliantly for me and makes 
good sense.


 Allow the second `case` keyword to be removed, which would then 
 have the same semantics as the range in foreach.


That would be ok too.
Oct 14 2014
parent "John Colvin" <john.loughran.colvin gmail.com> writes:
On Tuesday, 14 October 2014 at 21:33:36 UTC, Adam D. Ruppe wrote:

 On Tuesday, 14 October 2014 at 21:29:59 UTC, John Colvin wrote:

 You have to duplicate the case keyword when declaring case 
 ranges. Why?

 Case ranges are inclusive at both ends of the range, unlike in 
 foreach. Again, why?


 It comes from writing:

 switch(foo) {
   case 1:
   case 2:
   case 3:
   case 4:
      // code
 }

 Then just replacing the case 2 and case 3 with .. collapsing 
 the repetition to just the beginning and the end. If you write 
 it as:

 switch(foo) {
   case 1:
   ..
   case 4:
      // code
 }

 vertically, that is, I think the rationale becomes a lot more 
 clear. I like this a lot, it works brilliantly for me and makes 
 good sense.


Ah, yeah, that does make sense.


 Allow the second `case` keyword to be removed, which would 
 then have the same semantics as the range in foreach.


 That would be ok too.
Oct 15 2014
prev sibling next sibling parent Andrei Alexandrescu <SeeWebsiteForEmail erdani.org> writes:
On 10/14/14, 2:29 PM, John Colvin wrote:

 if code like this worked: http://dpaste.dzfl.pl/7ea4eb03f02e

 A few reasons why it doesn't:

 You have to duplicate the case keyword when declaring case ranges. Why?

 Case ranges are inclusive at both ends of the range, unlike in foreach.
 Again, why?

 exponential notation (e.g. `2e9`) returns a double, not a long.


 The exponential notation isn't really a problem, declaring some enums
 `enum i2e9 = cast(long)2e9;` deals with it fine. The case ranges are a
 wart though.

 Solution:

 Allow the second `case` keyword to be removed, which would then have the
 same semantics as the range in foreach.

 E.g.
 case 0 .. 4: // matches 0,1,2,3
 case 0: .. case 4: // matches 0,1,2,3,4

 No breakage, greater consistency, neater code, good stuff. At least as
 good as pascal.

 Even better, the .. operator would become general (overloadable, too)
 and the case range would just be a special case of it.


The short answer is it's all good as it is. -- Andrei
Oct 14 2014
prev sibling parent "eles" <eles eles.com> writes:
On Tuesday, 14 October 2014 at 21:29:59 UTC, John Colvin wrote:

 if code like this worked: http://dpaste.dzfl.pl/7ea4eb03f02e

 A few reasons why it doesn't:

 You have to duplicate the case keyword when declaring case 
 ranges. Why?

 Case ranges are inclusive at both ends of the range, unlike in 
 foreach. Again, why?


Actually, the latter is the reason for the former. If incluseve 
ranges were defined (let's say with ...), then that becomes 
possible.
Oct 15 2014