www.digitalmars.com         C & C++   DMDScript  

digitalmars.D.learn - What's wrong with my usage of std.algorithm.map in this code example?

reply pineapple <meapineapple gmail.com> writes:
I would've expected this to work, but instead I get a compile 
error. Is my syntax wrong? Is this just not a case that map can 
handle, and I should be doing something else?

     import std.algorithm : map;
     import std.conv : to;
     import std.stdio : writeln;
     import std.string : join;

     string test(Args...)(in Args items){
         immutable string[items.length] itemstrings = 
map!(to!string)(items);
         return join(itemstrings, ", ");
     }

     unittest{
         writeln(test(1, 2, 3, 4));
     }
May 24 2016
next sibling parent reply Steven Schveighoffer <schveiguy yahoo.com> writes:
On 5/24/16 4:03 PM, pineapple wrote:
 I would've expected this to work, but instead I get a compile error. Is
 my syntax wrong? Is this just not a case that map can handle, and I
 should be doing something else?

      import std.algorithm : map;
      import std.conv : to;
      import std.stdio : writeln;
      import std.string : join;

      string test(Args...)(in Args items){
          immutable string[items.length] itemstrings =
 map!(to!string)(items);
Slice assignment from range to array is not supported. In your example, I'm curious why the efforts to specify the type? I think it would work with just saying auto itemstrings = ... -Steve
May 24 2016
parent reply pineapple <meapineapple gmail.com> writes:
On Tuesday, 24 May 2016 at 20:18:34 UTC, Steven Schveighoffer 
wrote:
 Slice assignment from range to array is not supported.

 In your example, I'm curious why the efforts to specify the 
 type? I think it would work with just saying auto itemstrings = 
 ...

 -Steve
I still get an error if I use auto instead.
May 25 2016
next sibling parent Chris <wendlec tcd.ie> writes:
On Wednesday, 25 May 2016 at 10:24:19 UTC, pineapple wrote:
 On Tuesday, 24 May 2016 at 20:18:34 UTC, Steven Schveighoffer 
 wrote:
 Slice assignment from range to array is not supported.

 In your example, I'm curious why the efforts to specify the 
 type? I think it would work with just saying auto itemstrings 
 = ...

 -Steve
I still get an error if I use auto instead.
If you really need it, this works: import std.algorithm : map; import std.conv : to; import std.stdio : writeln; import std.string : join; string test(Args...)(in Args items) { writeln(items.stringof); string[] itemstrings; foreach (const ref i; items) itemstrings ~= i.to!string; // auto itemstrings = items.map!(a => a.to!string); return join(itemstrings, ", "); } unittest { writeln(test(1, 2, "v", 4, 'c')); } If you use map!(), you get this error: Error: template map_error.test!(int, int, string, int, char).test.map!((a) => a.to!string).map cannot deduce function from argument types !()(const(int), const(int), const(string), const(int), const(char)), candidates are: /home/christoph/.dvm/compilers/dmd-2.071.0/linux/bin/../../src/phobos/std/algorit m/iteration.d(450): map_error.test!(int, int, string, int, char).test.map!((a) => a.to!string).map(Range)(Range r) if (isInputRange!(Unqual!Range)) The argument types don't match, i.e. they are const(int) etc. instead of int. The arguments are passed as a tuple: cf. writeln(items.stringof); tuple(_param_0, _param_1, _param_2, _param_3, _param_4)
May 25 2016
prev sibling parent reply Steven Schveighoffer <schveiguy yahoo.com> writes:
On 5/25/16 6:24 AM, pineapple wrote:
 On Tuesday, 24 May 2016 at 20:18:34 UTC, Steven Schveighoffer wrote:
 Slice assignment from range to array is not supported.

 In your example, I'm curious why the efforts to specify the type? I
 think it would work with just saying auto itemstrings = ...
I still get an error if I use auto instead.
OK, I see the other issue now. map takes a range, whereas you are giving it a tuple. -Steve
May 25 2016
parent reply Chris <wendlec tcd.ie> writes:
On Wednesday, 25 May 2016 at 12:08:20 UTC, Steven Schveighoffer 
wrote:
 On 5/25/16 6:24 AM, pineapple wrote:
 On Tuesday, 24 May 2016 at 20:18:34 UTC, Steven Schveighoffer 
 wrote:
 Slice assignment from range to array is not supported.

 In your example, I'm curious why the efforts to specify the 
 type? I
 think it would work with just saying auto itemstrings = ...
I still get an error if I use auto instead.
OK, I see the other issue now. map takes a range, whereas you are giving it a tuple. -Steve
Why can the tuple be iterated with foreach, as in my quick fix, and indexed with tuple[0..], but is not accepted as a range? What are the differences? Is there a way to rangify a tuple?
May 25 2016
next sibling parent Chris <wendlec tcd.ie> writes:
On Wednesday, 25 May 2016 at 13:27:28 UTC, Chris wrote:
 On Wednesday, 25 May 2016 at 12:08:20 UTC, Steven Schveighoffer 
 wrote:
 On 5/25/16 6:24 AM, pineapple wrote:
 On Tuesday, 24 May 2016 at 20:18:34 UTC, Steven Schveighoffer 
 wrote:
 Slice assignment from range to array is not supported.

 In your example, I'm curious why the efforts to specify the 
 type? I
 think it would work with just saying auto itemstrings = ...
I still get an error if I use auto instead.
OK, I see the other issue now. map takes a range, whereas you are giving it a tuple. -Steve
Why can the tuple be iterated with foreach, as in my quick fix, and indexed with tuple[0..], but is not accepted as a range? What are the differences? Is there a way to rangify a tuple?
I should add : a homogeneous tuple, e.g. Tuple!(int, int, int);
May 25 2016
prev sibling next sibling parent reply ag0aep6g <anonymous example.com> writes:
On 05/25/2016 03:27 PM, Chris wrote:
 Why can the tuple be iterated with foreach, as in my quick fix, and
 indexed with tuple[0..], but is not accepted as a range? What are the
 differences?
popFront doesn't make sense with a tuple (aka expression list). When you remove the first element of a tuple, you get another tuple of a different type, but popFront can't change the type of the range. In code: ---- void main() { import std.meta: AliasSeq; AliasSeq!(int, int, int) tuple = AliasSeq!(1, 2, 3); tuple.popFront(); /* How would this be implemented? Would have to change tuple's type to AliasSeq!(int, int). */ } ----
 Is there a way to rangify a tuple?
std.range.only: ---- void main() { import std.meta: AliasSeq; import std.range: only; AliasSeq!(int, int, int) tuple = AliasSeq!(1, 2, 3); auto range = only(tuple); range.popFront(); /* ok */ } ----
May 25 2016
parent reply Chris <wendlec tcd.ie> writes:
On Wednesday, 25 May 2016 at 14:32:11 UTC, ag0aep6g wrote:
 On 05/25/2016 03:27 PM, Chris wrote:
 Why can the tuple be iterated with foreach, as in my quick 
 fix, and
 indexed with tuple[0..], but is not accepted as a range? What 
 are the
 differences?
popFront doesn't make sense with a tuple (aka expression list). When you remove the first element of a tuple, you get another tuple of a different type, but popFront can't change the type of the range. In code: ---- void main() { import std.meta: AliasSeq; AliasSeq!(int, int, int) tuple = AliasSeq!(1, 2, 3); tuple.popFront(); /* How would this be implemented? Would have to change tuple's type to AliasSeq!(int, int). */ } ----
 Is there a way to rangify a tuple?
std.range.only: ---- void main() { import std.meta: AliasSeq; import std.range: only; AliasSeq!(int, int, int) tuple = AliasSeq!(1, 2, 3); auto range = only(tuple); range.popFront(); /* ok */ } ----
I see. Maybe it would be worth adding a wrapper to typecons.Tuple or std.range that helps to rangify tuples. I cannot think of any use case right now. I never needed this and in the example that started this thread, it would keep the function from converting mixed tuples (cf. my example above).
May 25 2016
parent reply ag0aep6g <anonymous example.com> writes:
On 05/25/2016 04:39 PM, Chris wrote:
 I see. Maybe it would be worth adding a wrapper to typecons.Tuple or
 std.range that helps to rangify tuples.
std.range.only is that wrapper.
 I cannot think of any use case
 right now. I never needed this and in the example that started this
 thread, it would keep the function from converting mixed tuples (cf. my
 example above).
I'm not sure what you're saying here. Should the wrapper support tuples with different element types? Can't be a range then as a range has just one element type.
May 25 2016
parent Chris <wendlec tcd.ie> writes:
On Wednesday, 25 May 2016 at 14:48:14 UTC, ag0aep6g wrote:
 On 05/25/2016 04:39 PM, Chris wrote:
 I see. Maybe it would be worth adding a wrapper to 
 typecons.Tuple or
 std.range that helps to rangify tuples.
std.range.only is that wrapper.
Duh! Of course! :-)
 I cannot think of any use case
 right now. I never needed this and in the example that started 
 this
 thread, it would keep the function from converting mixed 
 tuples (cf. my
 example above).
I'm not sure what you're saying here. Should the wrapper support tuples with different element types? Can't be a range then as a range has just one element type.
I'm saying that for the above example something like std.range.only doesn't make sense, because the user might want to turn anything into string as in test(1, 2, "v", 4, 'c'); Mixed type tuples cannot be rangified, of course (and std.range.only takes care of that).
May 25 2016
prev sibling parent reply Era Scarecrow <rtcvb32 yahoo.com> writes:
On Wednesday, 25 May 2016 at 13:27:28 UTC, Chris wrote:
 Why can the tuple be iterated with foreach, as in my quick fix, 
 and indexed with tuple[0..], but is not accepted as a range? 
 What are the differences? Is there a way to rangify a tuple?
The tuple is identified/used at compile-time, as such it's a compiler primitive and not a range. Foreach in this case will unroll the loop regardless how it looks. So... test(Args...)(Args args) { ... foreach (const ref i; items) itemstrings ~= i.to!string; Will become: (const and ref are pointless in this example, unless the args are referenced) test(int arg1, int arg2, int arg3, int arg4) { ... itemstrings ~= arg1.to!string; itemstrings ~= arg2.to!string; itemstrings ~= arg3.to!string; itemstrings ~= arg4.to!string; Trying to use map on it was literally expanding the entire input to map. itemstrings => map!(to!string)(items); will become (all as individual items, not an array or range): itemstrings => map!(to!string)(arg1, arg2, arg3, arg4); You could probably use Array() inside which might work, assuming they are all the same type (or does it support a multi-type? I don't think so, unless you use Variant). There's a good section on Homogeneous vs Variadic behavior. Starts on page TDPL pg. 159-164 If you had the function declared as: test(int[] args...) then numbers passed with or without an array would be converted as an array. so your calling function would have an array which can be used with map, however it loses it's flexible template and variable types...
May 25 2016
parent Chris <wendlec tcd.ie> writes:
On Wednesday, 25 May 2016 at 19:07:32 UTC, Era Scarecrow wrote:
 On Wednesday, 25 May 2016 at 13:27:28 UTC, Chris wrote:
 Why can the tuple be iterated with foreach, as in my quick 
 fix, and indexed with tuple[0..], but is not accepted as a 
 range? What are the differences? Is there a way to rangify a 
 tuple?
The tuple is identified/used at compile-time, as such it's a compiler primitive and not a range. Foreach in this case will unroll the loop regardless how it looks. So... test(Args...)(Args args) { ... foreach (const ref i; items) itemstrings ~= i.to!string; Will become: (const and ref are pointless in this example, unless the args are referenced) test(int arg1, int arg2, int arg3, int arg4) { ... itemstrings ~= arg1.to!string; itemstrings ~= arg2.to!string; itemstrings ~= arg3.to!string; itemstrings ~= arg4.to!string; Trying to use map on it was literally expanding the entire input to map.
Ah, I didn't know that it was just unrolled. That makes sense, of course. [snip]
May 26 2016
prev sibling parent reply FreeSlave <freeslave93 gmail.com> writes:
On Tuesday, 24 May 2016 at 20:03:14 UTC, pineapple wrote:
 I would've expected this to work, but instead I get a compile 
 error. Is my syntax wrong? Is this just not a case that map can 
 handle, and I should be doing something else?

     import std.algorithm : map;
     import std.conv : to;
     import std.stdio : writeln;
     import std.string : join;

     string test(Args...)(in Args items){
         immutable string[items.length] itemstrings = 
 map!(to!string)(items);
         return join(itemstrings, ", ");
     }

     unittest{
         writeln(test(1, 2, 3, 4));
     }
Works with 'only', 'array' and static array slicing. import std.algorithm : map; import std.range : only; import std.conv : to; import std.stdio : writeln; import std.string : join; import std.array : array; string test(Args...)(in Args items){ immutable string[items.length] itemstrings = map!(to!string)(only(items)).array; return join(itemstrings[], ", "); } unittest{ writeln(test(1, 2, 3, 4)); }
May 25 2016
parent Chris <wendlec tcd.ie> writes:
On Wednesday, 25 May 2016 at 11:14:26 UTC, FreeSlave wrote:
 Works with 'only', 'array' and static array slicing.

 import std.algorithm : map;
 import std.range : only;
 import std.conv : to;
 import std.stdio : writeln;
 import std.string : join;
 import std.array : array;

 string test(Args...)(in Args items){
     immutable string[items.length] itemstrings = 
 map!(to!string)(only(items)).array;
     return join(itemstrings[], ", ");
 }

 unittest{
     writeln(test(1, 2, 3, 4));
 }
Nice, but I think it doesn't work with varying types. writeln(test(1, 2, "v", 4, 'c'));
May 25 2016