www.digitalmars.com         C & C++   DMDScript  

digitalmars.D - Can [] be made to work outside contexts of binary operators?

reply Shriramana Sharma <samjnaa_dont_spam_me gmail.com> writes:
I tried:

import std.stdio;
void main()
{
	int [5] vals = [1, 2, 3, 4, 5];
	writefln("A = %d, B = %d, C = %d, D = %d, E = %d", vals []);
}

but got thrown an exception that "%d is not a valid specifier for a range".

The Python equivalent to flatten a list works:

vals = [1, 2, 3, 4, 5]
print("A = {}, B = {}, C = {}, D = {}, E = {}".format(*vals))

Output:

A = 1, B = 2, C = 3, D = 4, E = 5

Question:

Can D's [] be made to work that way? I recently had to write custom 
functions since I had an array representing numerical fields and wanted to 
print them out with individual labels but I wasn't able to use a single 
writefln with sufficient specifiers for that purpose because of this 
limitation.

--
Oct 22 2015
next sibling parent reply Idan Arye <GenericNPC gmail.com> writes:
On Thursday, 22 October 2015 at 15:57:05 UTC, Shriramana Sharma 
wrote:

 I tried:

 import std.stdio;
 void main()
 {
 	int [5] vals = [1, 2, 3, 4, 5];
 	writefln("A = %d, B = %d, C = %d, D = %d, E = %d", vals []);
 }

 but got thrown an exception that "%d is not a valid specifier 
 for a range".

 The Python equivalent to flatten a list works:

 vals = [1, 2, 3, 4, 5]
 print("A = {}, B = {}, C = {}, D = {}, E = {}".format(*vals))

 Output:

 A = 1, B = 2, C = 3, D = 4, E = 5

 Question:

 Can D's [] be made to work that way? I recently had to write 
 custom functions since I had an array representing numerical 
 fields and wanted to print them out with individual labels but 
 I wasn't able to use a single writefln with sufficient 
 specifiers for that purpose because of this limitation.


D's `writefln` is a template-variadic function. Each time you use 
it, the compiler looks at the arguments you send to it, and 
compiles a new instantiation of it based on the number and types 
of these arguments. This means that it would have to know at 
compile time how many values `vals[]` holds - but that number is 
only known at runtime!

Now, in your case, since vals is a static array, it should be 
possible to know it's value at compile time. Maybe if there was a 
`tupleof` for static arrays?

At any rate, you can always use the range formatters %( and %) to 
print the array. See http://dpaste.dzfl.pl/47e3e5a9e5c4
Oct 22 2015
parent Meta <jared771 gmail.com> writes:
On Thursday, 22 October 2015 at 17:19:07 UTC, Idan Arye wrote:

 Now, in your case, since vals is a static array, it should be 
 possible to know it's value at compile time. Maybe if there was 
 a `tupleof` for static arrays?


Ask and ye shall receive.

http://forum.dlang.org/post/gkdqakdogqevwzntpgtu forum.dlang.org
Oct 22 2015
prev sibling parent Dmitry Olshansky <dmitry.olsh gmail.com> writes:
On 22-Oct-2015 18:57, Shriramana Sharma wrote:

 I tried:

 import std.stdio;
 void main()
 {
 	int [5] vals = [1, 2, 3, 4, 5];
 	writefln("A = %d, B = %d, C = %d, D = %d, E = %d", vals []);
 }

 but got thrown an exception that "%d is not a valid specifier for a range".

 The Python equivalent to flatten a list works:

 vals = [1, 2, 3, 4, 5]
 print("A = {}, B = {}, C = {}, D = {}, E = {}".format(*vals))

 Output:

 A = 1, B = 2, C = 3, D = 4, E = 5

 Question:

 Can D's [] be made to work that way? I recently had to write custom
 functions since I had an array representing numerical fields and wanted to
 print them out with individual labels but I wasn't able to use a single
 writefln with sufficient specifiers for that purpose because of this
 limitation.



Hm writeln supportы cool printing of any range might be not quite what 
you want though.

e.g.

auto arr = [1.23, 5.46, 6.21, 7.7711, 9.81121];
writeln("%(%.2f, %)", arr); // would print coma separated list



-- 
Dmitry Olshansky
Oct 22 2015