digitalmars.D.learn - Check template parameter whether it has "length"
- tcak (18/18) Oct 08 2015 I am "trying" to write a function that takes an array of items,
- John Colvin (21/39) Oct 08 2015 I'm 99% sure something like __traits(hasMember, int[], "length" )
- tcak (13/36) Oct 08 2015 __traits(compiles, { size_t n = A.init.length; }); did the trick.
- John Colvin (3/4) Oct 08 2015 yeah, that's because __traits(hasMember, ...) should be good
- Meta (3/4) Oct 08 2015 You're just looking in the wrong place =)
- Artur Skawina via Digitalmars-d-learn (5/19) Oct 08 2015 Use `A.init.length` instead of `A.length`.
- Nicholas Wilson (3/21) Oct 08 2015 iirc there's a hasLength trait in std.traits e.g. static
I am "trying" to write a function that takes an array of items,
and returns the length of longest item.
[code]
size_t maxLength(A)( const A[] listOfString ) if( __traits(
hasMember, A, "length" ) )
{
return 0; // not implemented yet
}
[/code]
I tried it with
if( __traits( compiles, A.length ) )
as well. But compiler doesn't match it.
writeln("Max Length: ", maxLength( ["foo", "123456789"] ));
Compilers says it cannot deduce function from argument types ...
I do not want to check whether the type "A" is string, char[],
etc. As long as it has length (please do not put me into ranges,
library functions etc as much as possible), I want the function
to accept it.
Oct 08 2015
On Thursday, 8 October 2015 at 09:29:30 UTC, tcak wrote:
I am "trying" to write a function that takes an array of items,
and returns the length of longest item.
[code]
size_t maxLength(A)( const A[] listOfString ) if( __traits(
hasMember, A, "length" ) )
{
return 0; // not implemented yet
}
[/code]
I tried it with
if( __traits( compiles, A.length ) )
as well. But compiler doesn't match it.
writeln("Max Length: ", maxLength( ["foo", "123456789"] ));
Compilers says it cannot deduce function from argument types ...
I do not want to check whether the type "A" is string, char[],
etc. As long as it has length (please do not put me into
ranges, library functions etc as much as possible), I want the
function to accept it.
I'm 99% sure something like __traits(hasMember, int[], "length" )
should evaluate to true. Please file a bug at issues.dlang.org
I notice it also doesn't work for "ptr".
The correct workaround:
__traits(compiles, A.init.length ));
or
__traits(compiles, listOfStrings.length ));
A.length doesn't work because length is not a static member, so
it's only accessible from an instance.
The __traits(compiles, ...) solution is actually more general
because it will work if .length is implemented via UFCS and
opDispatch.
FYI:
If you want to check whether a statement will compile, as opposed
to an expression, make a function/delegate out of it, e.g.:
__traits(compiles, { size_t n = A.init.length; });
to check that A has a member length that can be assigned to
size_t.
P.S. always check std.traits for solutions all your static
reflection problems, there's a lot of good stuff in there.
Oct 08 2015
On Thursday, 8 October 2015 at 09:50:12 UTC, John Colvin wrote:On Thursday, 8 October 2015 at 09:29:30 UTC, tcak wrote:__traits(compiles, { size_t n = A.init.length; }); did the trick. [code] size_t maxLength(A)( const A[] listOfString ) if( __traits( compiles, { size_t len = A.init.length; } ) ) { size_t len = 0; foreach(A str; listOfString) if( str.length > len ) len = str.length; return len; } [/code] BTW, there is nothing like std.traits.hasLength.[...]I'm 99% sure something like __traits(hasMember, int[], "length" ) should evaluate to true. Please file a bug at issues.dlang.org I notice it also doesn't work for "ptr". The correct workaround: __traits(compiles, A.init.length )); or __traits(compiles, listOfStrings.length )); A.length doesn't work because length is not a static member, so it's only accessible from an instance. The __traits(compiles, ...) solution is actually more general because it will work if .length is implemented via UFCS and opDispatch. FYI: If you want to check whether a statement will compile, as opposed to an expression, make a function/delegate out of it, e.g.: __traits(compiles, { size_t n = A.init.length; }); to check that A has a member length that can be assigned to size_t. P.S. always check std.traits for solutions all your static reflection problems, there's a lot of good stuff in there.
Oct 08 2015
On Thursday, 8 October 2015 at 15:22:02 UTC, tcak wrote:BTW, there is nothing like std.traits.hasLength.yeah, that's because __traits(hasMember, ...) should be good enough, but obviously not in this case at the moment.
Oct 08 2015
On Thursday, 8 October 2015 at 15:22:02 UTC, tcak wrote:BTW, there is nothing like std.traits.hasLength.You're just looking in the wrong place =) http://dlang.org/phobos/std_range_primitives.html#hasLength
Oct 08 2015
On 10/08/15 11:29, tcak via Digitalmars-d-learn wrote:
I am "trying" to write a function that takes an array of items, and returns
the length of longest item.
[code]
size_t maxLength(A)( const A[] listOfString ) if( __traits( hasMember, A,
"length" ) )
{
return 0; // not implemented yet
}
[/code]
I tried it with
if( __traits( compiles, A.length ) )
as well. But compiler doesn't match it.
Use `A.init.length` instead of `A.length`.
You could also use something like
if(is(typeof(A.init.length):size_t))
artur
Oct 08 2015
On Thursday, 8 October 2015 at 09:29:30 UTC, tcak wrote:
I am "trying" to write a function that takes an array of items,
and returns the length of longest item.
[code]
size_t maxLength(A)( const A[] listOfString ) if( __traits(
hasMember, A, "length" ) )
{
return 0; // not implemented yet
}
[/code]
I tried it with
if( __traits( compiles, A.length ) )
as well. But compiler doesn't match it.
writeln("Max Length: ", maxLength( ["foo", "123456789"] ));
Compilers says it cannot deduce function from argument types ...
I do not want to check whether the type "A" is string, char[],
etc. As long as it has length (please do not put me into
ranges, library functions etc as much as possible), I want the
function to accept it.
iirc there's a hasLength trait in std.traits e.g. static
assert(hasLength!int[])
Oct 08 2015









John Colvin <john.loughran.colvin gmail.com> 