On 1/2/19 10:17 PM, ToddAndMargo via perl6-users wrote:
 > On Wed, Jan 2, 2019 at 8:41 PM ToddAndMargo via perl6-users
 > <perl6-users@perl.org> wrote:
 >>
 >> Hi All,
 >>
 >> Looking at
 >>
 >>      https://docs.perl6.org/routine/print
 >>
 >> I see
 >>
 >>      multi sub print(**@args --> True)
 >>
 >> Question.  If I wanted to create my own print routine
 >> using **@args, how would I declare it?
 >>
 >>      sub printx( **@args data ) {...}
 >>
 >>
 >> Many thanks,
 >> -T


On 1/2/19 10:06 PM, Brad Gilbert wrote:
The variable name in :(**@args) is @args, it could be any valid array
identifier like @data

     sub printx( **@data ){…}

Note that there are several forms for slurpy positional arguments.

     :(   @a ) # not slurpy at all, takes one Positional argument and
aliases it as @a.

     :(  *@a ) # flattening slurpy  (1,),2,3 → 1,2,3 and ((1,),2,3) → 1,2,3
     :( **@a ) # structured slurpy (1,),2,3 → (1,),2,3 and ((1,),2,3) →
((1,),2,3) # note there is no change
     :(  +@a ) # one-arg rule slurpy (1,),2,3 → (1,),2,3 and ((1,),2,3)
→ (1,),2,3 # note that the second one changes

Note that itemized values do not get flattened for :(*@a), and that
Array values are itemized.

I like to consider the one-arg rule slurpy :(+@a) to be like a
combination between non-slurpy :(@a) and structured slurpy :(**@a)
That is a one-arg rule slurpy will sometimes be like an alias to a
singular positional, otherwise it will act like a structured slurpy.

Note that all but the [aliasing] non-slurpy :(@a) are [almost] always Array's.

---

The one-arg rule slurpy is the oddest one of the bunch so here is some
brief information:

The one-arg rule slurpy can be sigiless :(+a), in which case it will
be a List instead of an Array or an alias to the single argument
depending on what it was.

     sub one-arg-rule ( +args ){
         say args.^name, " # ", args.perl;
     }

     one-arg-rule(  (1,),2,3  ); # List # ((1,), 2, 3)
     one-arg-rule( ((1,),2,3) ); # List # ((1,), 2, 3) # one-arg rule
     one-arg-rule( ((1,),2,3).Seq ); # Seq # ((1,), 2, 3).Seq # one-arg rule
     one-arg-rule(  1..3   ); # List # (1, 2, 3)
     one-arg-rule( (1..3,) ); # List # (1..3,)

     sub one-arg-Array ( +@args ){
         say @args.^name, " # ", @args.perl;
     }

     one-arg-Array(  (1,),2,3  ); # Array # [(1,), 2, 3]
     one-arg-Array( ((1,),2,3) ); # Array # [(1,), 2, 3]
     one-arg-Array( ((1,),2,3).Seq ); # List # ((1,), 2, 3)
     one-arg-Array(  1..3   ); # Array # [1, 2, 3]
     one-arg-Array( (1..3,) ); # Array # [1..3,]

The one-arg rule exists because people tend to write the following:

     my @a = [1,2,3];

When the correct way to write it is:

     my @a = 1,2,3;

There are various other places where the one-arg rule is also used, so
it was made available to everyone.

---

Don't get too flustered if you don't understand all of the nuances, it
take everybody a while to understand them.


Thank you!

The term "slurpy" did help a lot.

:-)

I am writing your explanation down for my records.


Well Golly!

$ p6 'sub printx(**@args){print(@args)}; printx("abc",1,"def\n");'
abc 1 def

$ p6 'sub printx(**@args){print @args, "\n"}; printx("abc",1,"def");'
abc 1 def

Question:
$ p6 'sub printx(**@args){print(@args)}; printx("abc","xyz","def\n");'
abc xyz def

Are the spaces suppose to be there?

Reply via email to