On Sun, Nov 29, 2020 at 6:38 PM Ralph Mellor <ralphdjmel...@gmail.com> wrote:
> > Zen slicing as a possible way of 'de-containerizing' : > > https://docs.raku.org/language/subscripts#index-entry-Zen_slices > > A zen-slice only affects the single reference it's applied to. > > And it is a no op when applied to anything other than a `Scalar`. > > So it'll have no effect when applied directly to an array, list, hash, > or other non `Scalar`. > > Perhaps you're thinking "decontainerizing" does something other > than what it does? It's really a very simple operation. An analogy > is having a bunch of folk, some of whom always go by their legal > name, but others who sometimes go by their nickname. And let's > say there's an operation called "use legal name". When applied > to a nickname it yields the corresponding legal name. But when > applied to a legal name it just yields that name, i.e. it's a no op. > > That's all that decont does, except the analog to a nickname is > a `Scalar`, and the analog to a legal name is anything else. > > > Even if the ",=" postfix operator were to gain this ability on non-hash > > objects, then hash-objects would be special-cased in **not** requiring > > a Zen-sliced decontainerized object on the LHS, so people would have > > to consider that outcome. > > I think the focus on `,=`.is really unfortunate, because the real issue is > just about what `=` does with a non-`Scalar` on its LHS and a list of > values on its RHS. It's really got nothing whatsoever to do with `,`. > > It just depends on the type of non-`Scalar` on the LHS of `=`. > > Look what happens without not a comma in sight: > > my %hash = do for ^1 { %hash } > say %hash; # {} > > If it's a hash,, the only sensible thing to do with a list of values that > may > each be a pair, or a list of pairs, or an array of pairs, or a hash of > pairs, > is to flatten and concatenate all the individual pairs into the hash. > > my @array = do for ^1 { @array } > say @array; # (\Array_73652568 = [Array_73652568]) > > But if it's an array, the sensible thing to do is to *not* flatten by > default. > This topic was a subject of discussion during our (SF_Perl) Raku Meetup Group today. We reviewed the three URL references below. One member of our Meetup went throught the 2017 Advent Calendar reference extensively (hopefully we'll see a post on that). Afterwards, I played a bit with indexing constructs. Posting these for posterity's sake: https://docs.raku.org/language/subscripts#index-entry-Zen_slices https://docs.raku.org/language/glossary#index-entry-decont https://perl6advent.wordpress.com/2017/12/02/#theoneandonly user@mbook:~$ raku Welcome to 𝐑𝐚𝐤𝐮𝐝𝐨™ v2020.10. Implementing the 𝐑𝐚𝐤𝐮™ programming language v6.d. Built on MoarVM version 2020.10. You may want to `zef install Readline` or `zef install Linenoise` or use rlwrap for a line editor To exit type 'exit' or '^D' > my %hash = do for ^1 { %hash } {} > my %hash2 = do for ^1 { %hash2<> } {} > my %hash3 = do for ^1 { %hash3<*> } Odd number of elements found where hash initializer expected: Only saw 1 element in block <unit> at <unknown file> line 1 > my @array = do for ^1 { @array } (\Array_140587615723064 = [Array_140587615723064]) > my @array2 = do for ^1 { @array2[] } (\Array_140587615723904 = [Array_140587615723904]) > my @array3 = do for ^1 { @array3[*] } [()] Best Regards, Bill.
