Re^2: Merging arrays

... elements of @array must be unique...

That strikes me as a drawback. After all, there seems to be nothing inherent in the organization of the two source arrays to suggest such a limitation. Why shouldn't one be allowed to end up with data like:

my @AoH = (
   { 111 => "acct1" },
   { 111 => "hoo"   },
   { 111 => "ha"    },
   { 111 => "foo"   },
   { ... => "..."   },
   { 888 => "acct8" },
   );
[download]

Comment on Re^2: Merging arrays Download Code

Replies are listed 'Best First'.
Re^3: Merging arrays by LanX (Saint) on Jun 28, 2013 at 13:50 UTC
> Why shouldn't one be allowed to end up with data like: No special reason. Sometimes one has data structure which have to be unique, then using a hash is handy. Sometimes not ... I'm just listing another way to do it for one of those loosely defined PM requirements... :) FWIW I'll change my post to make the restriction clearer ... update The question is rather how to design functional constructs like `map` which can easily (idiomatically) be used/combined to create nested data-structures which solve all possible requirements (sorted, nested, unique, non-string keys, ...). Thats more a design than an implementation question, I rarely see people using List::MoreUtils here. Cheers Rolf ( addicted to the Perl Programming Language)	[reply] [d/l]
Re^4: Merging arrays by AnomalousMonk (Archbishop) on Jun 28, 2013 at 19:26 UTC
... loosely defined ... requirements... Yes, that's always the tricky bit: to divine what the OPer really wants... or, better yet, needs!	[reply]

update