I took a look through my codebase to see where I have used flatmap. Every
place I think I would prefer map({ }).flat on second look.so +1 LL On Fri, Jan 6, 2017 at 4:47 PM Sam S. <perl6-bugs-follo...@perl.org> wrote: > # New Ticket Created by Sam S. > # Please include the string: [perl #130520] > # in the subject line of all future correspondence about this issue. > # <URL: https://rt.perl.org/Ticket/Display.html?id=130520 > > > > So, Perl 6 has a built-in `flatmap` routine: > > .flatmap({ ... }) > > Show of hands: Who is reasonably sure, without testing it or looking > it up, which of the following four expressions it corresponds to? > > .flat.map({ ... }) > > .map({ ... }).flat > > .flat.map({ ... }).flat > > .map({ slip ... }) > > And if Perl 6 regulars and core developers can't tell, do you think a > typical Perl 6 newbie will guess correctly, and that code reviewers > will notice when the wrong behavior was assumed? > > Note that the four behaviors are the same (or at least easy to > confuse) for many simple cases - so it's quite possible to assume the > wrong one even after a quick test, if one didn't test the more > complicated inputs for which where they do differ. > > --- > > Second strike: Whoever wrote the p6doc entry > <https://docs.perl6.org/routine/flatmap> assumed the wrong behavior as > well, and no one else seems to have noticed or cared. (See my initial > report at <https://github.com/perl6/doc/issues/851>.) > > Third strike: Its behavior isn't actually tested in Roast. The single > Roast test that exists for it > <https://github.com/perl6/roast/blob/master/S03-metaops/hyper.t#L410> > merely tests that it is nodal, and doesn't distinguish between the > four possible behaviors listed above. > > Fourth strike: Whereas `deepmap` and `duckmap` differ from plain `map` > in how they iterate the input and how often they call the lambda, > `flatmap` is the same as `map` in those regards and differs, rather, > in how it collects the output. Users who take the naming scheme > "deepmap, duckmap, flatmap" as an indication that these are three of a > kind, will assume the wrong behavior for `flatmap`. > > --- > > Is all that confusion (and the needless proliferation of the Perl 6 > core setting) worth it, just to save typing a single character in a > specific circumstance? > > Unless I'm missing something vital, I'd say it's hard to argue that > the answer is "yes". > > I therefore suggest initiating the deprecation cycle in order to > eventually drop of this routine from the Perl 6 language and Rakudo > implementation. >
