Repository navigation
New named parameters to .classify #6
Description
Id like to suggest to add:
:&reduce:&produce:$initial-value
so,
^10.classify: * %% 2, :reduce(&[+])would return
{True => 25, False => 20},
^10.classify: * %% 2, :produce(&[+])would return
({}, {True => 0}, {True => 0, False => 1}, {True => 2, False => 1}, {True => 2, False => 4}, {True => 6, False => 4}, {True => 6, False => 9}, {True => 12, False => 9}, {True => 12, False => 16}, {True => 20, False => 16}, {True => 20, False => 25}).Seqand
^5.classify: * %% 2, :produce(-> %agg, $i { %agg{$i} = <a e i o u>[$i]; %agg }), :initial-value{}would return
{True => {0 => "a", 2 => "I", 4 => "u"}, False => {1 => "e", 3 => "o"}}That would also be interesting to add the :$initial-value (or with any name) to .reduce and .produce
Activity
AlexDaniel commentedon Apr 5, 2019
And which problem are we trying to solve here exactly?
FCO commentedon Apr 5, 2019
I had to do that by hand some times, and ended up not using classify because it would be worst to write and have to loop throw each key and reduce each value (on reduce case)... making the classify do that, you can reduce/produce iterating only once...
Altai-man commentedon Apr 5, 2019
Why not just:
for the first one? Because it adds some symmetry with
produce?FCO commentedon Apr 5, 2019
It iterates twice each element...
and what about multi level classify?
AlexDaniel commentedon Apr 5, 2019
Do we have some examples of this functionality being used/implemented in real code? Maybe in the ecosystem?
jnthn commentedon Apr 5, 2019
Does this proposal apply to
categorizealso?jnthn commentedon Apr 5, 2019
How would this interact with the current semantics of things like
[+] ()and[+] 1, which use multi-dispatch on the target operator?Further, how is it better than
(flat $initial-value, @values).reduce(&reducer)?AlexDaniel commentedon Apr 5, 2019
IRC discussion: https://colabti.org/irclogger/irclogger_log/perl6-dev?date=2019-04-05#l181
lizmat commentedon Jan 13, 2025
Updated IRC discussion link: https://irclogs.raku.org/perl6-dev/2019-04-05.html#23:03-0003
lizmat commentedon Feb 22, 2026
@FCO is this still something that would need to be pursued this way?
If not, then suggest closing this issue.
FCO commentedon Feb 22, 2026
I think that is/would be useful. Maybe I should create a module for that to help on the discussion…
lizmat commentedon Feb 22, 2026
Going to close this issue in the meantime, ok?