01:25 shareable6 left 01:27 shareable6 joined 09:53 bloatable6__ left, Voldenet left, lizmat left, coleman left, evalable6 left, gfldex left, disbot2 left, kurahaupo left, ShimmerFairy left, sivoais left, summerisle left, [Coke] left, rba left, SmokeMachine left, JRaspass left 10:13 kurahaupo joined, bloatable6__ joined, Voldenet joined, ShimmerFairy joined, [Coke] joined, lizmat joined, rba joined, coleman joined, evalable6 joined, gfldex joined, SmokeMachine joined, disbot2 joined, sivoais joined, summerisle joined, JRaspass joined 10:18 sivoais left 12:52 timo left
Geth rakudo/main: 42ef8973b5 | (Elizabeth Mattijsen)++ | src/core.c/Exception.rakumod
Small optimization in error messaging

Hopefully handling strange Mu values more user-friendly
13:45
rakudo/main: f2273bf17a | (Elizabeth Mattijsen)++ | src/core.c/CompUnit/Repository/Installation.rakumod
Fix more trailing whitespace
lizmat m: dd <a b c>.Set (|) Map.new(:42a, :666d) 13:56
evalable6 Bool = Set.new("d","a","c","b")
disbot2 <librasteve> interesting… can you make a Bag or Mix this way too? 14:47
lizmat m: dd <a b c> (+) <b c d> 15:11
evalable6 UInt = ("c"=>2,"a"=>1,"b"=>2,"d"=>1).Bag
lizmat m: dd <a b c>.MixHash (|) Map.new(:42a, :666d)
evalable6 Real = ("c"=>1,"a"=>42,"d"=>666,"b"=>1).MixHash
lizmat rules are: upgrade to the highest degree of value (Set -> Bag -> Mix) = (Bool -> Uint -> Real) 15:12
mutability is determined by the left hand side
Geth rakudo/main: 5c37ccea1e | (Elizabeth Mattijsen)++ | 24 files
Initial completion of what became the Great QuantHash Refactor

What started out as implementing support for Baggies that were parameterized with a coercive type, turned into a basically full-blown refactor of anything QuantHash related.
This work could probably have been done in separate commits, but ... (48 more lines)
16:31
16:31 timo joined
Geth roast: 9b48149b63 | (Elizabeth Mattijsen)++ | S03-operators/set_difference.t
Disable some tests to make the file as a whole pass

This appears to tickle a dispatch bug somewhere causing strange VMNull errors. Testing with (-) by itself works, and testing with
  ∖ also works. But testing them both in a loop after each other,
breaks tests on the second iteration.
Since this is repeatable, it should probably be investigated further
16:34
roast: e0dde57800 | (Elizabeth Mattijsen)++ | 3 files
Change immutable QuantHash errors

Following the changes in github.com/rakudo/rakudo/commit/5c37ccea1e
Basically,
   <a b c>.Bag<a> = 42
will now complain with:
   Cannot modify an immutable Bag (Bag(a b c))
rather than:
   Cannot change immutable Int 1
lizmat *phew* that's a load off... :-) 16:37
[Coke] wonders why releaseable isn't seeing 5c37ccea1e
lizmat too soon? 16:38
[Coke] yah, sometimes it's immediate, sometimes it takes a minute. 16:40
anyway, changelog up to date. 16:44
Geth roast: d997363b38 | (Elizabeth Mattijsen)++ | S02-types/set.t
Make sure we test for Str, not potential allomorphs
18:22
roast: de734dfe4d | (Elizabeth Mattijsen)++ | S03-operators/set_addition.t
Use the normal way to create a mutable QuantHash

  .new() used to *not* create the internal IterationSet, but now it
does, so no special action necessary anymore
rakudo/main: d8f907ee5c | (Elizabeth Mattijsen)++ | src/core.c/Rakudo/QuantHash.rakumod
Make sure OBJECTIFIER for coercive type parameterization throws

Turns out that ^coerce will return a Failure on coercion failure, rather than throw. Make sure that that objectifier throws when that happens
19:47
rakudo/main: 9e6d078e6c | (Elizabeth Mattijsen)++ | 3 files
Make sure AT-KEY on mutable quanthashes caches object/which

Turns out that the FETCH part of a Proxy may be called many times for the same key: so make sure we do the objectification and the whichification only once when the Proxy is made, instead of every time we do a FETCH or a STORE
roast: 7cbea54923 | (Elizabeth Mattijsen)++ | 2 files
Add some basic QuantHash coercive type parameterization tests

Just the really basic features for now.
19:48
japhb lizmat++ # Your turn to kick ass it looks like :-D 19:50
lizmat heh... well, it was a big project: about 1800 lines added and 2150 lines removed 19:54
japhb Nice that removed was nicely above added, while improving functionality and correctness at the same time. 19:59
lizmat time for some R&R :-) 20:00