Advent of bugfixes

[berni44]

On Monday, 9 December 2019 at 14:17:43 UTC, FeepingCreature wrote:
> Maybe add a line for each bug with the title of the bug report, 
> so people can see what got fixed?

Most of them are of the sort "wontfix", "invalid", "allready 
fixed", "duplicate", "is enhancement and not a bug", "wrong tool" 
and so on.  I think it's not really interesting to see these 
"fixes". That's the reason why I write "removed" and not 
"fixed"...

Having said this, there have been a few real fixes:

Day 1:
Issue 19105 - Bogus recursive template expansion via 
getSymbolsByUDA
               (unfortunatley meanwhile reopened)
Issue 20054 - getSymbolsByUDA no longer works on modules
Issue 20354 - interface is not supported by CanCAS in 
core.internal.atomic

Day 4:
Issue 16223 - BigUint: undefined shift for small instantiation 
type

Day 6:
Issue 19733 - expi documentation links broken
Issue 20396 - format!"%a" leeds to wrong result for denormalized 
float

Day 7:
Issue 9588 - format prints context pointer for struct
Issue 20350 - JSONType deprecations should be undeprecated
Issue 20416 - [Regression 2.073.2] compiler complains about 
escaping reference
               in certain cases

Day 8:
Issue 20364 - [REG2.069] changing length for typeof(null)[] array 
seg faults in
               _d_arraysetlengthiT()
Issue 20438 - [Reg 2.086] GC: memory not reusable when calling 
GC.collect after GC.free

And just to make it clear: Not all of these fixes have been made 
by me (indeed I made only 3 of these real ones) - I count *all* 
changes, looking at this as some sort of a community afford.