https://issues.dlang.org/show_bug.cgi?id=20798
ZombineDev <[email protected]> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |[email protected] --- Comment #1 from ZombineDev <[email protected]> --- The following works: struct S { string s; int field; } void main() { import std; auto someArr = 100.iota .map!(i => S("asd", i)) .array .randomShuffle; someArr.sort!((a, b) => a.field < b.field); someArr .map!(x => x.field) .assumeSorted!((a, b) => a < b) .equalRange(42) .writeln; } So where does this approach fall short? Performance, convenience, and accessibility? --
