From 5e0caa6110cafad2cd09dadc4811e35a98987c62 Mon Sep 17 00:00:00 2001 From: Michele Caini Date: Wed, 17 Jan 2024 17:41:25 +0100 Subject: [PATCH] *: updated TODO --- TODO | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/TODO b/TODO index 51c9ddc78..3d1b0a9af 100644 --- a/TODO +++ b/TODO @@ -21,11 +21,13 @@ TODO: * view: update natvis as needed after the last rework, merge pools/filter in the same array, drop check (?) and turn view into a position * view: type-only view_iterator (dyn get/excl sizes), type-only basic_common_view (dyn get/excl sizes with pointer to array from derived) * combine version-mask-vs-version-bits tricks with reserved bits to allow things like enabling/disabling -* basic_sparse_set(basic_sparse_set &&other, const allocator_type &allocator) uses moved-from packed in assert (same for storage) +* basic_sparse_set(basic_sparse_set &&other, const allocator_type &allocator) uses moved-from packed in assert +* basic_storage(basic_storage &&other, const allocator_type &allocator) uses moved-from packed in assert * maybe drop begin(v)/end(v) from sparse set and only use scoped begin/end at call sites -* view unchecked_refresh loop is never executed if Get is 1u * review all // NOLINT + linter workflow + review clang-tidy file * self contained entity traits to avoid explicit specializations (ie enum constants) * review assure conflicts check, hash doesn't fit the purpose maybe * allow to zero-sized versions (with non-regression tests) * auto type info data from types if present +* boxed_int -> boxed_type with known specializations +* sparse set sort_as should return the length to be a proper drop-in replacement for pack (then review registry destroy)