#2014 Pool NativeUI widgets instead of abandoning their frames

[LarryThiessen]

Fixes #2014

Release() only hid a widget's frame — and WoW frames can never be destroyed — so every editor redraw abandoned its entire widget set: measured +238 widgets per version click, permanently. Long sessions accumulate thousands of dead frames and megabytes of unreclaimable state. AceGUI pooled widgets for exactly this reason; the NativeUI rewrite dropped it.

Widgets are plain tables of closures over their own frame, so the same table is reused: UI:Create hands out a banked widget when one exists; Release banks poolable types. The reset contract — every part of which covers a real bug found during soak testing:

• every key added after construction is removed (caller decorations, wrapped methods);
• callbacks/children become fresh tables;
• frame scripts recorded at construction are restored on every subframe the widget exposes;
• regions and child frames callers created directly on the widget's frame are hidden on reuse — the key sweep cannot see those; without this a reused frame keeps rendering its previous life (header rows bleeding into macro blocks, doubled character counters, dead scrollbars intercepting the mouse wheel);
• a double Release cannot bank the same widget twice.

Only the 13 high-churn leaf/container types pool; windows, scroll frames, tab groups and the tree keep their old behaviour. Diagnostics kill switch: /run GSE_NoWidgetPool = true + /reload.

Best merged together with #2013 — pooled reuse of macro edit boxes is what makes the scrollbar re-show race fixed there common.

Testing

• luac -p clean.
• Confirmed in a running client: >95% pool reuse once warm, widget creation per click ~0, session memory stops climbing; soak-tested through normal editing (blocks, Tab builder, icons, checkboxes, dropdowns, config pages) with the state-bleed bugs above found and fixed during that soak.

🤖 Generated with Claude Code