feedback1b WG: item merges re-checked; returned equipment auto-merges
Player report after 0.1.0: "源石溶剂两个白的合成一个金色的之后又多给一个白的". Not reproduced: buying next to a white (hand / equipped), buying twice, equip on another or a full carrier, reward pick, 道具补给, 列装, 整备, 画卷 and the sale of a carrier all end with one golden and no white; random intents over all 50 mergeable items conserve the white-equivalents and never leave a pair. The one gap: equipment a promotion (PlayerState._mergeChess) or an effect that removes an operator (ctx.destroyPiece) returns to the hand skipped the auto-merge. Unreachable from a normal state (a pair never exists), closed: both now run checkItemMerges.
This commit is contained in:
1 parent
b4fa21a0a6
commit
8fa7a85a52
4 files changed
+185
-2
No files matched your search
+3
-1
@@ -331,7 +331,9 @@ Other equipment: 2 slots; a third replaces the equipped item the player picks in
|
||||
equipped on the target ⇒ `BAD_TARGET`, nothing changes) — and the replaced item is destroyed. Equipped items are
|
||||
otherwise locked (research 04 §2 / addendum: they leave the operator only on promotion, merge or sale): `g.destroy`
|
||||
refuses them (`BAD_TARGET 'equipped items are locked'`). A second copy of an equipped normal item merges into the golden
|
||||
item in the hand.
|
||||
item in the hand. Every path that hands an item to the player (buy, reward, 机变, grants, equip, the equipment a sale,
|
||||
a promotion or `ctx.destroyPiece` returns) ends with the auto-merge (`acquireItem` / `checkItemMerges`: "已拥有2件同一初始
|
||||
装备时…自动合并"), so a player never holds two identical mergeable normal items (`test/match/feedback1b-items.test.js`).
|
||||
|
||||
Built-ins (builtinMeta.js, overridable): 盟约之币 / 骑士储蓄罐 (random funds), 随身身份牌 (layers of the target's bonds),
|
||||
紧急调度券 (take shop chess), 精打细算玩偶 (+funds each round), 简易通讯机 / 拟态物质 (same-bond chess), 见钱眼开玩偶
|
||||
|
||||
Reference in new issue
Block a user