DNF90 服务端与启动器完整源码(含 GM 工具功能增强)
服务端(Go): - GM 塞背包/带属性装备发放:在线角色推送 op14 增量物品行,背包实时刷新,无需回角色选择界面 - GM 改金币/改等级:去掉在线拒绝,落库后推送 op14 钱包/op37 等级状态,在线实时刷新 - 等级端点统一切换到新版 GMSetCharacterLevel(自带在线刷新) - 无限/锁死疲劳模式 + 燃烧疲劳:每房间消耗 2 点、打怪与通关经验双倍 - 燃烧疲劳开关时给所有在线角色推送系统提示(op12 mode 0) - 深渊史诗掉率:GM 全局倍率 + 城镇增益药剂窗口倍率,真实接入掉落结算 启动器(Python/JS): - 标题栏自绘、窗口 1204x717、禁止拖拽改大小、仅最小化/关闭 - GM 工具:疲劳与燃烧、深渊史诗掉率、拍卖行上架(单一定价对齐) - 拍卖行上架前端对齐服务端单一价格语义,修复 400 报错
This commit is contained in:
commit
645b64fe21
4388 files changed
+1405457
No files matched your search
@@ -0,0 +1,109 @@
|
||||
package dnfbridge
|
||||
|
||||
// 攻坚阶段死亡计数 —— 0x0259 (601 RAID_RESULT) 的 deaths 字段。
|
||||
//
|
||||
// 为什么要有这个文件
|
||||
//
|
||||
// 客户端的阶段结算条把 0x0259 的第 4 个字段画成「死亡次数」(`raid.RaidSettlementResult`
|
||||
// 的 `DeathCount`,u16,`raid/result_flow_protocol.go:44` + `:117`),而
|
||||
// `publishRuntimeRaidPhaseSettlement` 一直只填 ResultType/PhaseIndex/ClearSeconds
|
||||
// —— 这个 u16 恒为 0。服主 2026-09-29 实测单人卢克:「我确定死亡了并且用复活币了的」
|
||||
// 但结算条写着「死亡次数 0」。
|
||||
//
|
||||
// luke.etc 的 `[CONDITION FOR RANK]` 正是按 `[DEAD COUNT]` 分档
|
||||
// (`{0,1}→0 / {2,5}→1 / {6,9}→2 / {10,14}→3`,`luke_pvf_test.go:136-141`),
|
||||
// 86JP 也是 `RAID_RESULT(0, phase, clearTime, deaths, rank+2, rewardOption)`
|
||||
// (`86raid_flow_report.md:100`)—— 所以这个字段本来就是"本阶段全团死亡次数"。
|
||||
//
|
||||
// 口径(本实现的约定,写明以免后人猜)
|
||||
//
|
||||
// 计数边沿 = **服务端接受的 C2S op40 `DIE_CHARACTER`**
|
||||
// (`handleDungeonDieCharacterLocal` 通过全部守卫、真正发出 op32 的那一步)。
|
||||
// 一次死亡一条,与"是否用复活币站起来"无关:两种死亡都该被统计。
|
||||
// 粒度 = (RaidKey, Phase),即本阶段整团;阶段结束时随攻坚一起清零。
|
||||
//
|
||||
// 适用范围:**所有团本**(卢克 + 安徒恩,团队与单人阶段都是)。
|
||||
//
|
||||
// 2026-09-29 服主拍板:这个值本来就是该有值的字段,恒 0 是缺陷。依据是
|
||||
// 86JP 的 `RAID_RESULT(0, phase, clearTime, deaths, rank+2, rewardOption)`
|
||||
// (`86raid_flow_report.md:100`)——第 4 个参数就是死亡次数;而 luke.etc 的
|
||||
// `[CONDITION FOR RANK]` 也正是按 `[DEAD COUNT]` 分档
|
||||
// (`{0,1}→0 / {2,5}→1 / {6,9}→2 / {10,14}→3`,`luke_pvf_test.go:136-141`),
|
||||
// 两者都说明这个 u16 是"本阶段全团死亡次数",与阵营无关。
|
||||
//
|
||||
// 在 2026-09-29 之前这里有一道 `runtimeRaidSettlementCarriesDeathCount` 的
|
||||
// owner 判断,只有卢克写这个字段(当时安徒恩 0x0259 要求逐字节不变)。服主
|
||||
// 拍板后该判断已删除:**安徒恩本次唯一有意的行为变化就是这个 u16**,其它字节
|
||||
// (header / ClearSeconds / Score / Flags / ResultType / PhaseIndex)逐字节不变,
|
||||
// 由 TestAntonPhaseSettlementCarriesThePhasesDeathCount 逐字节对照钉住。
|
||||
|
||||
// runtimeRaidPhaseDeathKey names one raid phase's death counter.
|
||||
type runtimeRaidPhaseDeathKey struct {
|
||||
RaidKey uint32
|
||||
Phase byte
|
||||
}
|
||||
|
||||
// recordRuntimeRaidPhaseDeath counts one accepted character death inside a raid
|
||||
// gate for its phase.
|
||||
//
|
||||
// The caller (handleDungeonDieCharacterLocal) holds session.dungeon.mu and this
|
||||
// takes s.raidMu, which is the order the raid lifecycle already uses -- the
|
||||
// inverse is what would deadlock (same rule as recordRuntimeRaidReviveCoinUseLocked).
|
||||
func (s *Service) recordRuntimeRaidPhaseDeath(session *gameSession, runtime *runtimeDungeonState) {
|
||||
if s == nil || session == nil || runtime == nil {
|
||||
return
|
||||
}
|
||||
binding := runtime.raidDungeon
|
||||
if binding.RaidKey == 0 || binding.MapID == 0 {
|
||||
return
|
||||
}
|
||||
key := runtimeRaidPhaseDeathKey{RaidKey: binding.RaidKey, Phase: binding.Phase}
|
||||
s.raidMu.Lock()
|
||||
state := s.raids[binding.RaidKey]
|
||||
if state == nil || state.Lifecycle.Phase != binding.Phase {
|
||||
// The gate belongs to a phase that is no longer running: a death that
|
||||
// arrives after the settlement is not part of any result body.
|
||||
s.raidMu.Unlock()
|
||||
return
|
||||
}
|
||||
if s.raidPhaseDeaths == nil {
|
||||
s.raidPhaseDeaths = make(map[runtimeRaidPhaseDeathKey]uint32, 4)
|
||||
}
|
||||
s.raidPhaseDeaths[key]++
|
||||
total := s.raidPhaseDeaths[key]
|
||||
s.raidMu.Unlock()
|
||||
s.logGameEvent(session, "game-raid-phase-death-counted",
|
||||
"char_id", session.selectedCharacterID,
|
||||
"raid_key", binding.RaidKey,
|
||||
"raid_phase", binding.Phase,
|
||||
"raid_map_id", binding.MapID,
|
||||
"phase_deaths", total)
|
||||
}
|
||||
|
||||
// runtimeRaidPhaseDeathCount is the u16 the 0x0259 result body carries for this
|
||||
// raid and phase.
|
||||
func (s *Service) runtimeRaidPhaseDeathCount(raidKey uint32, phase byte) uint16 {
|
||||
if s == nil || raidKey == 0 {
|
||||
return 0
|
||||
}
|
||||
s.raidMu.Lock()
|
||||
defer s.raidMu.Unlock()
|
||||
count := s.raidPhaseDeaths[runtimeRaidPhaseDeathKey{RaidKey: raidKey, Phase: phase}]
|
||||
if count > 0xffff {
|
||||
return 0xffff
|
||||
}
|
||||
return uint16(count)
|
||||
}
|
||||
|
||||
// purgeRuntimeRaidPhaseDeathsLocked drops one raid's phase death counters.
|
||||
// Caller holds s.raidMu.
|
||||
func (s *Service) purgeRuntimeRaidPhaseDeathsLocked(raidKey uint32) {
|
||||
if s == nil || raidKey == 0 {
|
||||
return
|
||||
}
|
||||
for key := range s.raidPhaseDeaths {
|
||||
if key.RaidKey == raidKey {
|
||||
delete(s.raidPhaseDeaths, key)
|
||||
}
|
||||
}
|
||||
}
|
||||
Reference in new issue
Block a user