Restore dev-style FullDPS execution semantics

Author: Nostrademous

src/Modules/Calcs.lua

@@ -207,10 +207,6 @@ end
 function calcs.calcFullDPS(build, mode, override, specEnv)
 	local fullEnv, cachedPlayerDB, cachedEnemyDB, cachedMinionDB = calcs.initEnv(build, mode, override, specEnv)
 	local usedEnv = nil
-	local prepassState = calcs.performActorPrepass(fullEnv, { snapshotDBs = true })
-	local skillPassPlayerDB = prepassState.cachedPlayerDB or cachedPlayerDB
-	local skillPassEnemyDB = prepassState.cachedEnemyDB or cachedEnemyDB
-	local skillPassMinionDB = prepassState.cachedMinionDB or cachedMinionDB
 
 	local fullDPS = {
 		combinedDPS = 0,
@@ -233,16 +229,12 @@ function calcs.calcFullDPS(build, mode, override, specEnv)
 	local igniteSource = ""
 	local burningGroundSource = ""
 	local causticGroundSource = ""
-
-	-- initEnv() is the shared actor-global prepass for the FullDPS run. Each included
-	-- skill then runs through the hot skill-local path on the reused env, while perform()
-	-- continues to populate the slim cache for trigger/source reuse within the generation.
 	for _, activeSkill in ipairs(fullEnv.player.activeSkillList) do
 		if activeSkill.socketGroup and activeSkill.socketGroup.includeInFullDPS then
 			local activeSkillCount, enabled = getActiveSkillCount(activeSkill)
 			if enabled then
 				fullEnv.player.mainSkill = activeSkill
-				calcs.performSkillPass(fullEnv, true, prepassState)
+				calcs.perform(fullEnv, true)
 				usedEnv = fullEnv
 				local output = usedEnv.player.output
 				local minionOutput = usedEnv.minion and usedEnv.minion.output or nil
@@ -376,9 +368,9 @@ function calcs.calcFullDPS(build, mode, override, specEnv)
 					everything = true,
 				}
 				fullEnv, _, _, _ = calcs.initEnv(build, mode, override, {
-					cachedPlayerDB = skillPassPlayerDB,
-					cachedEnemyDB = skillPassEnemyDB,
-					cachedMinionDB = skillPassMinionDB,
+					cachedPlayerDB = cachedPlayerDB,
+					cachedEnemyDB = cachedEnemyDB,
+					cachedMinionDB = cachedMinionDB,
 					env = fullEnv,
 					accelerate = accelerationTbl,
 					cacheGeneration = fullEnv.cacheGeneration,
@@ -453,26 +445,13 @@ function calcs.buildActiveSkill(env, mode, skill, targetUUID, limitedProcessingF
 	end
 	recursionGuards.buildingSkills[guardKey] = true
 
-	local accelerationTbl = env.cachedSkillBuildEnv and {
-		nodeAlloc = true,
-		requirementsItems = true,
-		requirementsGems = true,
-		skills = true,
-		everything = true,
-	} or nil
-	local cachedSkillPrepassState = env.cachedSkillBuildPrepassState
 	local fullEnv, _, _, _ = calcs.initEnv(env.build, mode, env.override, {
-		cachedPlayerDB = cachedSkillPrepassState and cachedSkillPrepassState.cachedPlayerDB or env.cachedPlayerDB,
-		cachedEnemyDB = cachedSkillPrepassState and cachedSkillPrepassState.cachedEnemyDB or env.cachedEnemyDB,
-		cachedMinionDB = cachedSkillPrepassState and cachedSkillPrepassState.cachedMinionDB or env.cachedMinionDB,
-		env = env.cachedSkillBuildEnv,
-		accelerate = accelerationTbl,
+		cachedPlayerDB = env.cachedPlayerDB,
+		cachedEnemyDB = env.cachedEnemyDB,
+		cachedMinionDB = env.cachedMinionDB,
 		cacheGeneration = env.cacheGeneration,
 		recursionGuards = recursionGuards,
 	})
-	env.cachedSkillBuildEnv = fullEnv
-	local prepassState = cachedSkillPrepassState or calcs.performActorPrepass(fullEnv, { snapshotDBs = true })
-	env.cachedSkillBuildPrepassState = prepassState
 
 	-- env.limitedSkills contains a map of uuids that should be limited in calculation
 	-- this is in order to prevent infinite recursion loops
@@ -489,7 +468,7 @@ function calcs.buildActiveSkill(env, mode, skill, targetUUID, limitedProcessingF
 		if activeSkillUUID == targetUUID then
 			fullEnv.player.mainSkill = activeSkill
 			fullEnv.requestedSkillUUID = targetUUID
-			calcs.performSkillPass(fullEnv, true, prepassState)
+			calcs.perform(fullEnv, true)
 			fullEnv.requestedSkillUUID = nil
 			recursionGuards.buildingSkills[guardKey] = nil
 			return true