feat(agent): ✨ 重构智能排程分流与双通道交付,补齐周级预算并接入连续微调复用 - 🔀 通用路由升级为 action 分流(chat/quick_note_create/task_query/schedule_plan),路由失败直接返回内部错误,不再回落聊天 - 🧭 智能排程链路重构:统一图编排与节点职责,完善日级/周级调优协作与提示词约束 - 📊 周级预算改为“有效周保底 + 负载加权分配”,避免有效周零预算并提升资源利用率 - ⚙️ 日级并发优化细化:按天拆分 DayGroup 并发执行,低收益天(suggested<=2)跳过,单天失败仅回退该天结果并继续全局 - 🧵 周级并发优化细化:按周并发 worker 执行,单周“单步动作”循环(每轮仅 1 个 Move/Swap 或 done),失败周保留原方案不影响其它周 - 🛰️ 新增排程预览双通道:聊天主链路输出终审文本,结构化 candidate_plans 通过 /api/v1/agent/schedule-preview 拉取 - 🗃️ 增补 Redis 预览缓存读写与清理逻辑,新增对应 API、路由、模型与错误码支持 - ♻️ 接入连续对话微调复用:命中同会话历史预览时复用上轮 HybridEntries,避免每轮重跑粗排 - 🛡️ 增加复用保护:仅当本轮与上轮 task_class_ids 集合一致才复用;不一致回退全量粗排 - 🧰 扩展预览缓存字段(task_class_ids/hybrid_entries/allocated_items),支撑微调承接链路 - 🗺️ 更新 README 5.4 Mermaid(总分流图 + 智能排程流转图)并补充决策文档 - ⚠️ 新增“连续微调复用”链路我尚未完成测试,且文档状态目前较为混乱,待连续对话微调功能真正测试完成后再统一更新
566 lines
17 KiB
Go
566 lines
17 KiB
Go
package logic
|
||
|
||
import (
|
||
"fmt"
|
||
|
||
"github.com/LoveLosita/smartflow/backend/conv"
|
||
"github.com/LoveLosita/smartflow/backend/model"
|
||
"github.com/LoveLosita/smartflow/backend/respond"
|
||
)
|
||
|
||
type slotStatus int
|
||
|
||
const (
|
||
Free slotStatus = iota // 0: 纯空闲
|
||
Occupied // 1: 已有课/任务,不可动
|
||
Blocked // 2: 用户屏蔽时段
|
||
Filler // 3: 水课,允许嵌入
|
||
)
|
||
|
||
type slotNode struct {
|
||
Status slotStatus
|
||
EventID uint // 🚀 关键:记录课程 ID,用于识别水课边界
|
||
}
|
||
|
||
type grid struct {
|
||
data map[int]map[int][13]slotNode
|
||
startWeek int
|
||
startDay int
|
||
endWeek int
|
||
endDay int
|
||
}
|
||
|
||
// getNode 和 setNode 是对 grid 数据结构的封装,确保我们在访问时能正确处理默认值(Free)和边界情况
|
||
func (g *grid) getNode(w, d, s int) slotNode {
|
||
if dayMap, ok := g.data[w]; ok {
|
||
return dayMap[d][s]
|
||
}
|
||
return slotNode{Status: Free, EventID: 0}
|
||
}
|
||
|
||
func (g *grid) setNode(w, d, s int, node slotNode) {
|
||
if _, ok := g.data[w]; !ok {
|
||
g.data[w] = make(map[int][13]slotNode)
|
||
}
|
||
dayData := g.data[w][d]
|
||
dayData[s] = node
|
||
g.data[w][d] = dayData
|
||
}
|
||
|
||
// 检查是否可用 (Free 或 Filler 且不在 Blocked 时段内)
|
||
func (g *grid) isAvailable(w, d, s int) bool {
|
||
node := g.getNode(w, d, s)
|
||
return node.Status == Free || node.Status == Filler
|
||
}
|
||
|
||
// countAvailableSlots 统计指定周次范围内所有可用的原子节次总数
|
||
func (g *grid) countAvailableSlots(currW, currD, currS int) int {
|
||
count := 0
|
||
if currW == 0 && currD == 0 && currS == 0 {
|
||
currW, currD, currS = g.startWeek, g.startDay, 1
|
||
}
|
||
for w := currW; w <= g.endWeek; w++ {
|
||
dayMap, hasData := g.data[w]
|
||
for d := 1; d <= 7; d++ {
|
||
// 🚀 头部裁剪:过滤开始日期前的天数
|
||
if w == currW && d < currD {
|
||
continue
|
||
}
|
||
// 🚀 尾部裁剪:过滤结束日期后的天数
|
||
if w == g.endWeek && d > g.endDay {
|
||
break
|
||
}
|
||
var dayData [13]slotNode
|
||
if hasData {
|
||
dayData = dayMap[d]
|
||
}
|
||
for s := 1; s <= 12; s++ {
|
||
if w == currW && d == currD && s < currS {
|
||
continue
|
||
}
|
||
if dayData[s].Status == Free || dayData[s].Status == Filler {
|
||
count++
|
||
}
|
||
}
|
||
}
|
||
}
|
||
return count
|
||
}
|
||
|
||
// FindNextAvailable 从当前时间点开始,按周、天、节次顺序查找下一个可用格子
|
||
func (g *grid) FindNextAvailable(currW, currD, currS int) (int, int, int) {
|
||
// 基础越界检查
|
||
if currW > g.endWeek || (currW == g.endWeek && currD > g.endDay) {
|
||
return -1, -1, -1
|
||
}
|
||
|
||
for w := currW; w <= g.endWeek; w++ {
|
||
dayMap, hasData := g.data[w]
|
||
for d := 1; d <= 7; d++ {
|
||
if w == currW && d < currD {
|
||
continue
|
||
}
|
||
if w == g.endWeek && d > g.endDay {
|
||
break
|
||
} // 🚀 守住结束天
|
||
|
||
var dayData [13]slotNode
|
||
if hasData {
|
||
dayData = dayMap[d]
|
||
}
|
||
|
||
for s := 1; s <= 12; s++ {
|
||
if w == currW && d == currD && s < currS {
|
||
continue
|
||
}
|
||
if w == g.endWeek && d == g.endDay {
|
||
break
|
||
} // 🚀 守住结束节
|
||
|
||
if dayData[s].Status == Free || dayData[s].Status == Filler {
|
||
return w, d, s
|
||
}
|
||
}
|
||
}
|
||
}
|
||
return -1, -1, -1
|
||
}
|
||
|
||
// 辅助函数:向后跳过指定数量的可用坑位
|
||
func (g *grid) skipAvailableSlots(w, d, s, skipCount int) (int, int, int) {
|
||
if skipCount <= 0 {
|
||
// 即使 gap 为 0,也要至少移到下一节
|
||
s++
|
||
if s > 12 {
|
||
s = 1
|
||
d++
|
||
if d > 7 {
|
||
d = 1
|
||
w++
|
||
}
|
||
}
|
||
return w, d, s
|
||
}
|
||
found := 0
|
||
currW, currD, currS := w, d, s+1
|
||
for currW <= g.endWeek {
|
||
if currS > 12 {
|
||
currS = 1
|
||
currD++
|
||
if currD > 7 {
|
||
currD = 1
|
||
currW++
|
||
}
|
||
continue
|
||
}
|
||
// 如果已经跳到了最后一天,不要再跳了,直接返回终点坐标
|
||
if currW == g.endWeek && currD > g.endDay {
|
||
return g.endWeek, g.endDay, 12
|
||
}
|
||
if g.isAvailable(currW, currD, currS) {
|
||
found++
|
||
if found > skipCount {
|
||
return currW, currD, currS
|
||
}
|
||
}
|
||
currS++
|
||
}
|
||
return currW, currD, currS
|
||
}
|
||
|
||
func SmartPlanningMainLogic(schedules []model.Schedule, taskClass *model.TaskClass) ([]model.UserWeekSchedule, error) {
|
||
//1.先构建时间格子
|
||
g := buildTimeGrid(schedules, taskClass)
|
||
//2.根据时间格子和排课策略计算每个任务块的具体安排时间
|
||
allocatedItems, err := computeAllocation(g, taskClass.Items, *taskClass.Strategy)
|
||
if err != nil {
|
||
return nil, err
|
||
}
|
||
//3.把这些时间通过DTO函数回填到涉<E588B0><E6B689>周的 UserWeekSchedule 结构中,供前端展示
|
||
return conv.PlanningResultToUserWeekSchedules(schedules, allocatedItems), nil
|
||
}
|
||
|
||
// SmartPlanningRawItems 执行粗排算法并直接返回已分配的任务项列表。
|
||
//
|
||
// 与 SmartPlanningMainLogic 共享完全相同的构建网格和分配逻辑,
|
||
// 但不做展示格式转换,直接返回 allocatedItems(每项的 EmbeddedTime 已回填)。
|
||
// 供 Agent 排程链路使用,避免从展示结构反向解析导致信息丢失。
|
||
func SmartPlanningRawItems(schedules []model.Schedule, taskClass *model.TaskClass) ([]model.TaskClassItem, error) {
|
||
g := buildTimeGrid(schedules, taskClass)
|
||
return computeAllocation(g, taskClass.Items, *taskClass.Strategy)
|
||
}
|
||
|
||
// SmartPlanningRawItemsMulti 执行“多任务类共享资源池”粗排。
|
||
//
|
||
// 职责边界:
|
||
// 1. 复用现有 SmartPlanningRawItems 的单任务类分配能力,不重写核心算法;
|
||
// 2. 通过“增量占位”把前一个任务类的建议结果写入共享工作日程,供后续任务类避让;
|
||
// 3. 返回聚合后的 allocatedItems(每项 EmbeddedTime 已回填);
|
||
// 4. 不负责展示结构转换(由 service/conv 层处理)。
|
||
func SmartPlanningRawItemsMulti(schedules []model.Schedule, taskClasses []*model.TaskClass) ([]model.TaskClassItem, error) {
|
||
if len(taskClasses) == 0 {
|
||
return []model.TaskClassItem{}, nil
|
||
}
|
||
|
||
// 1. 构建“工作副本”:
|
||
// 1.1 原始 schedules 不直接修改,避免污染调用方数据;
|
||
// 1.2 后续每完成一个任务类分配,就把结果增量写入 workingSchedules。
|
||
workingSchedules := cloneSchedulesForPlanning(schedules)
|
||
allAllocated := make([]model.TaskClassItem, 0)
|
||
|
||
// 2. syntheticEventID 用于给“虚拟占位任务”分配唯一 EventID。
|
||
// 2.1 采用负数区间,避免和数据库自增正数 EventID 冲突;
|
||
// 2.2 每个任务块占用一个 synthetic event,跨节次共享同一 eventID。
|
||
nextSyntheticEventID := -1
|
||
|
||
for _, taskClass := range taskClasses {
|
||
if taskClass == nil {
|
||
continue
|
||
}
|
||
if taskClass.Strategy == nil {
|
||
return nil, fmt.Errorf("task_class_id=%d 缺少 strategy 配置", taskClass.ID)
|
||
}
|
||
|
||
// 3. 复用单任务类粗排。
|
||
allocatedItems, err := SmartPlanningRawItems(workingSchedules, taskClass)
|
||
if err != nil {
|
||
// 3.1 明确标注失败任务类,便于上层快速定位。
|
||
return nil, fmt.Errorf("task_class_id=%d 粗排失败: %w", taskClass.ID, err)
|
||
}
|
||
allAllocated = append(allAllocated, allocatedItems...)
|
||
|
||
// 4. 把本任务类分配结果转成“虚拟 Schedule 占位”追加回工作副本。
|
||
// 4.1 目的:让后续任务类把这些已分配任务当成 Occupied,避免重叠;
|
||
// 4.2 若某任务块没有 EmbeddedTime,直接跳过,不阻断后续。
|
||
virtualSchedules, nextID := buildVirtualSchedulesFromAllocated(allocatedItems, taskClass, nextSyntheticEventID)
|
||
nextSyntheticEventID = nextID
|
||
if len(virtualSchedules) > 0 {
|
||
workingSchedules = append(workingSchedules, virtualSchedules...)
|
||
}
|
||
}
|
||
|
||
return allAllocated, nil
|
||
}
|
||
|
||
// cloneSchedulesForPlanning 深拷贝 schedules,确保后续在算法中安全修改。
|
||
//
|
||
// 说明:
|
||
// 1. 主要拷贝 Schedule 结构体本身;
|
||
// 2. Event 指针做浅字段复制,避免共享同一 Event 指针导致意外改写;
|
||
// 3. EmbeddedTask 在粗排阶段不参与状态写入,保留原值即可。
|
||
func cloneSchedulesForPlanning(src []model.Schedule) []model.Schedule {
|
||
if len(src) == 0 {
|
||
return []model.Schedule{}
|
||
}
|
||
dst := make([]model.Schedule, len(src))
|
||
for i := range src {
|
||
dst[i] = src[i]
|
||
if src[i].Event != nil {
|
||
eventCopy := *src[i].Event
|
||
dst[i].Event = &eventCopy
|
||
}
|
||
}
|
||
return dst
|
||
}
|
||
|
||
// buildVirtualSchedulesFromAllocated 将已分配任务块转成“虚拟占位 schedules”。
|
||
//
|
||
// 设计目的:
|
||
// 1. 让后续任务类在共享资源池里自动避让已分配任务;
|
||
// 2. 不落库,仅用于内存中的粗排冲突控制;
|
||
// 3. 通过 Type=task + CanBeEmbedded=false 强制标记为不可再嵌入。
|
||
func buildVirtualSchedulesFromAllocated(allocatedItems []model.TaskClassItem, taskClass *model.TaskClass, eventIDStart int) ([]model.Schedule, int) {
|
||
if len(allocatedItems) == 0 {
|
||
return []model.Schedule{}, eventIDStart
|
||
}
|
||
|
||
userID := 0
|
||
if taskClass != nil && taskClass.UserID != nil {
|
||
userID = *taskClass.UserID
|
||
}
|
||
|
||
virtual := make([]model.Schedule, 0)
|
||
nextEventID := eventIDStart
|
||
for _, item := range allocatedItems {
|
||
if item.EmbeddedTime == nil {
|
||
continue
|
||
}
|
||
|
||
taskName := "未命名任务"
|
||
if item.Content != nil && *item.Content != "" {
|
||
taskName = *item.Content
|
||
}
|
||
location := ""
|
||
event := &model.ScheduleEvent{
|
||
ID: nextEventID,
|
||
UserID: userID,
|
||
Name: taskName,
|
||
Location: &location,
|
||
Type: "task",
|
||
CanBeEmbedded: false,
|
||
}
|
||
for section := item.EmbeddedTime.SectionFrom; section <= item.EmbeddedTime.SectionTo; section++ {
|
||
virtual = append(virtual, model.Schedule{
|
||
EventID: nextEventID,
|
||
UserID: userID,
|
||
Week: item.EmbeddedTime.Week,
|
||
DayOfWeek: item.EmbeddedTime.DayOfWeek,
|
||
Section: section,
|
||
Event: event,
|
||
Status: "normal",
|
||
})
|
||
}
|
||
nextEventID--
|
||
}
|
||
return virtual, nextEventID
|
||
}
|
||
|
||
// buildTimeGrid 构建一个时间格子,标记出哪些时间段被占用、哪些被屏蔽、哪些是水课
|
||
|
||
func buildTimeGrid(schedules []model.Schedule, taskClass *model.TaskClass) *grid {
|
||
// 🚀 核心修正:获取精确的起始坐标
|
||
startW, startD, _ := conv.RealDateToRelativeDate(taskClass.StartDate.Format(conv.DateFormat))
|
||
endW, endD, _ := conv.RealDateToRelativeDate(taskClass.EndDate.Format(conv.DateFormat))
|
||
// 将信息初始化到 grid 结构中
|
||
g := &grid{
|
||
data: make(map[int]map[int][13]slotNode),
|
||
startWeek: startW,
|
||
startDay: startD,
|
||
endWeek: endW,
|
||
endDay: endD,
|
||
}
|
||
//标记屏蔽时段 (Blocked)
|
||
for _, blockIdx := range taskClass.ExcludedSlots {
|
||
sFrom, sTo := (blockIdx-1)*2+1, blockIdx*2
|
||
for w := startW; w <= endW; w++ {
|
||
for d := 1; d <= 7; d++ { //🚀 注意:这里的屏蔽是针对每天的,所以直接循环 1-7 天
|
||
for s := sFrom; s <= sTo; s++ {
|
||
g.setNode(w, d, s, slotNode{Status: Blocked})
|
||
}
|
||
}
|
||
}
|
||
}
|
||
|
||
// 映射日程 (尊重 Blocked 且只处理范围内的数据)
|
||
for _, s := range schedules {
|
||
if s.Week >= startW && s.Week <= endW {
|
||
if g.getNode(s.Week, s.DayOfWeek, s.Section).Status == Blocked {
|
||
continue
|
||
}
|
||
status := Occupied
|
||
// 只有当课程允许嵌入且当前事件支持嵌入时,才标记为 Filler
|
||
if *taskClass.AllowFillerCourse && s.Event.CanBeEmbedded {
|
||
status = Filler
|
||
}
|
||
g.setNode(s.Week, s.DayOfWeek, s.Section, slotNode{Status: status, EventID: uint(s.EventID)})
|
||
}
|
||
}
|
||
return g
|
||
}
|
||
|
||
// computeAllocation 是核心函数,负责根据当前的时间格子状态和排课策略,计算出每个任务块的具体安排时间
|
||
/*func computeAllocation(g *grid, items []model.TaskClassItem, strategy string) ([]model.TaskClassItem, error) {
|
||
if len(items) == 0 {
|
||
return items, nil
|
||
}
|
||
|
||
// 🚀 核心修正 1:获取真正的开始坐标(周、天、节)
|
||
// 这里假设你已经通过 conv 把 StartDate 换成了 w1, d1, s1
|
||
startW := g.startWeek
|
||
startD := g.startDay // 建议从 conv 传入具体的 DayOfWeek
|
||
startS := 1
|
||
|
||
// 1. 获取可用资源总量
|
||
totalAvailable := g.countAvailableSlots(0, 0, 0)
|
||
// 假设每个任务块至少占用 2 个原子槽位
|
||
totalRequired := len(items) * 2
|
||
|
||
// 🚀 核心改进:容量预判
|
||
if totalAvailable < totalRequired {
|
||
// 如果连最基本的坑位都不够,直接报错,不进行任何编排
|
||
return nil, respond.TimeNotEnoughForAutoScheduling
|
||
}
|
||
|
||
// 🚀 核心修正 2:步长改为“逻辑间隔”,不再是物理跳跃
|
||
// gap 表示:每两个任务之间,我们要故意空出多少个“可用位”
|
||
gap := 0
|
||
if strategy == "steady" && totalAvailable > totalRequired {
|
||
gap = (totalAvailable - totalRequired) / (len(items) + 1)
|
||
}
|
||
|
||
currW, currD, currS := startW, startD, startS
|
||
lastPlacedIndex := -1
|
||
for i := range items {
|
||
w, d, s := g.FindNextAvailable(currW, currD, currS)
|
||
if w == -1 || w > g.endWeek {
|
||
break
|
||
}
|
||
|
||
node := g.getNode(w, d, s)
|
||
slotLen := 2
|
||
if node.Status == Filler {
|
||
slotLen = 1
|
||
currID := node.EventID
|
||
for checkS := s + 1; checkS <= 12; checkS++ {
|
||
if next := g.getNode(w, d, checkS); next.Status == Filler && next.EventID == currID {
|
||
slotLen++
|
||
} else {
|
||
break
|
||
}
|
||
}
|
||
}
|
||
|
||
endS := s + slotLen - 1
|
||
items[i].EmbeddedTime = &model.TargetTime{
|
||
SectionFrom: s, SectionTo: endS,
|
||
Week: w, DayOfWeek: d,
|
||
}
|
||
|
||
for sec := s; sec <= endS; sec++ {
|
||
g.setNode(w, d, sec, slotNode{Status: Occupied})
|
||
}
|
||
|
||
// 🚀 核心修正 3:基于“可用位”推进指针,而非物理索引
|
||
// 我们要在 grid 中向后数出 gap 个可用位置,作为下一个任务的起点
|
||
currW, currD, currS = g.skipAvailableSlots(w, d, endS, gap)
|
||
|
||
lastPlacedIndex = i // 记录最后一个成功安放的任务索引
|
||
}
|
||
// 🚀 核心改进:结果完整性校验
|
||
if lastPlacedIndex < len(items)-1 {
|
||
return nil, fmt.Errorf("排程中断:由于时间片碎片化,仅成功安排了 %d/%d 个任务块,请尝试扩充时间范围或删减屏蔽位", lastPlacedIndex+1, len(items))
|
||
return nil, respond.TimeNotEnoughForAutoScheduling
|
||
}
|
||
|
||
return items, nil
|
||
}*/
|
||
|
||
type slotCoord struct {
|
||
w, d, s int
|
||
}
|
||
|
||
// getAllAvailable 获取窗口内所有可用的原子节次坐标(逻辑一维化)
|
||
func (g *grid) getAllAvailable() []slotCoord {
|
||
var coords []slotCoord
|
||
for w := g.startWeek; w <= g.endWeek; w++ {
|
||
dayMap, hasData := g.data[w]
|
||
for d := 1; d <= 7; d++ {
|
||
// 边界裁剪逻辑
|
||
if w == g.startWeek && d < g.startDay {
|
||
continue
|
||
}
|
||
if w == g.endWeek && d > g.endDay {
|
||
break
|
||
}
|
||
|
||
var dayData [13]slotNode
|
||
if hasData {
|
||
dayData = dayMap[d]
|
||
}
|
||
|
||
for s := 1; s <= 12; s++ {
|
||
// 顺着你的逻辑,不限开始节次,但需注意状态判定
|
||
if dayData[s].Status == Free || dayData[s].Status == Filler {
|
||
coords = append(coords, slotCoord{w, d, s})
|
||
}
|
||
}
|
||
}
|
||
}
|
||
return coords
|
||
}
|
||
|
||
func computeAllocation(g *grid, items []model.TaskClassItem, strategy string) ([]model.TaskClassItem, error) {
|
||
if len(items) == 0 {
|
||
return items, nil
|
||
}
|
||
|
||
// 1. 预处理:提取所有可用坑位
|
||
coords := g.getAllAvailable()
|
||
totalAvailable := len(coords)
|
||
totalRequired := len(items) * 2 // 基础需求:每个任务 2 节
|
||
|
||
if totalAvailable < totalRequired {
|
||
return nil, respond.TimeNotEnoughForAutoScheduling
|
||
}
|
||
|
||
// 2. 计算精准步长
|
||
gap := 0
|
||
if strategy == "steady" {
|
||
gap = (totalAvailable - totalRequired) / (len(items) + 1)
|
||
}
|
||
|
||
// 3. 线性映射分配
|
||
// cursor 是我们在逻辑切片中的“指针”
|
||
cursor := gap
|
||
lastPlacedIndex := -1
|
||
|
||
for i := range items {
|
||
if cursor >= totalAvailable {
|
||
break
|
||
}
|
||
|
||
// 获取当前逻辑位置对应的物理坐标
|
||
startLoc := coords[cursor]
|
||
w, d, s := startLoc.w, startLoc.d, startLoc.s
|
||
|
||
// 4. 计算本次任务块的落点区间。
|
||
// 4.1 默认按 2 节处理(普通空闲位优先遵循“每任务2节”的主策略);
|
||
// 4.2 命中 Filler(可嵌入课程)时,必须先回溯到同课程块起点,再计算完整连续跨度;
|
||
// 4.3 失败兜底:若普通空闲位后继不可用,只能退化为 1 节,避免越界或覆盖占用位。
|
||
node := g.getNode(w, d, s)
|
||
sectionFrom := s
|
||
slotLen := 2
|
||
if node.Status == Filler {
|
||
// 4.2.1 先向左回溯到“同一课程块”的起点。
|
||
// 目的:修复“指针落在课程中间节次时被错误切成 1 节”的问题。
|
||
// 例如课程占 9-10 节,若 cursor 命中 10 节,必须回溯到 9 节再整体计算。
|
||
currID := node.EventID
|
||
for checkS := s - 1; checkS >= 1; checkS-- {
|
||
prev := g.getNode(w, d, checkS)
|
||
if prev.Status == Filler && prev.EventID == currID {
|
||
sectionFrom = checkS
|
||
continue
|
||
}
|
||
break
|
||
}
|
||
|
||
// 4.2.2 再从起点向右扩展,拿到同一课程块的完整连续节次长度。
|
||
sectionTo := sectionFrom
|
||
for checkS := sectionFrom + 1; checkS <= 12; checkS++ {
|
||
if next := g.getNode(w, d, checkS); next.Status == Filler && next.EventID == currID {
|
||
sectionTo = checkS
|
||
} else {
|
||
break
|
||
}
|
||
}
|
||
slotLen = sectionTo - sectionFrom + 1
|
||
} else if s == 12 || !g.isAvailable(w, d, s+1) {
|
||
// 如果是 Free 区域,但下一节不可用,则被迫设为 1 节
|
||
slotLen = 1
|
||
}
|
||
|
||
// 回填时间
|
||
endS := sectionFrom + slotLen - 1
|
||
items[i].EmbeddedTime = &model.TargetTime{
|
||
SectionFrom: sectionFrom, SectionTo: endS,
|
||
Week: w, DayOfWeek: d,
|
||
}
|
||
|
||
// 标记占用 (物理网格)
|
||
for sec := sectionFrom; sec <= endS; sec++ {
|
||
g.setNode(w, d, sec, slotNode{Status: Occupied})
|
||
}
|
||
|
||
// 🚀 核心进步:逻辑跳跃
|
||
// 既然任务占用了 slotLen 节,我们在逻辑切片中也向后推 slotLen 个位置,再加 gap
|
||
cursor += slotLen + gap
|
||
lastPlacedIndex = i
|
||
}
|
||
|
||
if lastPlacedIndex < len(items)-1 {
|
||
return nil, respond.TimeNotEnoughForAutoScheduling
|
||
}
|
||
|
||
return items, nil
|
||
}
|