Agent-Intent/intent/slots.go

package intent

import (
	"regexp"
	"strings"
	"sync"
)

// SlotValidator 宿主注册的槽位值校验器。必须纯函数、确定性、无 I/O。
type SlotValidator func(value string) error

var slotValidators sync.Map // name → SlotValidator

// RegisterSlotValidator 注册进程级具名校验器；spec 通过 SlotSpec.Validator 引用。
// 必须在快照构建前注册（构建期会校验名字存在）。
//
// Deprecated: 校验器应随 Registry 隔离。请改用 Options.SlotValidators。
// 全局表只作回落：构建与校验先查作用域表，查不到再查这里。
func RegisterSlotValidator(name string, fn SlotValidator) {
	slotValidators.Store(name, fn)
}

// LookupSlotValidator 按名查找已注册的全局校验器。
//
// Deprecated: 请改用 Options.SlotValidators。本函数只服务全局回落。
func LookupSlotValidator(name string) (SlotValidator, bool) {
	fn, ok := slotValidators.Load(name)
	if !ok {
		return nil, false
	}
	v, ok := fn.(SlotValidator)
	return v, ok
}

// resolveSlotValidator 先查作用域表，查不到再回落已弃用的全局表。
func resolveSlotValidator(scoped map[string]SlotValidator, name string) (SlotValidator, bool) {
	if fn, ok := scoped[name]; ok && fn != nil {
		return fn, true
	}
	return LookupSlotValidator(name)
}

// 内置归一化步骤：trim / lower / upper。
var builtinNormalizers = map[string]func(string) string{
	"trim":  strings.TrimSpace,
	"lower": strings.ToLower,
	"upper": strings.ToUpper,
}

func normalizeSlotValue(v string, ops []string) string {
	for _, op := range ops {
		if fn, ok := builtinNormalizers[op]; ok {
			v = fn(v)
		}
	}
	return v
}

// compiledSlot 编译后的槽位抽取与校验产物。
type compiledSlot struct {
	key       string
	required  bool
	label     string
	question  string
	re        *regexp.Regexp
	group     int
	enumSet   map[string]bool
	enum      []string
	formatRe  *regexp.Regexp
	normalize []string
	validator SlotValidator
}

// validate 对归一化后的值做确定性校验；返回（原因, 是否合法）。
func (s *compiledSlot) validate(v string) (string, bool) {
	if s.formatRe != nil && !s.formatRe.MatchString(v) {
		return "format mismatch: " + s.formatRe.String(), false
	}
	if s.enumSet != nil && !s.enumSet[v] {
		return "value not in enum", false
	}
	if s.validator != nil {
		if err := s.validator(v); err != nil {
			return "validator: " + err.Error(), false
		}
	}
	return "", true
}

// slotCapture 一次槽位抽取的捕获区间。多槽位各自第一次匹配；区间重叠时
// 两边都标 span overlap，不写入 slots。库内不做重叠消解。
type slotCapture struct {
	key      string
	start    int
	end      int
	value    string
	ok       bool
	reason   string
	required bool
	matched  bool
}

// extractSlots 抽取并归一化槽位。
// missing：必填槽位未捕获（模式没匹配上）；invalid：捕获到但校验不合法，或与另一槽位区间重叠（key→原因）。
// 必填且非法只进 invalid（宿主可区分"没给"与"给错了"来决定澄清话术）。
func (cs *compiledSpec) extractSlots(input string) (slots map[string]string, missing []string, invalid map[string]string) {
	if len(cs.slots) == 0 {
		return nil, nil, nil
	}
	caps := make([]slotCapture, 0, len(cs.slots))
	for _, s := range cs.slots {
		loc := s.re.FindStringSubmatchIndex(input)
		g := s.group
		if loc == nil || 2*g+1 >= len(loc) || loc[2*g] < 0 || loc[2*g] == loc[2*g+1] {
			caps = append(caps, slotCapture{key: s.key, required: s.required})
			continue
		}
		raw := input[loc[2*g]:loc[2*g+1]]
		v := normalizeSlotValue(raw, s.normalize)
		reason, ok := s.validate(v)
		caps = append(caps, slotCapture{
			key: s.key, start: loc[2*g], end: loc[2*g+1], value: v,
			ok: ok, reason: reason, required: s.required, matched: true,
		})
	}
	overlap := map[int]bool{}
	for i := range caps {
		if !caps[i].matched {
			continue
		}
		for j := i + 1; j < len(caps); j++ {
			if !caps[j].matched {
				continue
			}
			if caps[i].start < caps[j].end && caps[j].start < caps[i].end {
				overlap[i] = true
				overlap[j] = true
			}
		}
	}
	for i, c := range caps {
		if !c.matched {
			if c.required {
				missing = append(missing, c.key)
			}
			continue
		}
		if overlap[i] {
			if invalid == nil {
				invalid = map[string]string{}
			}
			invalid[c.key] = "span overlap"
			continue
		}
		if !c.ok {
			if invalid == nil {
				invalid = map[string]string{}
			}
			invalid[c.key] = c.reason
			continue
		}
		if slots == nil {
			slots = map[string]string{}
		}
		slots[c.key] = c.value
	}
	return slots, missing, invalid
}

func compileSlot(spec SlotSpec, scoped map[string]SlotValidator) (compiledSlot, error) {
	re, err := regexp.Compile(spec.Pattern)
	if err != nil {
		return compiledSlot{}, newError(InvalidSpec, "slot pattern 编译失败",
			"field", "slots", "key", spec.Key, "error", err.Error())
	}
	cs := compiledSlot{
		key:       spec.Key,
		required:  spec.Required,
		label:     spec.Label,
		question:  spec.Question,
		re:        re,
		group:     effectiveGroup(spec.Group),
		normalize: append([]string(nil), spec.Normalize...),
	}
	if len(spec.Enum) > 0 {
		cs.enum = append([]string(nil), spec.Enum...)
		cs.enumSet = make(map[string]bool, len(spec.Enum))
		for _, e := range spec.Enum {
			cs.enumSet[e] = true
		}
	}
	if spec.Format != "" {
		fre, err := regexp.Compile(spec.Format)
		if err != nil {
			return compiledSlot{}, newError(InvalidSpec, "slot format 编译失败",
				"field", "slots", "key", spec.Key, "format", spec.Format, "error", err.Error())
		}
		cs.formatRe = fre
	}
	if spec.Validator != "" {
		fn, ok := resolveSlotValidator(scoped, spec.Validator)
		if !ok {
			return compiledSlot{}, newError(InvalidSpec, "slot validator 未注册",
				"field", "slots", "key", spec.Key, "validator", spec.Validator,
				"hint", "在 Options.SlotValidators 登记，或调用已弃用的 intent.RegisterSlotValidator")
		}
		cs.validator = fn
	}
	return cs, nil
}