fix: 优化核心包

gig-poc/apps/api/app/domain/schemas.py

@@ -1,6 +1,6 @@
 from __future__ import annotations
 
-from datetime import datetime
+from datetime import datetime, timedelta, timezone
 from enum import Enum
 
 from pydantic import BaseModel, ConfigDict, Field, field_validator, model_validator
@@ -46,6 +46,16 @@ class JobCard(BaseModel):
     tags: list[str] = Field(default_factory=list, description="业务标签列表")
     confidence: float = Field(ge=0, le=1, description="数据置信度，范围 0~1")
 
+    @field_validator("start_time", mode="after")
+    @classmethod
+    def normalize_start_time(cls, value: datetime) -> datetime:
+        shanghai_tz = timezone(timedelta(hours=8))
+        if value.tzinfo is None:
+            value = value.replace(tzinfo=shanghai_tz)
+        else:
+            value = value.astimezone(shanghai_tz)
+        return value.replace(second=0, microsecond=0)
+
 
 class WorkerCard(BaseModel):
     worker_id: str = Field(description="工人唯一 ID")

gig-poc/apps/api/app/services/extraction_service.py

@@ -2,6 +2,7 @@ from __future__ import annotations
 
 import json
 import re
+from collections import Counter
 from datetime import datetime, timedelta, timezone
 from pathlib import Path
 
@@ -22,7 +23,19 @@ class ExtractionService:
         self.skills = json.loads((self.settings.sample_data_dir / "skills.json").read_text(encoding="utf-8"))
         self.categories = json.loads((self.settings.sample_data_dir / "categories.json").read_text(encoding="utf-8"))
         self.regions = json.loads((self.settings.sample_data_dir / "regions.json").read_text(encoding="utf-8"))
+        self.sample_jobs = json.loads((self.settings.sample_data_dir / "jobs.json").read_text(encoding="utf-8"))
+        self.sample_workers = json.loads((self.settings.sample_data_dir / "workers.json").read_text(encoding="utf-8"))
+        self.default_region = self._build_default_region()
+        self.default_category = self._build_default_category()
+        self.default_salary_amount = self._build_default_salary_amount()
+        self.default_job_tags = self._build_default_job_tags()
+        self.default_worker_skills = self._build_default_worker_skills()
+        self.default_experience_tags = self._build_default_experience_tags()
+        self.category_skill_defaults = self._build_category_skill_defaults()
+        self.city_region_defaults = self._build_city_region_defaults()
+        self.tag_candidates = self._build_tag_candidates()
         self.llm_client = LLMClient(self.settings)
+        self.shanghai_tz = timezone(timedelta(hours=8))
 
     def extract_job(self, text: str) -> ExtractResponse:
         logger.info("extract_job request text=%s", text)
@@ -110,12 +123,12 @@ class ExtractionService:
 
     def _extract_job_rule(self, text: str) -> JobCard:
         skill_hits = [item for item in self.skills if item in text]
-        category = next((item for item in self.categories if item in text), "活动执行")
+        category = next((item for item in self.categories if item in text), self.default_category)
         region = self._extract_region(text)
         salary = self._extract_salary(text)
         headcount = self._extract_number(text, [r"(\d+)\s*[个名人位]"], default=1)
         duration = self._extract_number(text, [r"(\d+(?:\.\d+)?)\s*小时"], default=4.0, cast=float)
-        tags = [tag for tag in ["女生优先", "男生优先", "有经验优先", "沟通好", "可连做优先"] if tag in text]
+        tags = [tag for tag in self.tag_candidates if tag in text][:3]
         title = next((f"{category}{skill_hits[0]}兼职" for _ in [0] if skill_hits), f"{category}兼职")
         card = JobCard(
             job_id=generate_id("job"),
@@ -131,7 +144,7 @@ class ExtractionService:
             headcount=int(headcount),
             salary=salary,
             work_mode="排班制" if "排班" in text else "兼职",
-            tags=tags or ["有经验优先"],
+            tags=tags or self.default_job_tags,
             confidence=self._compute_confidence(skill_hits, region, salary.amount > 0),
         )
         return card
@@ -139,19 +152,29 @@ class ExtractionService:
     def _extract_worker_rule(self, text: str) -> WorkerCard:
         skill_hits = [item for item in self.skills if item in text][:6]
         region_hits = [item for item in self.regions if item["region"] in text or item["city"] in text]
-        city_names = list(dict.fromkeys([item["city"] for item in region_hits])) or ["深圳"]
-        region_names = list(dict.fromkeys([item["region"] for item in region_hits])) or ["南山"]
+        if not region_hits:
+            city_hits = [item["city"] for item in self.regions if item["city"] in text]
+            unique_city_hits = list(dict.fromkeys(city_hits))
+            region_hits = [
+                {"city": city, "region": self.city_region_defaults.get(city, self.default_region["region"])}
+                for city in unique_city_hits
+            ]
+        city_names = list(dict.fromkeys([item["city"] for item in region_hits])) or [self.default_region["city"]]
+        region_names = list(dict.fromkeys([item["region"] for item in region_hits])) or [self.default_region["region"]]
         availability = self._extract_availability(text)
-        experience = [item for item in ["商场", "会展", "活动执行", "物流", "零售", "客服中心", "快消", "校园推广"] if item in text]
+        experience = [item for item in self.default_experience_tags if item in text]
         card = WorkerCard(
             worker_id=generate_id("worker"),
             name=self._extract_name(text),
             description=text,
-            skills=[SkillScore(name=item, score=round(0.72 + index * 0.04, 2)) for index, item in enumerate(skill_hits or ["活动执行", "引导", "登记"])],
+            skills=[
+                SkillScore(name=item, score=round(0.72 + index * 0.04, 2))
+                for index, item in enumerate(skill_hits or self.default_worker_skills)
+            ],
             cities=city_names,
             regions=region_names,
             availability=availability,
-            experience_tags=experience or ["活动执行"],
+            experience_tags=experience or self.default_experience_tags[:2],
             reliability_score=0.76,
             profile_completion=0.68,
             confidence=self._compute_confidence(skill_hits, {"city": city_names[0], "region": region_names[0]}, True),
@@ -165,7 +188,10 @@ class ExtractionService:
         for item in self.regions:
             if item["region"] in text:
                 return item
-        return {"city": "深圳", "region": "南山"}
+        city_match = next((item["city"] for item in self.regions if item["city"] in text), "")
+        if city_match:
+            return {"city": city_match, "region": self.city_region_defaults.get(city_match, self.default_region["region"])}
+        return self.default_region
 
     def _extract_location(self, text: str, region: dict) -> str:
         markers = ["会展中心", "商场", "地铁站", "园区", "写字楼", "仓库", "门店"]
@@ -175,7 +201,7 @@ class ExtractionService:
         return f"{region['city']}{region['region']}待定点位"
 
     def _extract_salary(self, text: str) -> Salary:
-        amount = self._extract_number(text, [r"(\d+(?:\.\d+)?)\s*(?:元|块)"], default=150.0, cast=float)
+        amount = self._extract_number(text, [r"(\d+(?:\.\d+)?)\s*(?:元|块)"], default=self.default_salary_amount, cast=float)
         salary_type = "hourly" if "小时" in text and "/小时" in text else "daily"
         return Salary(type=salary_type, amount=amount, currency="CNY")
 
@@ -187,28 +213,72 @@ class ExtractionService:
         return default
 
     def _extract_job_time(self, text: str) -> datetime:
-        shanghai_tz = timezone(timedelta(hours=8))
-        now = datetime.now(shanghai_tz)
+        now = datetime.now(self.shanghai_tz)
+        for candidate in self._time_candidates(text, now):
+            parsed = self._parse_datetime(candidate, now)
+            if parsed:
+                return parsed
+        return self._normalize_datetime(now + timedelta(days=1))
+
+    def _time_candidates(self, text: str, now: datetime) -> list[str]:
+        candidates = [text]
+
+        if any(token in text for token in ("今天", "今日")):
+            candidates.append(text.replace("今日", now.strftime("%Y-%m-%d")).replace("今天", now.strftime("%Y-%m-%d")))
         if "明天" in text:
-            base = now + timedelta(days=1)
-        elif "后天" in text:
-            base = now + timedelta(days=2)
+            tomorrow = now + timedelta(days=1)
+            candidates.append(text.replace("明天", tomorrow.strftime("%Y-%m-%d")))
+        if "后天" in text:
+            day_after = now + timedelta(days=2)
+            candidates.append(text.replace("后天", day_after.strftime("%Y-%m-%d")))
+
+        weekday_map = {"一": 0, "二": 1, "三": 2, "四": 3, "五": 4, "六": 5, "日": 6, "天": 6}
+        week_match = re.search(r"(下周|本周|这周|周)([一二三四五六日天])", text)
+        if week_match:
+            week_token, weekday_token = week_match.groups()
+            target_weekday = weekday_map[weekday_token]
+            days_ahead = (target_weekday - now.weekday()) % 7
+            if week_token == "下周":
+                days_ahead = days_ahead + 7
+            elif week_token == "周" and days_ahead == 0:
+                days_ahead = 7
+            target_day = now + timedelta(days=days_ahead)
+            candidates.append(text.replace(week_match.group(0), target_day.strftime("%Y-%m-%d")))
+        return candidates
+
+    def _parse_datetime(self, text: str, now: datetime) -> datetime | None:
+        normalized = self._replace_time_words(text)
+        cleaned = re.sub(r"[，、。；,;]", " ", normalized)
+        cleaned = cleaned.replace("号", "日")
+        cleaned = re.sub(r"(\d{1,2})月(\d{1,2})日", rf"{now.year}-\1-\2", cleaned)
+        cleaned = re.sub(r"(\d{1,2})点半", r"\1:30", cleaned)
+        cleaned = re.sub(r"(\d{1,2})点", r"\1:00", cleaned)
+        cleaned = re.sub(r"(\d{1,2})时", r"\1:00", cleaned)
+
+        has_date = bool(re.search(r"\d{4}-\d{1,2}-\d{1,2}", cleaned))
+        if not has_date:
+            return None
+        try:
+            parsed = date_parser.parse(cleaned, fuzzy=True)
+        except Exception:
+            return None
+        return self._normalize_datetime(parsed)
+
+    def _replace_time_words(self, text: str) -> str:
+        replaced = text
+        replaced = re.sub(r"(今晚|晚上)", " 19:00 ", replaced)
+        replaced = re.sub(r"(下午)", " 14:00 ", replaced)
+        replaced = re.sub(r"(中午)", " 12:00 ", replaced)
+        replaced = re.sub(r"(早上|上午)", " 09:00 ", replaced)
+        replaced = re.sub(r"(凌晨)", " 01:00 ", replaced)
+        return replaced
+
+    def _normalize_datetime(self, value: datetime) -> datetime:
+        if value.tzinfo is None:
+            value = value.replace(tzinfo=self.shanghai_tz)
         else:
-            month_day = re.search(r"(\d{1,2})月(\d{1,2})日", text)
-            if month_day:
-                month, day = int(month_day.group(1)), int(month_day.group(2))
-                base = now.replace(month=month, day=day)
-            else:
-                base = now + timedelta(days=1)
-        hour = 9
-        if "下午" in text:
-            hour = 13
-        elif "晚上" in text:
-            hour = 19
-        explicit_hour = re.search(r"(\d{1,2})[:点时](\d{0,2})?", text)
-        if explicit_hour:
-            hour = int(explicit_hour.group(1))
-        return base.replace(hour=hour, minute=0, second=0, microsecond=0)
+            value = value.astimezone(self.shanghai_tz)
+        return value.replace(second=0, microsecond=0)
 
     def _extract_availability(self, text: str) -> list[str]:
         tags = []
@@ -230,13 +300,10 @@ class ExtractionService:
         return "匿名候选人"
 
     def _guess_category_skills(self, category: str) -> list[str]:
-        mapping = {
-            "活动执行": ["签到", "引导", "登记"],
-            "促销": ["促销", "导购", "陈列"],
-            "配送": ["配送", "装卸", "司机协助"],
-            "客服": ["客服", "电话邀约", "线上客服"],
-        }
-        return mapping.get(category, ["活动执行", "沟通"])
+        skills = self.category_skill_defaults.get(category)
+        if skills:
+            return skills
+        return self.default_worker_skills[:3]
 
     def _compute_confidence(self, skill_hits: list[str], region: dict, has_salary: bool) -> float:
         score = 0.55
@@ -250,3 +317,109 @@ class ExtractionService:
 
     def _missing_fields(self, exc: ValidationError) -> list[str]:
         return [".".join(str(part) for part in item["loc"]) for item in exc.errors()]
+
+    def _build_default_region(self) -> dict:
+        if self.sample_jobs:
+            pair_counter = Counter(
+                (item.get("city"), item.get("region"))
+                for item in self.sample_jobs
+                if item.get("city") and item.get("region")
+            )
+            if pair_counter:
+                city, region = pair_counter.most_common(1)[0][0]
+                return {"city": city, "region": region}
+        if self.regions:
+            return {"city": self.regions[0]["city"], "region": self.regions[0]["region"]}
+        return {"city": "深圳", "region": "南山"}
+
+    def _build_default_category(self) -> str:
+        counter = Counter(item.get("category") for item in self.sample_jobs if item.get("category"))
+        if counter:
+            return counter.most_common(1)[0][0]
+        return self.categories[0] if self.categories else "活动执行"
+
+    def _build_default_salary_amount(self) -> float:
+        amounts = sorted(
+            float(item["salary"]["amount"])
+            for item in self.sample_jobs
+            if isinstance(item.get("salary"), dict) and isinstance(item["salary"].get("amount"), (int, float))
+        )
+        if not amounts:
+            return 150.0
+        mid = len(amounts) // 2
+        if len(amounts) % 2 == 1:
+            return amounts[mid]
+        return round((amounts[mid - 1] + amounts[mid]) / 2, 2)
+
+    def _build_default_job_tags(self) -> list[str]:
+        counter = Counter(
+            tag
+            for item in self.sample_jobs
+            for tag in item.get("tags", [])
+            if isinstance(tag, str) and tag.strip()
+        )
+        top_tags = [tag for tag, _ in counter.most_common(3)]
+        return top_tags or ["有经验优先"]
+
+    def _build_default_worker_skills(self) -> list[str]:
+        counter = Counter(
+            skill.get("name")
+            for item in self.sample_workers
+            for skill in item.get("skills", [])
+            if isinstance(skill, dict) and isinstance(skill.get("name"), str) and skill.get("name")
+        )
+        top_skills = [name for name, _ in counter.most_common(4)]
+        return top_skills or ["活动执行", "引导", "登记"]
+
+    def _build_default_experience_tags(self) -> list[str]:
+        counter = Counter(
+            tag
+            for item in self.sample_workers
+            for tag in item.get("experience_tags", [])
+            if isinstance(tag, str) and tag.strip()
+        )
+        top_tags = [tag for tag, _ in counter.most_common(5)]
+        return top_tags or ["活动执行"]
+
+    def _build_category_skill_defaults(self) -> dict[str, list[str]]:
+        category_skills: dict[str, Counter] = {}
+        for item in self.sample_jobs:
+            category = item.get("category")
+            if not isinstance(category, str) or not category:
+                continue
+            counter = category_skills.setdefault(category, Counter())
+            for skill in item.get("skills", []):
+                if isinstance(skill, str) and skill:
+                    counter[skill] += 1
+        return {category: [name for name, _ in counter.most_common(4)] for category, counter in category_skills.items()}
+
+    def _build_city_region_defaults(self) -> dict[str, str]:
+        counter: dict[str, Counter] = {}
+        for item in self.regions:
+            city = item.get("city")
+            region = item.get("region")
+            if not city or not region:
+                continue
+            counter.setdefault(city, Counter())[region] += 1
+        for item in self.sample_jobs:
+            city = item.get("city")
+            region = item.get("region")
+            if city and region:
+                counter.setdefault(city, Counter())[region] += 3
+        defaults: dict[str, str] = {}
+        for city, regions in counter.items():
+            defaults[city] = regions.most_common(1)[0][0]
+        return defaults
+
+    def _build_tag_candidates(self) -> list[str]:
+        sample_tags = list(
+            dict.fromkeys(
+                tag
+                for item in self.sample_jobs
+                for tag in item.get("tags", [])
+                if isinstance(tag, str) and tag.strip()
+            )
+        )
+        baseline_tags = ["女生优先", "男生优先", "有经验优先", "沟通好", "可连做优先"]
+        merged = list(dict.fromkeys([*sample_tags, *baseline_tags]))
+        return merged[:30]

gig-poc/apps/web/src/pages/JobPage.tsx

@@ -1,16 +1,76 @@
-import { useState } from "react";
+import { useEffect, useState } from "react";
 import { api } from "../api/client";
 import { JsonPanel } from "../components/JsonPanel";
 import { MatchList } from "../components/MatchList";
 
-const DEFAULT_TEXT = "明天下午南山会展中心需要2个签到协助，5小时，150/人，女生优先，需要会签到、引导和登记。";
+const FALLBACK_TEXT = "明天下午南山会展中心需要2个签到协助，5小时，150/人，女生优先，需要会签到、引导和登记。";
+
+function pickRandom<T>(items: T[]): T {
+  return items[Math.floor(Math.random() * items.length)];
+}
+
+function asString(value: unknown): string {
+  return typeof value === "string" ? value.trim() : "";
+}
+
+function asNumber(value: unknown): number | null {
+  return typeof value === "number" && Number.isFinite(value) ? value : null;
+}
+
+function asStringArray(value: unknown): string[] {
+  if (!Array.isArray(value)) {
+    return [];
+  }
+  return value.filter((item): item is string => typeof item === "string" && item.trim().length > 0);
+}
+
+function buildAdaptiveJobText(items: Record<string, unknown>[]): string {
+  if (!items.length) {
+    return FALLBACK_TEXT;
+  }
+  const source = pickRandom(items);
+  const title = asString(source.title) || asString(source.category) || "活动兼职";
+  const city = asString(source.city) || "深圳";
+  const region = asString(source.region) || "南山";
+  const headcount = asNumber(source.headcount) ?? 2;
+  const duration = asNumber(source.duration_hours) ?? 4;
+  const location = asString(source.location_detail) || `${city}${region}待定点位`;
+  const skills = asStringArray(source.skills).slice(0, 3);
+  const tags = asStringArray(source.tags).slice(0, 2);
+  const salary = (source.salary as Record<string, unknown> | undefined) ?? {};
+  const amount = asNumber(salary.amount) ?? 150;
+  const skillText = skills.length ? `需要会${skills.join("、")}` : "有相关经验优先";
+  const tagText = tags.length ? `，${tags.join("，")}` : "";
+  return `明天下午${location}需要${headcount}个${title}，${duration}小时，${amount}/人${tagText}，${skillText}。`;
+}
 
 export function JobPage() {
-  const [text, setText] = useState(DEFAULT_TEXT);
+  const [text, setText] = useState("");
   const [jobCard, setJobCard] = useState<unknown>(null);
   const [matches, setMatches] = useState<any[]>([]);
   const [loading, setLoading] = useState(false);
 
+  useEffect(() => {
+    let active = true;
+    void (async () => {
+      try {
+        const result = await api.jobs();
+        if (!active) {
+          return;
+        }
+        setText((current) => current || buildAdaptiveJobText(result.items));
+      } catch {
+        if (!active) {
+          return;
+        }
+        setText((current) => current || FALLBACK_TEXT);
+      }
+    })();
+    return () => {
+      active = false;
+    };
+  }, []);
+
   const handleExtract = async () => {
     setLoading(true);
     try {

gig-poc/apps/web/src/pages/WorkerPage.tsx

@@ -1,16 +1,73 @@
-import { useState } from "react";
+import { useEffect, useState } from "react";
 import { api } from "../api/client";
 import { JsonPanel } from "../components/JsonPanel";
 import { MatchList } from "../components/MatchList";
 
-const DEFAULT_TEXT = "我做过商场促销和活动签到，也能做登记和引导，周末都能接，福田南山都方便。";
+const FALLBACK_TEXT = "我做过商场促销和活动签到，也能做登记和引导，周末都能接，福田南山都方便。";
+
+function pickRandom<T>(items: T[]): T {
+  return items[Math.floor(Math.random() * items.length)];
+}
+
+function asString(value: unknown): string {
+  return typeof value === "string" ? value.trim() : "";
+}
+
+function asStringArray(value: unknown): string[] {
+  if (!Array.isArray(value)) {
+    return [];
+  }
+  return value.filter((item): item is string => typeof item === "string" && item.trim().length > 0);
+}
+
+function buildAdaptiveWorkerText(items: Record<string, unknown>[]): string {
+  if (!items.length) {
+    return FALLBACK_TEXT;
+  }
+  const source = pickRandom(items);
+  const name = asString(source.name) || "我";
+  const regions = asStringArray(source.regions).slice(0, 2);
+  const experiences = asStringArray(source.experience_tags).slice(0, 2);
+  const skillObjects = Array.isArray(source.skills) ? source.skills : [];
+  const skills = skillObjects
+    .map((item) => (item && typeof item === "object" ? asString((item as Record<string, unknown>).name) : ""))
+    .filter(Boolean)
+    .slice(0, 3);
+  const availability = asStringArray(source.availability);
+  const expText = experiences.length ? experiences.join("和") : "活动执行";
+  const skillText = skills.length ? skills.join("、") : "沟通和执行";
+  const regionText = regions.length ? `${regions.join("、")}都方便` : "同城都方便";
+  const timeText = availability.some((item) => item.includes("weekend")) ? "周末都能接" : "时间比较灵活";
+  return `${name}做过${expText}，也能做${skillText}，${timeText}，${regionText}。`;
+}
 
 export function WorkerPage() {
-  const [text, setText] = useState(DEFAULT_TEXT);
+  const [text, setText] = useState("");
   const [workerCard, setWorkerCard] = useState<unknown>(null);
   const [matches, setMatches] = useState<any[]>([]);
   const [loading, setLoading] = useState(false);
 
+  useEffect(() => {
+    let active = true;
+    void (async () => {
+      try {
+        const result = await api.workers();
+        if (!active) {
+          return;
+        }
+        setText((current) => current || buildAdaptiveWorkerText(result.items));
+      } catch {
+        if (!active) {
+          return;
+        }
+        setText((current) => current || FALLBACK_TEXT);
+      }
+    })();
+    return () => {
+      active = false;
+    };
+  }, []);
+
   const handleExtract = async () => {
     setLoading(true);
     try {