refactor: Phase 1 抽出 shared/skills_data 解耦 daily 与 bot

daily 不再通过 sys.path 导入 bot/skills_service，skills feed 数据层上移到 shared/。

Co-authored-by: Cursor <cursoragent@cursor.com>

bot/env_config.py

@@ -3,11 +3,16 @@
 from __future__ import annotations
 
 import os
+import sys
 from pathlib import Path
 
 from dotenv import load_dotenv
 
 _BOT_DIR = Path(__file__).resolve().parent
+_REPO_ROOT = _BOT_DIR.parent
+if str(_REPO_ROOT) not in sys.path:
+    sys.path.insert(0, str(_REPO_ROOT))
+
 load_dotenv(_BOT_DIR / ".env")
 load_dotenv(_BOT_DIR / ".env.local", override=True)
 

bot/skills_service.py

@@ -1,32 +1,20 @@
-"""skills.sh 数据查询与命令解析。"""
+"""skills.sh 快查命令解析与企微回复格式化。"""
 
 from __future__ import annotations
 
-import json
-import logging
 import re
-import time
+import sys
 from dataclasses import dataclass
 from pathlib import Path
 from typing import Any, Literal
 
-import certifi
-import httpx
+_REPO_ROOT = Path(__file__).resolve().parent.parent
+if str(_REPO_ROOT) not in sys.path:
+    sys.path.insert(0, str(_REPO_ROOT))
 
-logger = logging.getLogger(__name__)
+from shared.skills_data import Board, board_items, format_installs, load_feed, warm_feed_cache
 
-FEED_URLS = [
-    # jsDelivr 在国内通常比 raw.githubusercontent.com 更稳定
-    "https://cdn.jsdelivr.net/gh/NeverSight/skills.sh_feed@main/data/feed.json",
-    "https://raw.githubusercontent.com/NeverSight/skills.sh_feed/main/data/feed.json",
-]
-CACHE_TTL_SECONDS = 600
-CACHE_DIR = Path(__file__).resolve().parent / ".cache"
-CACHE_FILE = CACHE_DIR / "feed.json"
-
-_cache: dict[str, Any] = {"data": None, "fetched_at": 0.0}
-
-Board = Literal["trending", "hot", "all"]
+__all__ = ["Command", "handle_command", "parse_command", "warm_feed_cache"]
 
 
 @dataclass
@@ -37,72 +25,6 @@ class Command:
     query: str = ""
 
 
-def _fetch_json(url: str) -> dict[str, Any]:
-    headers = {
-        "User-Agent": "skills-hot-bot/1.0",
-        "Accept": "application/json",
-    }
-    with httpx.Client(
-        timeout=httpx.Timeout(20.0, connect=10.0),
-        verify=certifi.where(),
-        follow_redirects=True,
-    ) as client:
-        resp = client.get(url, headers=headers)
-        resp.raise_for_status()
-        return resp.json()
-
-
-def _load_disk_cache() -> dict[str, Any] | None:
-    if not CACHE_FILE.exists():
-        return None
-    try:
-        return json.loads(CACHE_FILE.read_text(encoding="utf-8"))
-    except (OSError, json.JSONDecodeError) as exc:
-        logger.warning("读取本地缓存失败: %s", exc)
-        return None
-
-
-def _save_disk_cache(data: dict[str, Any]) -> None:
-    CACHE_DIR.mkdir(parents=True, exist_ok=True)
-    CACHE_FILE.write_text(json.dumps(data, ensure_ascii=False), encoding="utf-8")
-
-
-def load_feed(force: bool = False) -> dict[str, Any]:
-    now = time.time()
-    if not force and _cache["data"] and now - _cache["fetched_at"] < CACHE_TTL_SECONDS:
-        return _cache["data"]
-
-    errors: list[str] = []
-    for url in FEED_URLS:
-        for attempt in range(3):
-            try:
-                data = _fetch_json(url)
-                _cache["data"] = data
-                _cache["fetched_at"] = now
-                _save_disk_cache(data)
-                logger.info("skills 数据已更新: %s", url)
-                return data
-            except Exception as exc:
-                msg = f"{url} (#{attempt + 1}): {exc}"
-                errors.append(msg)
-                logger.debug("拉取失败 %s", msg)
-                time.sleep(0.5 * (attempt + 1))
-
-    stale = _load_disk_cache()
-    if stale:
-        logger.warning("网络不可用，回退到本地缓存")
-        _cache["data"] = stale
-        _cache["fetched_at"] = now
-        return stale
-
-    raise RuntimeError(f"无法获取 skills 数据。最近错误: {errors[-1]}")
-
-
-def warm_feed_cache() -> None:
-    """启动时预加载，避免首条消息才触发网络请求。"""
-    load_feed(force=True)
-
-
 def _normalize_text(text: str) -> str:
     text = re.sub(r"@\S+\s*", "", text)
     return text.strip().lower()
@@ -157,19 +79,6 @@ def parse_command(text: str) -> Command:
     return Command(kind="search", query=raw, limit=5)
 
 
-def _format_installs(n: int | float) -> str:
-    if n >= 1_000_000:
-        return f"{n / 1_000_000:.1f}M"
-    if n >= 1_000:
-        return f"{n / 1_000:.1f}K"
-    return str(int(n))
-
-
-def _board_items(feed: dict[str, Any], board: Board) -> list[dict[str, Any]]:
-    key = {"trending": "topTrending", "hot": "topHot", "all": "topAllTime"}[board]
-    return feed.get(key, [])
-
-
 def _board_title(board: Board) -> str:
     return {
         "trending": "Trending（近期增长）",
@@ -180,7 +89,7 @@ def _board_title(board: Board) -> str:
 
 def format_list(board: Board, limit: int) -> str:
     feed = load_feed()
-    items = _board_items(feed, board)[:limit]
+    items = board_items(feed, board)[:limit]
     updated = feed.get("updatedAt", "未知")[:10]
 
     lines = [
@@ -192,7 +101,7 @@ def format_list(board: Board, limit: int) -> str:
     for i, item in enumerate(items, 1):
         title = item.get("title", "?")
         source = item.get("source", "?")
-        installs = _format_installs(item.get("installs", 0))
+        installs = format_installs(item.get("installs", 0))
         desc = item.get("description", "")
         if len(desc) > 80:
             desc = desc[:77] + "..."
@@ -241,7 +150,7 @@ def format_search(query: str, limit: int) -> str:
     for i, item in enumerate(matches, 1):
         title = item.get("title", "?")
         source = item.get("source", "?")
-        installs = _format_installs(item.get("installs", 0))
+        installs = format_installs(item.get("installs", 0))
         link = item.get("link", "")
         lines.append(f"{i}. **{title}** · {installs} · `{source}`")
         if link:
@@ -270,7 +179,7 @@ def format_detail(name: str) -> str:
         [
             f"**{best.get('title', '?')}**",
             f"`{best.get('source', '?')}`",
-            f"安装量：**{_format_installs(best.get('installs', 0))}**",
+            f"安装量：**{format_installs(best.get('installs', 0))}**",
             "",
             desc,
             "",