项目初始化

2026-07-02 11:31:16 +08:00
commit eef4c76e3f
64 changed files with 13369 additions and 0 deletions
--- a/daily/format_wecom.py
+++ b/daily/format_wecom.py
@@ -0,0 +1,281 @@
+"""企微早报排版。"""
+
+from __future__ import annotations
+
+import re
+from typing import Any
+
+from daily.config import wecom_skill_desc_limit
+from daily.localize import LocalizeJob, localize_brief_descriptions, needs_chinese
+from daily.text_utils import trim_brief
+
+ICONS = {
+    "header": "📰",
+    "highlights": "💡",
+    "trending": "📈",
+    "hot": "🔥",
+    "github": "🐙",
+    "emerging": "🌱",
+    "topic": "🤖",
+    "ainews": "🌍",
+    "pick": "📦",
+    "theme": "🎯",
+    "file": "📄",
+}
+
+
+def _skill_line(rank: int, item: dict[str, Any], *, badge: str = "") -> list[str]:
+    source = item.get("source", "?")
+    installs = item.get("installs_fmt", "?")
+    link = item.get("link", "")
+    desc = item.get("desc_short", "")
+    badge_prefix = f"{badge} " if badge else ""
+    if item.get("cluster"):
+        count = int(item.get("cluster_count") or 1)
+        sample = item.get("cluster_titles") or item.get("title", "")
+        label = f"**{source}** · {count} skills · **{installs}**"
+        if link:
+            head = f"{rank}. {badge_prefix}[{label}]({link})"
+        else:
+            head = f"{rank}. {badge_prefix}{label}"
+        lines = [head]
+        if sample or desc:
+            hint = desc or sample
+            lines.append(f"   > {hint}")
+        return lines
+    title = item.get("title", "?")
+    if link:
+        head = f"{rank}. {badge_prefix}[**{title}**]({link}) · `{source}` · **{installs}**"
+    else:
+        head = f"{rank}. {badge_prefix}**{title}** · `{source}` · **{installs}**"
+    lines = [head]
+    if desc:
+        lines.append(f"   > {desc}")
+    return lines
+
+
+def _ai_news_lines(items: list[dict[str, Any]]) -> list[str]:
+    lines: list[str] = []
+    for i, item in enumerate(items, 1):
+        title = item.get("title", "?")
+        link = item.get("link", "")
+        source = item.get("source_name", "?")
+        pub = item.get("published_fmt", "")
+        desc = item.get("desc_short", "")
+        pub_suffix = f" · {pub}" if pub else ""
+        if link:
+            head = f"{i}. [**{title}**]({link}) · `{source}`{pub_suffix}"
+        else:
+            head = f"{i}. **{title}** · `{source}`{pub_suffix}"
+        lines.append(head)
+        if desc:
+            lines.append(f"   > {desc}")
+    return lines
+
+
+def _github_repo_lines(repos: list[dict[str, Any]], *, show_created: bool = False) -> list[str]:
+    lines: list[str] = []
+    for i, repo in enumerate(repos, 1):
+        name = repo["repo"]
+        url = repo["url"]
+        lang = repo.get("language", "")
+        stars_today = repo.get("stars_today_fmt", "")
+        total = repo.get("total_stars_fmt", "")
+        created = repo.get("created_at", "")
+        meta_parts: list[str] = []
+        if lang:
+            meta_parts.append(lang)
+        if stars_today:
+            meta_parts.append(f"+{stars_today} today")
+        elif total:
+            meta_parts.append(f"⭐{total}")
+        if show_created and created:
+            meta_parts.append(f"创建于 {created}")
+        meta = f" · {' · '.join(meta_parts)}" if meta_parts else ""
+        lines.append(f"{i}. [{name}]({url}){meta}")
+        desc = repo.get("desc_short") or repo.get("description", "")
+        if desc:
+            lines.append(f"   > {desc}")
+    return lines
+
+
+def _fallback_skill_desc(item: dict[str, Any]) -> str:
+    if item.get("cluster"):
+        count = int(item.get("cluster_count") or 1)
+        source = item.get("source") or "unknown"
+        sample = item.get("cluster_titles") or item.get("title") or ""
+        return f"{count} agent skills from {source}, including {sample}"
+    title = item.get("title") or "skill"
+    source = item.get("source") or "unknown"
+    return f"{title} skill from {source}"
+
+
+def _skill_group_key(item: dict[str, Any]) -> str:
+    return str(item.get("id") or f"{item.get('source')}/{item.get('title')}")
+
+
+def finalize_wecom_skill_groups(
+    items: list[dict[str, Any]],
+    *,
+    desc_limit: int | None = None,
+) -> list[dict[str, Any]]:
+    """为企微 Skills 榜生成一句简要中文简介（与完整版 .md 长描述分离）。"""
+    if desc_limit is None:
+        desc_limit = wecom_skill_desc_limit()
+    limit = desc_limit if desc_limit > 0 else 48
+    copies: list[tuple[str, dict[str, Any]]] = []
+    jobs: list[LocalizeJob] = []
+    for item in items:
+        copy = dict(item)
+        desc = (copy.get("description") or "").strip()
+        if not desc:
+            desc = _fallback_skill_desc(copy)
+        key = _skill_group_key(copy)
+        job_key = f"wecom:{key}"
+        if needs_chinese(desc) or len(desc) > limit:
+            jobs.append(LocalizeJob(job_key, desc, limit))
+        else:
+            copy["wecom_desc"] = desc
+        copies.append((job_key, copy))
+
+    zh_map = localize_brief_descriptions(jobs, archive=True)
+    out: list[dict[str, Any]] = []
+    for job_key, copy in copies:
+        if job_key in zh_map:
+            copy["wecom_desc"] = zh_map[job_key]
+        elif "wecom_desc" not in copy:
+            copy["wecom_desc"] = _brief_fallback_desc(
+                (copy.get("description") or "").strip() or _fallback_skill_desc(copy),
+                limit,
+            )
+        out.append(copy)
+    return out
+
+
+def _brief_fallback_desc(text: str, limit: int) -> str:
+    return trim_brief(text, limit)
+
+
+def _grouped_skill_to_wecom_item(
+    item: dict[str, Any],
+    *,
+    desc_limit: int | None = None,
+) -> dict[str, Any]:
+    if desc_limit is None:
+        desc_limit = wecom_skill_desc_limit()
+    limit = desc_limit if desc_limit > 0 else 48
+    installs = int(item.get("installs") or 0)
+    installs_fmt = item.get("installs_fmt") or str(installs)
+    desc = (item.get("wecom_desc") or item.get("description") or "").strip()
+    if not desc and item.get("cluster"):
+        desc = item.get("cluster_titles") or ""
+    if not item.get("wecom_desc"):
+        desc = _brief_fallback_desc(desc, limit)
+    return {
+        "title": item.get("title", ""),
+        "source": item.get("source", "?"),
+        "installs_fmt": installs_fmt,
+        "link": item.get("link", ""),
+        "desc_short": desc,
+        "cluster": bool(item.get("cluster")),
+        "cluster_count": item.get("cluster_count"),
+        "cluster_titles": item.get("cluster_titles"),
+    }
+
+
+def build_skills_board_section(icon_key: str, board_label: str, items: list[dict[str, Any]]) -> str:
+    prepared = finalize_wecom_skill_groups(items)
+    wecom_items = [_grouped_skill_to_wecom_item(x) for x in prepared]
+    lines = [f"{ICONS[icon_key]} **{board_label} Top {len(wecom_items)}**"]
+    for rank, item in enumerate(wecom_items, 1):
+        lines.extend(_skill_line(rank, item))
+    return "\n".join(lines)
+
+
+_SKILL_SECTIONS = re.compile(
+    r"📈 \*\*Skills Trending.*?(?=🐙 \*\*GitHub Trending)",
+    re.DOTALL,
+)
+
+
+def replace_wecom_skill_sections(
+    md: str,
+    *,
+    trending: list[dict[str, Any]],
+    hot: list[dict[str, Any]],
+) -> str:
+    """用 Python 合并后的 Skills 榜替换或插入 Agent 早报中的对应区块。"""
+    trending_sec = build_skills_board_section("trending", "Skills Trending", trending)
+    hot_sec = build_skills_board_section("hot", "Skills Hot", hot)
+    replacement = f"{trending_sec}\n\n{hot_sec}\n\n"
+    if _SKILL_SECTIONS.search(md):
+        return _SKILL_SECTIONS.sub(replacement, md)
+    github_marker = "🐙 **GitHub Trending"
+    idx = md.find(github_marker)
+    if idx >= 0:
+        return md[:idx] + replacement + md[idx:]
+    return md.rstrip() + "\n\n" + replacement
+
+
+def build_wecom_report(
+    *,
+    date_str: str,
+    time_str: str,
+    updated: str,
+    highlights: list[str],
+    theme_line: str,
+    trending: list[dict[str, Any]],
+    hot: list[dict[str, Any]],
+    repos: list[dict[str, Any]],
+    emerging: list[dict[str, Any]],
+    topic_name: str,
+    topic_repos: list[dict[str, Any]],
+    ai_news: list[dict[str, Any]] | None = None,
+    pick_command: str,
+) -> str:
+    lines = [
+        f"{ICONS['header']} **早报 · {date_str}**",
+        f"> ⏱ {time_str} · 数据截至 {updated}",
+        "",
+        f"{ICONS['highlights']} **今日速览**",
+    ]
+    for point in highlights[:3]:
+        lines.append(f"> {point}")
+    lines.append("")
+    lines.append(f"{ICONS['theme']} {theme_line}")
+    lines.append("")
+
+    if ai_news:
+        lines.append(f"{ICONS['ainews']} **国际 AI 时讯 Top {len(ai_news)}**")
+        lines.extend(_ai_news_lines(ai_news))
+        lines.append("")
+
+    lines.append(f"{ICONS['trending']} **Skills Trending Top {len(trending)}**")
+    for rank, item in enumerate(trending, 1):
+        lines.extend(_skill_line(rank, item, badge=item.get("badge", "")))
+    lines.append("")
+
+    lines.append(f"{ICONS['hot']} **Skills Hot Top {len(hot)}**")
+    for rank, item in enumerate(hot, 1):
+        lines.extend(_skill_line(rank, item, badge=item.get("badge", "")))
+    lines.append("")
+
+    if repos:
+        lines.append(f"{ICONS['github']} **GitHub Trending Top {len(repos)}**")
+        lines.extend(_github_repo_lines(repos))
+        lines.append("")
+
+    if emerging:
+        lines.append(f"{ICONS['emerging']} **新兴项目 Top {len(emerging)}**")
+        lines.extend(_github_repo_lines(emerging, show_created=True))
+        lines.append("")
+
+    if topic_repos:
+        lines.append(f"{ICONS['topic']} **Topic `{topic_name}` Top {len(topic_repos)}**")
+        lines.extend(_github_repo_lines(topic_repos))
+        lines.append("")
+
+    lines.append(f"{ICONS['pick']} **今日首推**")
+    lines.append(f"`{pick_command}`")
+
+    return "\n".join(lines)