项目初始化

daily/skills_group.py

@@ -0,0 +1,103 @@
+"""Skills 榜单：同 source 合并为一条（企微 Top N）。"""
+
+from __future__ import annotations
+
+from typing import Any
+
+
+def skill_id(item: dict[str, Any]) -> str:
+    return str(item.get("id") or f"{item.get('source')}/{item.get('title')}")
+
+
+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 _installs_range(items: list[dict[str, Any]]) -> tuple[int, int, str]:
+    values = [int(item.get("installs") or 0) for item in items]
+    lo, hi = min(values), max(values)
+    if lo == hi:
+        return lo, hi, format_installs(hi)
+    return lo, hi, f"{format_installs(lo)}–{format_installs(hi)}"
+
+
+def _best_description(items: list[dict[str, Any]]) -> str:
+    for item in items:
+        desc = (item.get("description") or "").strip()
+        if desc:
+            return desc
+    return ""
+
+
+def _cluster_skill(items: list[dict[str, Any]]) -> dict[str, Any]:
+    ranked = sorted(items, key=lambda x: int(x.get("installs") or 0), reverse=True)
+    top = ranked[0]
+    _lo, _hi, installs_fmt = _installs_range(ranked)
+    titles = [str(x.get("title") or "") for x in ranked if x.get("title")]
+    sample = ", ".join(titles[:4])
+    if len(titles) > 4:
+        sample = f"{sample}…"
+    return {
+        "id": skill_id(top),
+        "title": top.get("title", ""),
+        "source": top.get("source", ""),
+        "installs": int(top.get("installs") or 0),
+        "installs_min": _lo,
+        "installs_max": _hi,
+        "installs_fmt": installs_fmt,
+        "link": top.get("link", ""),
+        "description": _best_description(ranked),
+        "cluster": True,
+        "cluster_count": len(ranked),
+        "cluster_skills": titles,
+        "cluster_titles": sample,
+    }
+
+
+def _single_skill(item: dict[str, Any]) -> dict[str, Any]:
+    installs = int(item.get("installs") or 0)
+    return {
+        "id": skill_id(item),
+        "title": item.get("title", ""),
+        "source": item.get("source", ""),
+        "installs": installs,
+        "installs_fmt": format_installs(installs),
+        "link": item.get("link", ""),
+        "description": item.get("description", ""),
+        "cluster": False,
+    }
+
+
+def group_skills_by_source(
+    items: list[dict[str, Any]],
+    *,
+    limit: int = 10,
+    pool_size: int = 50,
+) -> list[dict[str, Any]]:
+    """按 source 去重合并；保留各 source 在榜内的最佳名次顺序。"""
+    if not items or limit <= 0:
+        return []
+
+    pool = items[: max(pool_size, limit)]
+    by_source: dict[str, list[dict[str, Any]]] = {}
+    first_rank: dict[str, int] = {}
+    for rank, item in enumerate(pool, 1):
+        source = (item.get("source") or "?").strip() or "?"
+        by_source.setdefault(source, []).append(item)
+        first_rank.setdefault(source, rank)
+
+    ordered_sources = sorted(by_source.keys(), key=lambda s: first_rank[s])
+    result: list[dict[str, Any]] = []
+    for source in ordered_sources:
+        group = by_source[source]
+        if len(group) == 1:
+            result.append(_single_skill(group[0]))
+        else:
+            result.append(_cluster_skill(group))
+        if len(result) >= limit:
+            break
+    return result