contentm_agent/执行会话/目标-调研5个开源内容工作台项目并生成分析文档-5199a6/取证/_easel_tree_stat.py

# -*- coding: utf-8 -*-
"""Easel 文件树统计（第 3 棒）—— 只读复用本目标目录内的 easel.tree.json，不重新下载。
产出：取证/easel.tree.stat.json（可复算）。
用法：python _easel_tree_stat.py
"""
import json, os, sys
from collections import Counter, defaultdict

HERE = os.path.dirname(os.path.abspath(__file__))
SRC = os.path.join(HERE, "api", "easel.tree.json")
OUT = os.path.join(HERE, "easel.tree.stat.json")

d = json.load(open(SRC, encoding="utf-8"))
tree = d["tree"]
blobs = [x for x in tree if x["type"] == "blob"]
paths = [x["path"] for x in blobs]

res = {
    "_source": "api/easel.tree.json（GitHub trees API, recursive=1, default branch main）",
    "_note": "路径级统计；truncated=%s" % d.get("truncated"),
    "blob_count": len(blobs),
    "tree_count": len(tree) - len(blobs),
    "total_entries": len(tree),
}

# 顶层目录分布
top = Counter(p.split("/")[0] if "/" in p else "(root)" for p in paths)
res["top_level"] = dict(sorted(top.items(), key=lambda kv: -kv[1]))

# 二级目录分布（用于 web/、skills/、docs/、easel/ 内部）
lvl2 = Counter("/".join(p.split("/")[:2]) for p in paths if p.count("/") >= 1)
res["level2"] = dict(sorted(lvl2.items(), key=lambda kv: -kv[1]))

# 扩展名
ext = Counter(os.path.splitext(p)[1].lower() or "(noext)" for p in paths)
res["ext_top"] = dict(sorted(ext.items(), key=lambda kv: -kv[1])[:25])

# 核心面：skills/openclaw 下的一级子目录 == 一个技能（SKILL.md 所在目录）
sk = [p for p in paths if p.startswith("skills/openclaw/")]
res["skills_openclaw_blobs"] = len(sk)
skill_dirs = sorted({p.split("/")[2] for p in sk if p.count("/") >= 2})
res["skills_openclaw_dir_count"] = len(skill_dirs)
res["skills_openclaw_dirs"] = skill_dirs
res["skills_openclaw_skillmd"] = sorted(p for p in sk if p.endswith("/SKILL.md"))
res["skills_openclaw_skillmd_count"] = len([p for p in sk if p.endswith("/SKILL.md")])

# skills/shared
sh = [p for p in paths if p.startswith("skills/shared/")]
res["skills_shared_blobs"] = len(sh)
res["skills_shared_top"] = dict(sorted(Counter("/".join(p.split("/")[:4]) for p in sh).items(), key=lambda kv: -kv[1])[:20])

# docs
dc = [p for p in paths if p.startswith("docs/")]
res["docs_files"] = sorted(dc)

# web
web = [p for p in paths if p.startswith("web/")]
res["web_blobs"] = len(web)
res["web_level3"] = dict(sorted(Counter("/".join(p.split("/")[:3]) for p in web).items(), key=lambda kv: -kv[1])[:25])
res["web_tsx_count"] = len([p for p in web if p.endswith(".tsx")])
res["web_ts_count"] = len([p for p in web if p.endswith(".ts")])
res["web_py_count"] = len([p for p in web if p.endswith(".py")])

# easel CLI
cli = [p for p in paths if p.startswith("easel/")]
res["easel_cli_files"] = cli

# profiles / outputs
res["profiles_files"] = sorted(p for p in paths if p.startswith("profiles/"))
res["outputs_files"] = len([p for p in paths if p.startswith("outputs/")])

# openclaw 同步面
res["openclaw_files"] = sorted(p for p in paths if p.startswith("openclaw/"))

# 媒体/素材
res["assets_blobs"] = len([p for p in paths if p.startswith("assets/")])
res["aux_scripts_py"] = sorted(p for p in paths if p.endswith(".py") and not p.startswith(("skills/", "web/", "easel/")))

with open(OUT, "w", encoding="utf-8") as f:
    json.dump(res, f, ensure_ascii=False, indent=1)

# 控制台摘要
def show(k):
    v = res[k]
    print(f"{k}: {v if not isinstance(v,(list,dict)) else (len(v) if isinstance(v,list) else v)}")

print("== blob/tree ==", res["blob_count"], res["tree_count"], "truncated:", d.get("truncated"))
print("\n== top_level ==")
for k, v in res["top_level"].items():
    print(f"  {k}: {v}")
print("\n== skills/openclaw: blobs=%d dirs=%d SKILL.md=%d ==" % (
    res["skills_openclaw_blobs"], res["skills_openclaw_dir_count"], res["skills_openclaw_skillmd_count"]))
print("\n== docs ==")
for p in res["docs_files"]:
    print("  ", p)
print("\n== web: blobs=%d tsx=%d ts=%d py=%d ==" % (
    res["web_blobs"], res["web_tsx_count"], res["web_ts_count"], res["web_py_count"]))
for k, v in res["web_level3"].items():
    print(f"  {k}: {v}")
print("\n== easel CLI ==")
for p in res["easel_cli_files"]:
    print("  ", p)
print("\n== profiles ==")
for p in res["profiles_files"]:
    print("  ", p)
print("\n== openclaw sync ==")
for p in res["openclaw_files"]:
    print("  ", p)
print("\n== ext_top ==")
for k, v in res["ext_top"].items():
    print(f"  {k}: {v}")
print("\nwrote", OUT)