256 lines
9.6 KiB
Python
Executable File
256 lines
9.6 KiB
Python
Executable File
#!/usr/bin/env python3
|
||
"""
|
||
process_data.py — 对 Holy Crab 任务结果进行数据分析
|
||
用法:
|
||
python3 process_data.py summarize <任务ID> 榜单摘要
|
||
python3 process_data.py sentiment <任务ID> 评论情感统计
|
||
python3 process_data.py compare <任务ID> 产品横向对比
|
||
python3 process_data.py detail <任务ID> 完整数据（含召回内容）
|
||
|
||
Examples:
|
||
python3 process_data.py summarize HC-20260727-ABC12
|
||
python3 process_data.py sentiment HC-20260727-ABC12
|
||
"""
|
||
|
||
import json
|
||
import os
|
||
import sys
|
||
from pathlib import Path
|
||
|
||
SCRIPT_DIR = Path(__file__).resolve().parent
|
||
DEFAULT_BACKEND = "http://localhost:8000"
|
||
|
||
|
||
def load_cookie() -> str | None:
|
||
env = os.environ.get("HC_SESSION")
|
||
if env:
|
||
return env
|
||
cookie_file = Path.home() / ".holy_crab_cookie"
|
||
if cookie_file.exists():
|
||
return cookie_file.read_text().strip()
|
||
return None
|
||
|
||
|
||
def fetch_task(task_id: str) -> dict:
|
||
backend = os.environ.get("HC_BACKEND_URL", DEFAULT_BACKEND).rstrip("/")
|
||
url = f"{backend}/api/v1/tasks/{task_id}"
|
||
cookie = load_cookie()
|
||
headers = {}
|
||
if cookie:
|
||
headers["Cookie"] = f"hc_session={cookie}"
|
||
|
||
import urllib.request
|
||
req = urllib.request.Request(url, headers=headers)
|
||
try:
|
||
with urllib.request.urlopen(req, timeout=15) as resp:
|
||
data = json.loads(resp.read().decode("utf-8"))
|
||
except urllib.error.HTTPError as exc:
|
||
body = exc.read().decode("utf-8", errors="replace")
|
||
try:
|
||
err = json.loads(body)
|
||
print(f"❌ API错误 {exc.code}: {err}", file=sys.stderr)
|
||
except Exception:
|
||
print(f"❌ HTTP {exc.code}: {body[:300]}", file=sys.stderr)
|
||
sys.exit(1)
|
||
|
||
result = data.get("data", {}).get("result", {})
|
||
extracted = result.get("extracted")
|
||
if not extracted:
|
||
print(f"❌ 任务 {task_id} 暂无 extracted 数据（任务未完成或格式不兼容）",
|
||
file=sys.stderr)
|
||
sys.exit(1)
|
||
return extracted
|
||
|
||
|
||
# ---------------------------------------------------------------------------
|
||
# 命令处理器
|
||
# ---------------------------------------------------------------------------
|
||
|
||
def cmd_summarize(extracted: dict) -> None:
|
||
"""榜单摘要"""
|
||
ranking = extracted.get("榜单", [])
|
||
products = extracted.get("产品详情", [])
|
||
src = extracted.get("参考来源笔记总量", {})
|
||
meta = extracted.get("提取元数据", {})
|
||
stats = meta.get("召回统计", {})
|
||
warns = extracted.get("提取警告", [])
|
||
|
||
print(f"\n{'='*60}")
|
||
print(f" 📊 榜单摘要")
|
||
print(f"{'='*60}")
|
||
print(f" 参考来源笔记总量 : {src.get('原始文本', 'N/A')} (提取: {src.get('提取数量', 'N/A')})")
|
||
print(f" 包含召回明细 : {'是' if meta.get('包含召回明细') else '否'}")
|
||
print(f" 标签总数 : {stats.get('标签总数', 0)}")
|
||
print(f" 实际召回总数 : {stats.get('实际召回内容总数', 0)}")
|
||
print(f"\n 【榜单排名】")
|
||
print(f" {'排名':<4} {'产品名称':<20} {'推荐比例':<8} {'产品详情'}")
|
||
print(f" {'─'*60}")
|
||
for item in ranking:
|
||
rank = item.get("当前排名", "")
|
||
name = item.get("产品名称", "")[:20]
|
||
ratio = item.get("推荐比例", "")
|
||
# 找对应的产品详情
|
||
prod = next((p for p in products if p.get("产品名称") == name), None)
|
||
if prod:
|
||
top3 = [t["内容标签"] for t in prod.get("标签数量Top3", [])]
|
||
detail = f"Top标签: {', '.join(top3)}"
|
||
else:
|
||
detail = ""
|
||
print(f" {str(rank):<4} {name:<20} {ratio:<8} {detail}")
|
||
|
||
if products:
|
||
print(f"\n 【各产品经验标签统计】")
|
||
for prod in products:
|
||
tags = prod.get("内容标签", [])
|
||
top3 = prod.get("标签数量Top3", [])
|
||
print(f"\n ▶ {prod.get('产品名称')} (排行{prod.get('当前排行')})")
|
||
print(f" 参考经验: {prod.get('参考经验人数')} | 推荐比例: {prod.get('推荐比例')}")
|
||
print(f" 标签 Top3: ", end="")
|
||
print(", ".join(f"{t['内容标签']}({t['经验数量']})" for t in top3) if top3 else "无")
|
||
recall_total = sum(t.get("实际召回数量", 0) for t in tags)
|
||
print(f" 实际召回总数: {recall_total}")
|
||
|
||
if warns:
|
||
print(f"\n ⚠️ 提取警告 ({len(warns)} 条):")
|
||
for w in warns[:5]:
|
||
print(f" - {w}")
|
||
if len(warns) > 5:
|
||
print(f" ... 共 {len(warns)} 条")
|
||
|
||
print(f"\n{'='*60}\n")
|
||
|
||
|
||
def cmd_sentiment(extracted: dict) -> None:
|
||
"""评论情感统计（从 comment 类型召回内容中提取）"""
|
||
all_comments = []
|
||
for prod in extracted.get("产品详情", []):
|
||
for tag in prod.get("内容标签", []):
|
||
for note in tag.get("召回内容", []):
|
||
if note.get("召回类型") == "comment":
|
||
all_comments.append({
|
||
"评论内容": note.get("评论内容"),
|
||
"所属笔记点赞": note.get("所属笔记点赞数量"),
|
||
"发布时间": note.get("发布时间"),
|
||
"产品": prod.get("产品名称"),
|
||
"标签": tag.get("内容标签"),
|
||
})
|
||
|
||
if not all_comments:
|
||
print(f"\n ⚠️ 当前任务无评论召回数据（comment 类型召回内容为空）")
|
||
print(f" 轻度/常规任务默认不包含评论召回，深度任务（deep模式）才有。\n")
|
||
return
|
||
|
||
print(f"\n{'='*60}")
|
||
print(f" 💬 评论召回统计 (共 {len(all_comments)} 条)")
|
||
print(f"{'='*60}")
|
||
|
||
# 按产品分组
|
||
by_product: dict[str, list] = {}
|
||
for c in all_comments:
|
||
p = c["产品"]
|
||
by_product.setdefault(p, []).append(c)
|
||
|
||
for product, comments in by_product.items():
|
||
print(f"\n 【{product}】({len(comments)} 条评论)")
|
||
for c in comments[:10]:
|
||
content = c["评论内容"] or ""
|
||
print(f" • {content[:60]}{'...' if len(content)>60 else ''}")
|
||
print(f" 所属笔记 👍{c['所属笔记点赞']} | {c['发布时间']}")
|
||
if len(comments) > 10:
|
||
print(f" ... 还有 {len(comments)-10} 条")
|
||
|
||
print(f"\n 📌 AI 情感分析建议：")
|
||
print(f" 请 AI 读取上述评论内容，判断每条评论的情感（正面/负面/中性），")
|
||
print(f" 并统计各情感类别的数量和占比。")
|
||
print(f"{'='*60}\n")
|
||
|
||
|
||
def cmd_compare(extracted: dict) -> None:
|
||
"""产品横向对比"""
|
||
products = extracted.get("产品详情", [])
|
||
|
||
if not products:
|
||
print(f"❌ 无产品详情数据", file=sys.stderr)
|
||
return
|
||
|
||
print(f"\n{'='*60}")
|
||
print(f" 🔍 产品横向对比")
|
||
print(f"{'='*60}")
|
||
|
||
headers = ["指标"] + [p.get("产品名称", f"产品{i+1}")[:12] for i, p in enumerate(products)]
|
||
col_w = 14
|
||
print(f" {'指标':<14} " + " ".join(f"{h:<{col_w}}" for h in headers[1:]))
|
||
print(f" {'─'*60}")
|
||
|
||
def col(val: str) -> str:
|
||
return f"{str(val):<{col_w}}"
|
||
|
||
print(f" {'当前排行':<14} " + " ".join(col(p.get("当前排行", "")) for p in products))
|
||
print(f" {'参考经验人数':<14} " + " ".join(col(p.get("参考经验人数", "")) for p in products))
|
||
print(f" {'推荐比例':<14} " + " ".join(col(p.get("推荐比例", "")) for p in products))
|
||
|
||
tag_counts = [len(p.get("内容标签", [])) for p in products]
|
||
print(f" {'标签数量':<14} " + " ".join(col(c) for c in tag_counts))
|
||
|
||
recall_totals = []
|
||
for p in products:
|
||
total = sum(t.get("实际召回数量", 0) for t in p.get("内容标签", []))
|
||
recall_totals.append(total)
|
||
print(f" {'实际召回总数':<14} " + " ".join(col(r) for r in recall_totals))
|
||
|
||
top_tags = []
|
||
for p in products:
|
||
top3 = p.get("标签数量Top3", [])
|
||
tags_str = ", ".join(t["内容标签"] for t in top3[:2]) if top3 else "—"
|
||
top_tags.append(tags_str)
|
||
print(f" {'Top1-2标签':<14} " + " ".join(f"{t[:col_w*2-2]:<{col_w*2}}" for t in top_tags))
|
||
|
||
print(f"\n 【各产品 Top3 标签详情】")
|
||
for i, prod in enumerate(products):
|
||
print(f"\n {i+1}. {prod.get('产品名称')}")
|
||
for t in prod.get("标签数量Top3", []):
|
||
recall = next(
|
||
(x.get("实际召回数量", 0)
|
||
for x in prod.get("内容标签", [])
|
||
if x.get("内容标签") == t["内容标签"]),
|
||
0
|
||
)
|
||
print(f" {t['排名']}. {t['内容标签']} {t['经验数量']}人 召回:{recall}")
|
||
|
||
print(f"\n{'='*60}\n")
|
||
|
||
|
||
def cmd_detail(extracted: dict) -> None:
|
||
"""完整数据（含所有召回内容）"""
|
||
print(json.dumps(extracted, ensure_ascii=False, indent=2))
|
||
|
||
|
||
# ---------------------------------------------------------------------------
|
||
# 主入口
|
||
# ---------------------------------------------------------------------------
|
||
|
||
COMMANDS = {
|
||
"summarize": ("榜单摘要", cmd_summarize),
|
||
"sentiment": ("评论情感统计", cmd_sentiment),
|
||
"compare": ("产品横向对比", cmd_compare),
|
||
"detail": ("完整数据", cmd_detail),
|
||
}
|
||
|
||
|
||
def main() -> None:
|
||
if len(sys.argv) < 3 or sys.argv[1] not in COMMANDS:
|
||
print(f"用法: {sys.argv[0]} <{'|'.join(COMMANDS)}> <任务ID>")
|
||
print(f"示例: {sys.argv[0]} summarize HC-20260727-ABC12")
|
||
sys.exit(1)
|
||
|
||
cmd_key = sys.argv[1]
|
||
task_id = sys.argv[2]
|
||
_, handler = COMMANDS[cmd_key]
|
||
|
||
extracted = fetch_task(task_id)
|
||
handler(extracted)
|
||
|
||
|
||
if __name__ == "__main__":
|
||
main()
|