
import sys
sys.path.insert(0, '/workspace/.local/lib/python3.11/site-packages')

# 先读取 PDF 原始字节，找 Creator 信息
with open('/workspace/skill/sql-report-generator/output/电力能耗趋势报告_1775543360.pdf', 'rb') as f:
    raw = f.read(2000).decode('latin-1', errors='replace')
    print("PDF 头部信息:")
    print(raw[:500])

# 再用 pypdf 读取元数据
try:
    from pypdf import PdfReader
    r = PdfReader('/workspace/skill/sql-report-generator/output/电力能耗趋势报告_1775543360.pdf')
    print("\n元数据:")
    print(r.metadata)
    print("\n第1页文本:")
    print(r.pages[0].extract_text())
except Exception as e:
    print("pypdf error:", e)
