check_files.py

# /home/wangxinxin/my_graphrag/check_files.py
import pandas as pd
import os

# Check text_units.parquet (stores raw text chunks)
FILE_PATH = "output/text_units.parquet"

if os.path.exists(FILE_PATH):
    df = pd.read_parquet(FILE_PATH)
    # Search for text blocks containing "AHG" or "顶峰" (Apex/Summit)
    mask = df['text'].str.contains("AHG|顶峰", case=False, na=False)
    results = df[mask]
    
    if not results.empty:
        print(f"Successfully found {len(results)} text chunks containing 'AHG/顶峰'!")
        print("Sample text snippet:")
        print(results.iloc[0]['text'][:100] + "...")
    else:
        print("No content containing 'AHG' or '顶峰' was found in text_units.parquet.")
        print("Possible reasons: The file was not indexed, or the filename is not in the processing list.")
else:
    print(f"File does not exist: {FILE_PATH}")