Why are there no operation symbols in the extracted table?

[ijjih2]

Catalyst,Reduction temperature (8C),Hydrogen uptake (mmol/g),Pd dispersion (%)
2% Pd/ZnO,200,8 � 1,8.6
2% Pd/ZnO,300,8 � 2,8.6
2% Pd/ZnO,400,2.7 � 0.4,2.9
2% Pd/ZnO,500,3.1 � 0.4,3.3
6% Pd/ZnO,200,17 � 2,6.2
10% Pd/ZnO,200,25 � 4,5.4
2% Pd/CeO2,200,45 � 9,48
2% Pd/CeO2,400,36 � 5,38
2% Pd/CeO2,550,12 � 2,13
6% Pd/CeO2,200,98 � 18,34
10% Pd/CeO2,200,116 � 18,25

page_4.pdf

import re
import fitz # PyMuPDF
from gmft.auto import TableDetector, AutoTableFormatter
import pandas as pd
from gmft.pdf_bindings import PyPDFium2Document

def extract_pdf_content(pdf_path):
# 使用gmft提取表格
detector = TableDetector()
formatter = AutoTableFormatter()

# 读取PDF文件
doc_tables = PyPDFium2Document(pdf_path)
tables = []

# 提取表格
for page_index, page in enumerate(doc_tables):
    detected_tables = detector.extract(page)
    for table in detected_tables:
        formatted_table = formatter.format(table)

        # 获取表格数据，调用df方法获取DataFrame
        table_data = formatted_table.df()  # 获取表格内容

        if table_data is not None:
            # 将表格数据转换为DataFrame并存储
            df = table_data
            tables.append({
                "page": page_index + 1,
                "position": table.bbox,
                "dataframe": df,  # 存储DataFrame
            })
doc_tables.close()

# 使用 PyMuPDF 提取文本
doc_text = fitz.open(pdf_path)
full_text = []
for i in range(len(doc_text)):
    page = doc_text[i]
    text = page.get_text("text")  # 获取页面文本
    full_text.append({
        "page": i + 1,
        "text": text
    })
doc_text.close()

return {"tables": tables, "text": full_text}

if name == "main":
pdf_path = "1.pdf"
result = extract_pdf_content(pdf_path)

print(f"提取到 {len(result['tables'])} 个表格:")
for table in result['tables']:
    print(f"\n第 {table['page']} 页表格（位置：{table['position']}）")
    print(table['dataframe'])  # 输出表格

print("\n提取文本示例:")
for text_block in result['text']:  # 输出所有页面的文本
    print(f"\n第 {text_block['page']} 页文本：\n{text_block['text']}")  # 输出完整文本