Python实现Word文档中提取表格数据并转换为CSV和JSON格式

import os
import csv
import json
from docx import Document
from collections import defaultdict
from lxml import etree

def extract_tables_from_docx(docx_path):
    doc = Document(docx_path)
    all_tables_data = []

    for table in doc.tables:
        table_data = []
        merged_cells = defaultdict(str)
        row_spans = defaultdict(lambda: 0)
        for i, row in enumerate(table.rows):
            row_data = []
            for j, cell in enumerate(row.cells):
                cell_text = cell.text.strip()
                cell_xml = etree.fromstring(cell._element.xml)

                nsmap = {'w': 'http://schemas.openxmlformats.org/wordprocessingml/2006/main'}
                grid_span = cell_xml.xpath(".//w:gridSpan/@w:val", namespaces=nsmap)
                if grid_span:
                    span = int(grid_span[0])
                    for k in range(span):
                        if k == 0:
                            row_data.append(cell_text)
                        else:
                            merged_cells[(i, j + k)] = cell_text
                else:
                    row_data.append(cell_text)

                v_merge = cell_xml.xpath(".//w:vMerge/@w:val", namespaces=nsmap)
                if v_merge:
                    if v_merge[0] == 'restart':
                        row_spans[(i, j)] = 1
                        merged_cells[(i, j)] = cell_text
                    elif v_merge[0] is None:
                        row_spans[(i, j)] += 1
                        row_data[-1] = merged_cells[(i - row_spans[(i, j)], j)]

            table_data.append(row_data)
        all_tables_data.append(table_data)

    return all_tables_data

def save_tables_to_csv(tables, output_dir, file_name):
    for i, table in enumerate(tables):
        csv_path = os.path.join(output_dir, f"{file_name}_table_{i+1}.csv")
        with open(csv_path, mode='w', newline='', encoding='utf-8') as file:
            writer = csv.writer(file)
            writer.writerows(table)
        print(f"表格 {i+1} 已保存为 CSV 文件，路径为 {csv_path}")

def save_tables_to_json(tables, output_dir, file_name):
    for i, table in enumerate(tables):
        json_path = os.path.join(output_dir, f"{file_name}_table_{i+1}.json")
        with open(json_path, mode='w', encoding='utf-8') as file:
            json.dump(table, file, ensure_ascii=False, indent=4)
        print(f"表格 {i+1} 已保存为 JSON 文件，路径为 {json_path}")

# 设置文件路径和输出目录
docx_path = r'E:\data\\测试表格'
output_dir = r'E:\data\\测试表格'

# 获取文件夹中所有 Word 文件的列表
word_files = [f for f in os.listdir(docx_path) if f.endswith('.doc') or f.endswith('.docx')]


# 提取表格数据并保存为 CSV 和 JSON 文件
for file in word_files:
    file_path = os.path.join(docx_path, file)
    tables = extract_tables_from_docx(file_path)
    file_name = os.path.splitext(file)[0]
    save_tables_to_csv(tables, output_dir, file_name)
    save_tables_to_json(tables, output_dir, file_name)