# pip install pywin32
# https://blog.csdn.net/weixin_42927998/article/details/115086797
import win32com
from win32com.client import Dispatch

docApp = win32com.client.Dispatch('Word.Application')
# 是不是打Word显示
docApp.Visible = False
docApp.DisplayAlerts = 0
# doc = docApp.Documents.Open('c:/1.docx')
# doc = docApp.Documents.Open('c:/昭通市人口变化及其对教育的影响20240416.docx')
# doc = docApp.Documents.Open('c:/昆明市人口变化及其对教育的影响20240419.docx')
doc = docApp.Documents.Open('c:/红河哈尼族彝族自治州人口变化及其对教育的影响20240419.docx')

# 遍历文档中的所有内嵌形状
# idx = 1
# for inline_shape in doc.InlineShapes:
#     if inline_shape.Type == win32com.client.constants.wdInlineShapeChart:  # 检查是否为内嵌图表
#         shape = doc.InlineShapes(idx)
#         # 获取图表的标题,此项目中图表没有标题
#         # print(shape.Chart.ChartTitle.Text)
#         sheet = shape.Chart.ChartData.Workbook.Worksheets("Sheet1")
#         # 行数
#         row_size = sheet.UsedRange.rows.Count
#         # 列数
#         col_size = sheet.UsedRange.columns.Count
#         # 遍历获取表格中的数据
#         for i in range(1, row_size + 1):
#             for j in range(1, col_size + 1):
#                 print(sheet.Cells(i, j).Value, end=" ")
#             print("")
#         print("")
#         # 下一个图表的索引号
#         idx = idx + 1
# print(idx)
# 遍历文档中所有的文字段落,判断是不是以 图+数字开头
idx = 1
for para in doc.Paragraphs:
    x = para.Range.Text.strip().replace("图 ", "图").replace("  ", " ")
    if x.startswith("图"):
        print(x)
        idx = idx + 1

# 关闭文档和Word应用
doc.Close()
docApp.Quit()