使用Python轻松获取Word文本框中的文字和图片

pip install Spire.Doc

from spire.doc import *
from spire.doc.common import *

from spire.doc import *
from spire.doc.common import *

inputFile = "/文本框.docx"
outputFile = "ExtractedText.txt"

# 加载文档
document = Document()
document.LoadFromFile(inputFile)

# 检查文档中是否包含文本框
if document.TextBoxes.Count > 0:
    with open(outputFile, 'w', encoding='utf-8') as sw:
        for i in range(document.TextBoxes.Count):
            textbox = document.TextBoxes.get_Item(i)
            # 遍历文本框内的子对象
            for j in range(textbox.Body.ChildObjects.Count):
                obj = textbox.Body.ChildObjects.get_Item(j)
                # 提取段落文字
                if obj.DocumentObjectType == DocumentObjectType.Paragraph:
                    para = obj if isinstance(obj, Paragraph) else None
                    sw.write(para.Text + "\n")
                # 提取表格中的文字
                if obj.DocumentObjectType == DocumentObjectType.Table:
                    table = obj if isinstance(obj, Table) else None
                    for row_idx in range(table.Rows.Count):
                        row = table.Rows[row_idx]
                        for cell_idx in range(row.Cells.Count):
                            cell = row.Cells[cell_idx]
                            for para_idx in range(cell.Paragraphs.Count):
                                para = cell.Paragraphs.get_Item(para_idx)
                                sw.write(para.Text + "\t")
                        sw.write("\n")

document.Close()

from spire.doc import *
from spire.doc.common import *

inputFile = "./Data/TextBoxTable.docx"
outputFile = "TableFromTextBox.txt"

# 加载文档
doc = Document()
doc.LoadFromFile(inputFile)

# 获取第一个文本框
textbox = doc.TextBoxes[0]

# 获取文本框中的第一个表格
table = textbox.Body.Tables[0] if isinstance(textbox.Body.Tables[0], Table) else None

result = ""
for i in range(table.Rows.Count):
    row = table.Rows.get_Item(i)
    for j in range(row.Cells.Count):
        cell = row.Cells.get_Item(j)
        for k in range(cell.Paragraphs.Count):
            paragraph = cell.Paragraphs.get_Item(k)
            result += paragraph.Text + "\t"
    result += "\n"

with open(outputFile, 'w', encoding='utf-8') as fp:
    fp.write(result)

doc.Close()

from spire.doc import *
from spire.doc.common import *
import os

inputFile = "./Data/TextBoxWithImages.docx"
outputDir = "ExtractedImages/"

if not os.path.exists(outputDir):
    os.makedirs(outputDir)

# 加载文档
document = Document()
document.LoadFromFile(inputFile)

image_index = 0

# 遍历所有文本框
for i in range(document.TextBoxes.Count):
    textbox = document.TextBoxes.get_Item(i)
    # 遍历文本框中的子对象
    for j in range(textbox.Body.ChildObjects.Count):
        obj = textbox.Body.ChildObjects.get_Item(j)
        # 检查是否为段落，图片通常嵌套在段落中
        if obj.DocumentObjectType == DocumentObjectType.Paragraph:
            para = obj if isinstance(obj, Paragraph) else None
            for k in range(para.ChildObjects.Count):
                child = para.ChildObjects.get_Item(k)
                if child.DocumentObjectType == DocumentObjectType.Picture:
                    picture = child if isinstance(child, DocPicture) else None
                    # 获取图片二进制数据并保存
                    image_bytes = picture.ImageBytes
                    image_path = os.path.join(outputDir, f"Image-{image_index}.png")
                    with open(image_path, 'wb') as img_file:
                        img_file.write(image_bytes)
                    image_index += 1
                    print(f"已保存图片: {image_path}")

document.Close()