Windows系统下python实现从Word中读取全部内容包含图片，并放入新的Word中

要实现从Word中读取全部内容包含图片，并放入新的Word中，需要使用Python的docx库和python-docx-image库。下面是一个基本的示例代码：

from docx import Document
from docx.shared import Inches
from docx.enum.text import WD_ALIGN_PARAGRAPH
from docx.oxml import OxmlElement
from docx.oxml.ns import qn
from docx.image.image import Image
from docx.image.exceptions import UnrecognizedImageError
from docx.text.paragraph import Paragraph

def add_picture(paragraph, picture_path):
    try:
        paragraph.alignment = WD_ALIGN_PARAGRAPH.CENTER
        run = paragraph.add_run()
        element = run._element
        element.tag = 'w:pict'
        element.set(qn('xmlns:wp'), 'http://schemas.openxmlformats.org/drawingml/2006/wordprocessingDrawing')
        element.set(qn('xmlns:r'), 'http://schemas.openxmlformats.org/officeDocument/2006/relationships')
        img = Image(picture_path)
        width, height = img.width, img.height
        inline = element.makeelement(qn('wp:inline'))
        inline.set(qn('distT'), "0")
        inline.set(qn('distB'), "0")
        inline.set(qn('distL'), "0")
        inline.set(qn('distR'), "0")
        extent = inline.makeelement(qn('wp:extent'))
        extent.set(qn('cx'), str(width * 9525))
        extent.set(qn('cy'), str(height * 9525))
        inline.append(extent)
        docpr = inline.makeelement(qn('wp:docPr'))
        docpr.set(qn('id'), "1")
        docpr.set(qn('name'), "Image")
        docpr.set(qn('descr'), "Image")
        inline.append(docpr)
        graphic = inline.makeelement(qn('a:graphic'))
        inline.append(graphic)
        graphic.set(qn('xmlns:a'), 'http://schemas.openxmlformats.org/drawingml/2006/main')
        graphicwrap = graphic.makeelement(qn('a:graphicData'))
        graphicwrap.set(qn('uri'), 'http://schemas.openxmlformats.org/drawingml/2006/picture')
        graphic.append(graphicwrap)
        pic = graphicwrap.makeelement(qn('pic:pic'))
        graphicwrap.append(pic)
        nvpicpr = pic.makeelement(qn('pic:nvPicPr'))
        pic.append(nvpicpr)
        cnvpr = nvpicpr.makeelement(qn('pic:cNvPr'))
        cnvpr.set(qn('id'), "0")
        cnvpr.set(qn('name'), "Image")
        cnvpr.set(qn('descr'), "Image")
        nvpicpr.append(cnvpr)
        nvpicpr.append(inline.makeelement(qn('pic:cNvPicPr')))
        blip = pic.makeelement(qn('pic:blipFill'))
        pic.append(blip)
        blip.set(qn('dpi'), "96, 96")
        blip.set(qn('rotWithShape'), "1")
        blipfill = blip.makeelement(qn('a:blip'))
        blip.append(blipfill)
        blipfill.set(qn('r:embed'), inline.makeelement(qn('a:extLst')))
        blipfill.set(qn('cstate'), "print")
        stretch = blip.makeelement(qn('a:stretch'))
        blip.append(stretch)
        fillrect = stretch.makeelement(qn('a:fillRect'))
        stretch.append(fillrect)
        run._r.append(element)
    except UnrecognizedImageError:
        pass

# 读取原始文档
doc = Document('original.docx')

# 创建新文档
new_doc = Document()

for element in doc.element.body:
    if isinstance(element, Paragraph):
        # 复制段落
        new_paragraph = new_doc.add_paragraph(element.text, element.style)

        # 处理段落中的图片
        for run in element.runs:
            if run._element.tag.endswith('}r'):
                for child in run._element:
                    if child.tag.endswith('}pict'):
                        for pic_child in child:
                            if pic_child.tag.endswith('}blip'):
                                picture_path = doc.part.related_parts[pic_child.attrib['{http://schemas.openxmlformats.org/officeDocument/2006/relationships}embed']].partname[1:]
                                add_picture(new_paragraph, picture_path)

new_doc.save('new.docx')

这个示例代码首先读取了原始文档，然后创建一个新的文档。它遍历原始文档中的所有段落，并复制它们到新文档中。在复制段落时，它会处理其中的图片，将它们添加到新文档中。处理图片时，它使用了add_picture函数，该函数会将图片添加到指定的段落中。