从基础到进阶详解Python处理Word文档的完全指南

pip install python-docx docxtpl docxcompose lxml  

uv add python-docx docxtpl docxcompose lxml  

from docx import document  
doc = document()  
doc.add_paragraph("python-docx是一个用于创建")  
doc.save("文件1.docx")  

from docx.oxml.ns import qn  
def set_chinese_font(run, zh_font_name="宋体", en_font_name="times new roman"):  
    run.font.name = en_font_name  
    run._element.rpr.rfonts.set(qn("w:eastasia"), zh_font_name)  
doc = document()  
paragraph = doc.add_paragraph()  
run = paragraph.add_run('这是一段设置了中文字体的文本。')  
set_chinese_font(run)  
doc.save("文件1.docx")  

doc = document('example.docx')  

# 遍历段落  
for para in doc.paragraphs[:3]:  
    print(para)  
    print(para.text)  
    print()  
# 遍历表格  
for table in doc.tables:  
    for row in table.rows:  
        for cell in row.cells:  
            print(cell.text)  

doc.add_heading("1.1 transformer整体工作流程", 2)  
doc.add_heading("transformer整体架构", 3)  

text = """transformer 模型由编码器（encoder）和解码器（decoder）组成。..."""  
paragraph1 = doc.add_paragraph(text)  

paragraph_format = paragraph1.paragraph_format  
paragraph_format.first_line_indent = 0  
paragraph_format.element.ppr.ind.set(qn("w:firstlinechars"), '200')  

para_format.first_line_indent = pt(10)  

from docx.enum.text import wd_paragraph_alignment  
paragraph1.alignment = wd_paragraph_alignment.left  

p = paragraph1._element  
p.getparent().remove(p)  

# 将文本按换行符分割成多个段落  
for one_paragraph_text in text.split("\n"):  
    temp_paragraph = doc.add_paragraph(one_paragraph_text)  
    paragraph_format = temp_paragraph.paragraph_format  
    paragraph_format.first_line_indent = 0  
    paragraph_format.element.ppr.ind.set(qn("w:firstlinechars"), "200")  

from docx.shared import pt  
para_format = temp_paragraph.paragraph_format  
para_format.line_spacing = pt(18)  # 行间距（固定值）  
para_format.space_before = pt(3)  # 段前距离  
para_format.space_after = pt(0)  # 段后距离  
para_format.right_indent = pt(20)  # 右侧缩进  
para_format.left_indent = pt(0)  # 左侧缩进  

from docx.shared import rgbcolor, pt  
# 加粗文本  
temp_paragraph.add_run('加粗文本').bold = true  
# 红色斜体文本  
run = temp_paragraph.add_run('红色斜体文本')  
run.font.color.rgb = rgbcolor(255,0,0)  # 设置红色  
run.font.size = pt(14)  # 字号14磅  
run.bold = true  # 加粗  
run.italic = true  # 斜体  
run.underline = true  # 下划线  
# 下标和上标  
run2 = temp_paragraph.add_run("1")  
run2.font.subscript = true  # 下标  
run3 = temp_paragraph.add_run("2")  
run3.font.superscript = true  # 上标  

table = doc.add_table(rows=4, cols=5)  
table.style = "grid table 1 light"  # 应用预定义样式  

# 方式1：直接指定单元格  
cell = table.cell(0, 1)  
cell.text = "parrot, possibly dead"  
# 方式2：通过行获取单元格  
row = table.rows[1]  
cells = row.cells  
cells[0].text = "foo bar to you."  
cells[1].text = "and a hearty foo bar to you too sir!"  

from docx.enum.style import wd_style_type  
styles = doc.styles  
for s in styles:  
    if s.type == wd_style_type.table:  
        print(s.name)  

# 增加一行  
row = table.add_row()  
# 删除一行  
def remove_row(table, row):  
    tbl = table._tbl  
    tr = row._tr  
    tbl.remove(tr)  
row = table.rows[len(table.rows) - 1]  
remove_row(table, row)  

# 方式1：一行一行添加  
items = (  
    (7, "1024", "plush kittens"),  
    (3, "2042", "furbees"),  
    (1, "1288", "french poodle collars, deluxe"),  
)  
for item in items:  
    cells = table.add_row().cells  
    cells[0].text = str(item[0])  
    cells[1].text = item[1]  
    cells[2].text = item[2]  
# 方式2：批量填充  
for row in table.rows:  
    for cell in row.cells:  
        cell.text = "数据单元"  

table.cell(0, 0).merge(table.cell(1, 1))  # 跨行列合并  

# 表格宽度自适应  
table.autofit = true  
# 指定行高  
from docx.shared import cm  
table.rows[0].height = cm(0.93)  
# 修改表格字体大小  
table.style.font.size = pt(15)  
# 设置单元格对齐  
from docx.enum.table import wd_align_vertical  
cell = table.cell(0, 0)  
cell.paragraphs[0].paragraph_format.alignment = wd_align_paragraph.center  
cell.vertical_alignment = wd_align_vertical.center  
# 复制表格  
from copy import deepcopy  
table_copy = deepcopy(doc.tables[0])  
para1 = doc.add_paragraph()  
para1._p.addnext(table_copy._element)  

from io import bytesio  
import base64  
# 普通插入  
doc.add_picture('图片1.png')  
doc.add_picture('图片2.png', width=inches(2.5), height=inches(2))  
# 使用base64插入  
picture2_base64 = open("图片2base64.txt").read()  
img2_buf = base64.b64decode(picture2_base64)  
doc.add_picture(bytesio(img2_buf))  
# 并排放图  
run = doc.add_paragraph().add_run()  
run.add_picture("图片1.png", width=inches(2.5), height=inches(2))  
run.add_picture("图片1.png", width=inches(2.5), height=inches(2))  

doc.add_page_break()  

# 修改已有样式  
doc.styles["normal"].font.size = pt(14)  
doc.styles['normal'].font.name = 'arial'  
doc.styles['normal']._element.rpr.rfonts.set(qn('w:eastasia'), '楷体')  
# 创建自定义段落样式  
from docx.enum.style import wd_style_type  
userstyle1 = doc.styles.add_style('userstyle1', wd_style_type.paragraph)  
userstyle1.font.size = pt(40)  
userstyle1.font.color.rgb = rgbcolor(0xff, 0xde, 0x00)  
userstyle1.paragraph_format.alignment = wd_align_paragraph.center  
userstyle1.font.name = '微软雅黑'  
userstyle1._element.rpr.rfonts.set(qn('w:eastasia'), '微软雅黑')  
# 使用自定义样式  
doc.add_paragraph('自定义段落样式', style=userstyle1)  

from docxtpl import docxtemplate, inlineimage, richtext  
tpl = docxtemplate("docxexample.docx")  
text = """transformer 模型由编码器（encoder）和解码器（decoder）组成..."""  
picture1 = inlineimage(tpl, image_descriptor="图片1.png")  
# 准备数据  
paragraphs1 = [  
    "步骤1：输入表示（input representation）",  
    "步骤2：编码器处理（encoder processing）",  
    "步骤3：解码器处理（decoder processing）",  
]  
paragraphs2 = [  
    {"step": 1, "text": "输入向量（词嵌入+位置编码）进入编码器层。"},  
    {"step": 2, "text": "自注意力子层。"},  
    {"step": 3, "text": "前馈网络子层。"},  
]  
table = [  
    {"character": "并行计算", "description": "编码器可并行处理整个序列（与rnn不同）"},  
    {"character": "自注意力", "description": "每个词直接关联所有词，捕获长距离依赖"},  
    {"character": "位置编码", "description": "为无顺序的注意力机制注入位置信息"},  
]  
alerts = [  
    {  
        "date": "2015-03-10",  
        "desc": richtext("very critical alert", color="ff0000", bold=true),  
        "type": "critical",  
        "bg": "ff0000",  
    },  
    # ... 其他数据  
]  
# 渲染模板  
context = {  
    "title": "transformer",  
    "text_body": text,  
    "picture1": picture1,  
    "picture2": picture2,  
    "paragraphs1": paragraphs1,  
    "paragraphs2": paragraphs2,  
    "runs": paragraphs1,  
    "display_paragraph": true,  
    "table1": table,  
    "table2": table,  
    "alerts": alerts,  
}  
tpl.render(context)  
tpl.save("文件3.docx")  

from docx.oxml import oxmlelement  
from docx.oxml.ns import qn  
def set_cell_border(cell, **kwargs):  
    tc = cell._tc  
    tcpr = tc.get_or_add_tcpr()  
    tcborders = tcpr.first_child_found_in("w:tcborders")  
    if tcborders is none:  
        tcborders = oxmlelement("w:tcborders")  
        tcpr.append(tcborders)  
      
    for edge in ("left", "top", "right", "bottom", "insideh", "insidev"):  
        edge_data = kwargs.get(edge)  
        if edge_data:  
            tag = "w:{}".format(edge)  
            element = tcborders.find(qn(tag))  
            if element is none:  
                element = oxmlelement(tag)  
                tcborders.append(element)  
              
            for key in ["sz", "val", "color", "space", "shadow"]:  
                if key in edge_data:  
                    element.set(qn("w:{}".format(key)), str(edge_data[key]))  
# 使用示例  
set_cell_border(  
    table.cell(0, 0),  
    top={"sz": 4, "val": "single", "color": "#000000", "space": "0"},  
    bottom={"sz": 4, "val": "single", "color": "#000000", "space": "0"},  
    left={"sz": 4, "val": "single", "color": "#000000", "space": "0"},  
    right={"sz": 4, "val": "single", "color": "#000000", "space": "0"},  
)  

def add_hyperlink(paragraph, url, text):  
    part = paragraph.part  
    r_id = part.relate_to(  
        url,  
        "http://schemas.openxmlformats.org/officedocument/2006/relationships/hyperlink",  
        is_external=true,  
    )  
    hyperlink = oxmlelement("w:hyperlink")  
    hyperlink.set(qn("r:id"), r_id)  
      
    run = oxmlelement("w:r")  
    run_text = oxmlelement("w:t")  
    run_text.text = text  
    run.append(run_text)  
    hyperlink.append(run)  
      
    paragraph._p.append(hyperlink)  
p = doc.add_paragraph("点击访问： ")  
add_hyperlink(p, "https://www.baidu.com", "示例链接")  

import zipfile  
from xml.etree.elementtree import fromstring  
def extract_images(docx_path, output_dir):  
    with zipfile.zipfile(docx_path) as z:  
        try:  
            doc_rels = z.read('word/_rels/document.xml.rels').decode('utf-8')  
        except keyerror:  
            return []  
          
        root = fromstring(doc_rels)  
        rels = []  
        for child in root:  
            if 'type' in child.attrib and child.attrib['type'] == rt.image:  
                rels.append((child.attrib['id'], child.attrib['target']))  
          
        images = []  
        for rel_id, target in rels:  
            try:  
                image_data = z.read('word/' + target)  
                image_name = target.split('/')[-1]  
                with open(f"{output_dir}/{image_name}", 'wb') as f:  
                    f.write(image_data)  
                images.append(image_name)  
            except keyerror:  
                continue  
        return images  
print(extract_images("transformer原理纯享版.docx", "pictures"))  

# 插入“衬于文字下方”的浮动图片  
# 如将 behinddoc="1" 改成0就是“浮于文字上方”了  
# refer to docx.oxml.shape.ct_inline  
class ct_anchor(baseoxmlelement):  
    """  
    ``<w:anchor>`` element, container for a floating image.  
    """  
    extent = oneandonlyone('wp:extent')  
    docpr = oneandonlyone('wp:docpr')  
    graphic = oneandonlyone('a:graphic')  
    @classmethod  
    def new(cls, cx, cy, shape_id, pic, pos_x, pos_y):  
        """  
        return a new ``<wp:anchor>`` element populated with the values passed  
        as parameters.  
        """  
        anchor = parse_xml(cls._anchor_xml(pos_x, pos_y))  
        anchor.extent.cx = cx  
        anchor.extent.cy = cy  
        anchor.docpr.id = shape_id  
        anchor.docpr.name = 'picture %d' % shape_id  
        anchor.graphic.graphicdata.uri = (  
            'http://schemas.openxmlformats.org/drawingml/2006/picture'  
        )  
        anchor.graphic.graphicdata._insert_pic(pic)  
        return anchor  
    @classmethod  
    def new_pic_anchor(cls, shape_id, rid, filename, cx, cy, pos_x, pos_y):  
        """  
        return a new `wp:anchor` element containing the `pic:pic` element  
        specified by the argument values.  
        """  
        pic_id = 0  # word doesn't seem to use this, but does not omit it  
        pic = ct_picture.new(pic_id, filename, rid, cx, cy)  
        anchor = cls.new(cx, cy, shape_id, pic, pos_x, pos_y)  
        anchor.graphic.graphicdata._insert_pic(pic)  
        return anchor  
    @classmethod  
    def _anchor_xml(cls, pos_x, pos_y):  
        return (  
            '<wp:anchor distt="0" distb="0" distl="0" distr="0" simplepos="0" relativeheight="0" \n'  
            '           behinddoc="1" locked="0" layoutincell="1" allowoverlap="1" \n'  
            '           %s>\n'  
            '  <wp:simplepos x="0" y="0"/>\n'  
            '  <wp:positionh relativefrom="page">\n'  
            '    <wp:posoffset>%d</wp:posoffset>\n'  
            '  </wp:positionh>\n'  
            '  <wp:positionv relativefrom="page">\n'  
            '    <wp:posoffset>%d</wp:posoffset>\n'  
            '  </wp:positionv>\n'                      
            '  <wp:extent cx="914400" cy="914400"/>\n'  
            '  <wp:wrapnone/>\n'  
            '  <wp:docpr id="666" name="unnamed"/>\n'  
            '  <wp:cnvgraphicframepr>\n'  
            '    <a:graphicframelocks nochangeaspect="1"/>\n'  
            '  </wp:cnvgraphicframepr>\n'  
            '  <a:graphic>\n'  
            '    <a:graphicdata uri="uri not set"/>\n'  
            '  </a:graphic>\n'  
            '</wp:anchor>' % ( nsdecls('wp', 'a', 'pic', 'r'), int(pos_x), int(pos_y) )  
        )  
# refer to docx.parts.story.basestorypart.new_pic_inline  
def new_pic_anchor(part, image_descriptor, width, height, pos_x, pos_y):  
    """return a newly-created `w:anchor` element.  
    the element contains the image specified by *image_descriptor* and is scaled  
    based on the values of *width* and *height*.  
    """  
    rid, image = part.get_or_add_image(image_descriptor)  
    cx, cy = image.scaled_dimensions(width, height)  
    shape_id, filename = part.next_id, image.filename      
    return ct_anchor.new_pic_anchor(shape_id, rid, filename, cx, cy, pos_x, pos_y)  
# refer to docx.text.run.add_picture  
def add_float_picture(p, image_path_or_stream, width=none, height=none, pos_x=0, pos_y=0):  
    """add float picture at fixed position `pos_x` and `pos_y` to the top-left point of page.  
    """  
    run = p.add_run()  
    anchor = new_pic_anchor(run.part, image_path_or_stream, width, height, pos_x, pos_y)  
    run._r.add_drawing(anchor)  
# refer to docx.oxml.__init__.py  
register_element_cls('wp:anchor', ct_anchor)  
document = document()  
# add a floating picture  
p = document.add_paragraph()  
add_float_picture(p, '图片1.png')  
# add text  
p.add_run('hello world '*50)  
document.save('文件2.docx')  
# https://www.cnblogs.com/dancesir/p/17788854.html  

# 分2栏  
section = doc.sections[0]  
sectpr = section._sectpr  
cols = sectpr.xpath('./w:cols')[0]  
cols.set(qn('w:num'),'2')  

# 普通页眉  
doc = document('transformer原理纯享版.docx')  
doc.sections[0].header.paragraphs[0].text = "这是第1节页眉"  
# 分奇偶设置页眉  
doc.settings.odd_and_even_pages_header_footer = true  
doc.sections[0].even_page_header.paragraphs[0].text = "这是偶数页页眉"  
doc.sections[0].header.paragraphs[0].text = "这是奇数页页眉"  
# 设置首页页眉  
doc.sections[0].different_first_page_header_footer = true  
doc.sections[0].first_page_header.paragraphs[0].text = "这是首页页眉"  

# 插入目录（不会更新域）  
paragraph = doc.paragraphs[0].insert_paragraph_before()  
run = paragraph.add_run()  
fldchar = oxmlelement('w:fldchar')  
fldchar.set(qn('w:fldchartype'), 'begin')  
instrtext = oxmlelement('w:instrtext')  
instrtext.set(qn('xml:space'), 'preserve')  
instrtext.text = r'toc \o "1-3" \h \z \u'  
fldchar2 = oxmlelement('w:fldchar')  
fldchar2.set(qn('w:fldchartype'), 'separate')  
fldchar3 = oxmlelement('w:t')  
fldchar3.text = "right-click to update field."  
fldchar2.append(fldchar3)  
fldchar4 = oxmlelement('w:fldchar')  
fldchar4.set(qn('w:fldchartype'), 'end')  
r_element = run._r  
r_element.append(fldchar)  
r_element.append(instrtext)  
r_element.append(fldchar2)  
r_element.append(fldchar4)  
# 自动更新目录  
import lxml  
name_space = "{http://schemas.openxmlformats.org/wordprocessingml/2006/main}"  
update_name_space = "%supdatefields" % name_space  
val_name_space = "%sval" % name_space  
try:  
    element_update_field_obj = lxml.etree.subelement(doc.settings.element, update_name_space)  
    element_update_field_obj.set(val_name_space, "true")  
except exception as e:  
    del e  

from docxcompose.composer import composer  
master = document("文件1.docx")  
composer = composer(master)  
doc1 = document("文件2.docx")  
composer.append(doc1)  
doc2 = document("文件3.docx")  
composer.append(doc2)  
composer.save("combined.docx")