Python-docx将Word文档的目录或文本框作为普通段落读入

📢作者： 小小明-代码实体

📢博客主页：https://blog.csdn.net/as604049322

📢欢迎点赞 👍 收藏 ⭐留言 📝 欢迎讨论！

昨天我们处理Word文档的自动编号，详见《Python解析Word文档的自动编号》
链接：https://xxmdmst.blog.csdn.net/article/details/139638262

今天我们处理一下Word文档的目录和文本框，例如：

如果我们按照传统方式读取，是无法读取目录和文本框的。

这时，我们只需要想办法目录和文本框中的P节点与普通P节点一起被读取即可。

查看其xml结构后知道，w:sdt是目录节点，文本框节点存在于p里面的v:textbox节点下。

可以写出如下代码：

from docx import Document
from docx.oxml import ns
from docx.text.paragraph import Paragraph

doc = Document('目录测试.docx')

ns.nsmap.update(doc.element.nsmap)
body = doc.element.body
paragraphs = []
for p in body.xpath('w:p | w:sdt/w:sdtContent/w:p | w:p//v:textbox//w:p'):
    paragraphs.append(Paragraph(p, body))

注意上面的代码使用Paragraph(p, body)封装是为了让paragraphs的结果类型与doc.paragraphs保存一致，可以看到其源码为：

如果不需要Paragraph的特殊功能，仅做基本的数据读取，也可以不封装。

然后就能将普通段落和目录内的段落以及文本框内的段落，都按顺序读取：

for paragraph in paragraphs:
    print(paragraph.text)

如果这时，我们需要将自动编号也读取进来，需要注意文本框内的段落是单独计数的。

最后我们将该功能整合到上次的代码中：

import logging
import re
import traceback
from functools import lru_cache

from docx import Document, ImagePart
from docx.oxml import register_element_cls, CT_DecimalNumber, CT_String, CT_Tbl
from docx.oxml.ns import nsmap
from docx.oxml.simpletypes import ST_DecimalNumber
from docx.oxml.xmlchemy import BaseOxmlElement, RequiredAttribute, ZeroOrMore, OneAndOnlyOne, ZeroOrOne
from docx.table import Table
from docx.text.paragraph import Paragraph


class CT_Lvl(BaseOxmlElement):
    ilvl = RequiredAttribute("w:ilvl", ST_DecimalNumber)
    start = ZeroOrOne("w:start")
    numFmt = ZeroOrOne("w:numFmt")
    lvlText = OneAndOnlyOne("w:lvlText")
    suff = ZeroOrOne("w:suff")

    def get_style(self):
        style = {
   
            "start": self.start.val if self.start is not None else 1,
            "numFmt": self.numFmt.val if self.numFmt is not None else "decimal",
            "lvlText": self.lvlText.val
        }
        if self.suff is not None:
            style["suff"] = self.suff.val
        return style


class CT_AbstractNum(BaseOxmlElement):
    abstractNumId = RequiredAttribute("w:abstractNumId", ST_DecimalNumber)
    lvl = ZeroOrMore("w:lvl")


class CT_Numbering(BaseOxmlElement):
    num = ZeroOrMore("w:num", successors=("w:numIdMacAtCleanup",))
    abstractNum = ZeroOrMore("w:abstractNum", successors=("w:numIdMacAtCleanup",))


register_element_cls("w:numbering", CT_Numbering)
register_element_cls("w:abstractNum", CT_AbstractNum)
register_element_cls("w:lvl", CT_Lvl)
register_element_cls("w:start", CT_DecimalNumber)
register_element_cls("w:numFmt", CT_String)
register_element_cls("w:lvlText", CT_String)
register_element_cls("w:suff", CT_String)


class WithNumberDocxReader:
    ideographTraditional = "甲乙丙丁戊己庚辛壬癸"
    ideographZodiac = "子丑寅卯辰巳午未申酉戌亥"

    def __init__(self, docx, gap_text="\t", include_table_content=True, style_optimization=False, debug=False):
        self.preview_cache = {
   }
        self.debug = debug
        self.style_optimization = style_optimization
        self.heading_level_numpr = {
   }
        self.style_level = {
   }
        self.style2numId = {
   }
        self.level2numpr = {
   }
        self.parts = []
        self.docx = Document(docx)
        self.include_table_content = include_table_content
        nsmap.update(self.docx.element.nsmap)
        self.numId2style = {
   }
        self.get_style_data()
        self.key2level = {
   }
        self.gap_text = gap_text
        self.cnt = {
   }
        self.cache = {
   }
        self.result = []

    def replace_number2text(self):
        body = self.docx.element.body
        for p in body.xpath('w:p | w:sdt/w:sdtContent/w:p | w:p//v:textbox//w:p | w:tbl//w:p'):
            paragraph = Paragraph(p, body)
            text = paragraph.text.strip()
            if not text:
                continue
            number_text = self.get_number_text(paragraph)
            line = number_text + text
            paragraph.text = line
            paragraph.paragraph_format.element.pPr._remove_numPr()
        for style in self.docx.part.styles.element.style_lst:
            if style.pPr is not None and style.pPr.numPr is not None:
                style.pPr._remove_numPr()

    def save(self, file):
        self.docx.save(file)

    @property
    def texts(self):
        if self.result:
            return self.result.copy()
        self.clear()
        for paragraph in self.tablesAndParagraphs:
            if isinstance(paragraph, Table):
                for row in paragraph.rows:
                    line = "\t".join([cell.text.strip() for cell in row.cells])
                    if line:
                        self.result.append(line)
            else:
                number_text = self.get_number_text(paragraph)
                line = number_text + paragraph.text.strip()
                if line:
                    self.result.append(line)
        if self.debug:
            print("---------------debug-------------------")
        return self.result.copy()

    def clear(self):
        self.result.clear()
        self.cnt