可能遇到过： docxtpl 模板占位的元素带入了 html 元素，wps能打开但是word打开报错的情况
这种情况下需要先把 html 转为 RichText 元素，然后再插入
但是现在按照这个思路去找，可能完全找不到相关的解决方案
转变了一下思路，用关键词 markdown to richtext 搜索了一下，找到了一个相关的 issue jhpyle/docassemble#72
虽说里面没有提及解决方案，但是他们提到是通过 markdown -> html -> richtext 来解决的，并且该项目中已经实现了相关功能
扣了出来实现了一下，见下面的 html_richtext.py
调用示例见 main.py
| import re | |
| import string | |
| from jinja2.runtime import Undefined | |
| import docx.opc.constants | |
| from bs4 import NavigableString, Tag | |
| from docxtpl import RichText | |
| # ref: https://github.com/jhpyle/docassemble/issues/72 | |
| # ref: https://github.com/jhpyle/docassemble/blob/c16c786c65186399584c9a30d0e3d7dcb3acb056/docassemble_base/docassemble/base/file_docx.py#LL645C21-L645C21 | |
| list_types = ['1', 'A', 'a', 'I', 'i'] | |
| def ensure_definition(*pargs, **kwargs): | |
| for val in pargs: | |
| if isinstance(val, Undefined): | |
| str(val) | |
| for val in kwargs.values(): | |
| if isinstance(val, Undefined): | |
| str(val) | |
| def roman(num, case=None): | |
| """Given an index between 0 and 3999, returns a roman numeral between 1 and 4000.""" | |
| ensure_definition(num, case) | |
| if case is None: | |
| case = 'upper' | |
| num = num + 1 | |
| if not isinstance(num, int): | |
| raise TypeError("expected integer, got %s" % type(num)) | |
| if not 0 < num < 4000: | |
| raise ValueError("Argument must be between 1 and 3999") | |
| ints = (1000, 900, 500, 400, 100, 90, 50, 40, 10, 9, 5, 4, 1) | |
| nums = ('M', 'CM', 'D', 'CD', 'C', 'XC', 'L', 'XL', 'X', 'IX', 'V', 'IV', 'I') | |
| result = "" | |
| for indexno, the_int in enumerate(ints): | |
| count = int(num / the_int) | |
| result += nums[indexno] * count | |
| num -= the_int * count | |
| if case == 'lower': | |
| return result.lower() | |
| return result | |
| def Alpha(number): | |
| multiplier = int((number - 1) / 26) | |
| indexno = (number - 1) % 26 | |
| return string.ascii_uppercase[indexno] * (multiplier + 1) | |
| def alpha(number): | |
| multiplier = int((number - 1) / 26) | |
| indexno = (number - 1) % 26 | |
| return string.ascii_lowercase[indexno] * (multiplier + 1) | |
| def Roman_Numeral(number): | |
| return roman((number - 1) % 4000, case='upper') | |
| def roman_numeral(number): | |
| return roman((number - 1) % 4000, case='lower') | |
| class InlineSoupParser: | |
| def __init__(self, tpl): | |
| self.runs = [RichText('')] | |
| self.run = self.runs[-1] | |
| self.bold = False | |
| self.italic = False | |
| self.underline = False | |
| self.indentation = 0 | |
| self.style = 'p' | |
| self.strike = False | |
| self.size = None | |
| self.charstyle = None | |
| self.color = None | |
| self.tpl = tpl | |
| self.at_start = True | |
| self.list_number = 1 | |
| self.list_type = list_types[-1] | |
| def new_paragraph(self): | |
| if self.at_start: | |
| self.at_start = False | |
| else: | |
| self.run.add("\n", italic=self.italic, bold=self.bold, underline=self.underline, strike=self.strike, size=self.size, style=self.charstyle, color=self.color) | |
| if self.indentation: | |
| self.run.add("\t" * self.indentation) | |
| if self.style == 'ul': | |
| self.run.add("•\t") | |
| if self.style == 'ol1': | |
| self.run.add(str(self.list_number) + ".\t") | |
| self.list_number += 1 | |
| elif self.style == 'olA': | |
| self.run.add(Alpha(self.list_number) + ".\t") | |
| self.list_number += 1 | |
| elif self.style == 'ola': | |
| self.run.add(alpha(self.list_number) + ".\t") | |
| self.list_number += 1 | |
| elif self.style == 'olI': | |
| self.run.add(Roman_Numeral(self.list_number) + ".\t") | |
| self.list_number += 1 | |
| elif self.style == 'oli': | |
| self.run.add(roman_numeral(self.list_number) + ".\t") | |
| self.list_number += 1 | |
| # else: | |
| # self.list_number = 1 | |
| def __str__(self): | |
| output = '' | |
| for run in self.runs: | |
| output += str(run) | |
| return output | |
| def start_link(self, url): | |
| ref = self.tpl.docx._part.relate_to(url, docx.opc.constants.RELATIONSHIP_TYPE.HYPERLINK, is_external=True) | |
| self.runs.append('<w:hyperlink r:id="%s">' % (ref, )) | |
| self.new_run() | |
| def end_link(self): | |
| self.runs.append('</w:hyperlink>') | |
| self.new_run() | |
| def new_run(self): | |
| self.runs.append(RichText('')) | |
| self.run = self.runs[-1] | |
| def traverse(self, elem): | |
| for part in elem.contents: | |
| if isinstance(part, NavigableString): | |
| self.run.add(str(part), italic=self.italic, bold=self.bold, underline=self.underline, strike=self.strike, size=self.size, style=self.charstyle, color=self.color) | |
| elif isinstance(part, Tag): | |
| if part.name in ('p', 'blockquote'): | |
| self.new_paragraph() | |
| self.traverse(part) | |
| elif part.name == 'li': | |
| self.new_paragraph() | |
| self.traverse(part) | |
| elif part.name == 'ul': | |
| oldstyle = self.style | |
| self.style = 'ul' | |
| self.indentation += 1 | |
| self.traverse(part) | |
| self.indentation -= 1 | |
| self.style = oldstyle | |
| elif part.name == 'ol': | |
| oldstyle = self.style | |
| oldlistnumber = self.list_number | |
| oldlisttype = self.list_type | |
| if part.get('type', None) in list_types: | |
| self.list_type = part['type'] | |
| else: | |
| self.list_type = list_types[(list_types.index(self.list_type) + 1) % 5] | |
| try: | |
| self.list_number = int(part.get('start', 1)) | |
| except: | |
| self.list_number = 1 | |
| self.style = 'ol' + self.list_type | |
| self.indentation += 1 | |
| self.traverse(part) | |
| self.indentation -= 1 | |
| self.list_type = oldlisttype | |
| self.list_number = oldlistnumber | |
| self.style = oldstyle | |
| elif part.name == 'strong': | |
| self.bold = True | |
| self.traverse(part) | |
| self.bold = False | |
| elif part.name == 'em': | |
| self.italic = True | |
| self.traverse(part) | |
| self.italic = False | |
| elif part.name == 'strike': | |
| self.strike = True | |
| self.traverse(part) | |
| self.strike = False | |
| elif part.name == 'u': | |
| self.underline = True | |
| self.traverse(part) | |
| self.underline = False | |
| elif re.match(r'h[1-6]', part.name): | |
| oldsize = self.size | |
| self.size = 60 - ((int(part.name[1]) - 1) * 10) | |
| self.bold = True | |
| self.traverse(part) | |
| self.bold = False | |
| self.size = oldsize | |
| elif part.name == 'a': | |
| self.start_link(part['href']) | |
| if self.tpl.da_hyperlink_style: | |
| self.charstyle = self.tpl.da_hyperlink_style | |
| else: | |
| self.underline = True | |
| self.color = '#0000ff' | |
| self.traverse(part) | |
| if self.tpl.da_hyperlink_style: | |
| self.charstyle = None | |
| else: | |
| self.underline = False | |
| self.color = None | |
| self.end_link() | |
| elif part.name == 'br': | |
| self.run.add("\n", italic=self.italic, bold=self.bold, underline=self.underline, strike=self.strike, size=self.size, style=self.charstyle, color=self.color) | |
| else: | |
| print("Encountered a " + part.__class__.__name__) | |
| def html_to_richtext(tpl, source_code): | |
| # ref: https://github.com/jhpyle/docassemble/blob/c16c786c65186399584c9a30d0e3d7dcb3acb056/docassemble_base/docassemble/base/file_docx.py#LL645C21-L645C21 | |
| source_code = re.sub(r"\n", ' ', source_code) | |
| source_code = re.sub(r">\s+<", '><', source_code) | |
| soup = BeautifulSoup('<html>' + source_code + '</html>', 'html.parser') | |
| parser = InlineSoupParser(tpl) | |
| for elem in soup.find_all(recursive=False): | |
| parser.traverse(elem) | |
| output = str(parser) | |
| return output |
| from docxtpl import DocxTemplate | |
| if __name__ == '__main__': | |
| tpl = DocxTemplate('tpl.docx') | |
| content html_to_richtext(tpl, '在 HTTP 事务的上下文中，基本访问身份验证是 HTTP 用户代理在发出请求时提供用户名和密码的方法。<br/><br/>通过 HTTP 连接使用基本身份验证保护此目录。使用基本身份验证时，用户凭据将以纯文本形式发送，并且由于未使用 HTTPS，它们很容易受到数据包嗅探的影响。') | |
| context = { | |
| 'content': 'content' | |
| } | |
| tpl.render(context) | |
| tpl.save('output.docx') | |
| # 注意，tpl.docx 中的占位符类似于 `{{r content }}` |