diff --git a/docutranslate/exporter/md/md2html_exporter.py b/docutranslate/exporter/md/md2html_exporter.py
index af42ddb..76770c3 100644
--- a/docutranslate/exporter/md/md2html_exporter.py
+++ b/docutranslate/exporter/md/md2html_exporter.py
@@ -1,22 +1,22 @@
from dataclasses import dataclass
-
import jinja2
import markdown
-
from docutranslate.exporter.md.base import MDExporter, MDExporterConfig
from docutranslate.ir.document import Document
from docutranslate.ir.markdown_document import MarkdownDocument
from docutranslate.utils.resource_utils import resource_path
+
@dataclass
class MD2HTMLExporterConfig(MDExporterConfig):
cdn: bool = True
+
class MD2HTMLExporter(MDExporter):
def __init__(self, config: MD2HTMLExporterConfig = None):
config = config or MD2HTMLExporterConfig()
super().__init__(config=config)
- self.cdn=config.cdn
+ self.cdn = config.cdn
def export(self, document: MarkdownDocument) -> Document:
cdn = self.cdn
@@ -26,41 +26,32 @@ class MD2HTMLExporter(MDExporter):
katex_css = f'' if not cdn else r""""""
katex_js = f'' if not cdn else r""""""
auto_render = f'' if not cdn else r""""""
- # language=javascript
+
+ # 修改 JavaScript 渲染配置，增加更多选项
render_math_in_element = r"""
- """ if cdn else r"""
- """
+ """
+
mermaid = f''
- # 使用 python-markdown 和 pymdown-extensions
- # Arithmatex 扩展专门用于处理 KaTeX/MathJax 公式
- # 它能智能识别 $...$, $$...$$, \(...\), \[...\] 等，并保护它们不受干扰
+ # 修改扩展配置
extensions = [
'markdown.extensions.tables',
'pymdownx.arithmatex',
@@ -69,14 +60,20 @@ class MD2HTMLExporter(MDExporter):
extension_configs = {
'pymdownx.arithmatex': {
- 'generic': True
+ 'generic': True,
+ 'block_tag': 'div',
+ 'inline_tag': 'span',
+ 'block_syntax': ['dollar', 'square'],
+ 'inline_syntax': ['dollar', 'round'],
+ 'tex_inline_wrap': ['\\(', '\\)'],
+ 'tex_block_wrap': ['\\[', '\\]'],
+ 'smart_dollar': True
},
'pymdownx.superfences': {
'custom_fences': [
{
'name': 'mermaid',
'class': 'mermaid',
- # 这个 format 函数确保输出的 HTML 结构是 Mermaid.js 期望的
'format': lambda source, language, css_class, options, md,
**kwargs: f'{source}'
}
@@ -84,9 +81,21 @@ class MD2HTMLExporter(MDExporter):
}
}
- # 注意：这里不再需要 .replace("\\", "\\\\")
+ # 预处理 markdown 内容，确保数学公式周围有正确的空行
+ content = document.content.decode()
+
+ # 处理 $$ 块公式，确保前后有空行
+ import re
+ # 匹配 $$ 块公式
+ def fix_block_math(match):
+ formula = match.group(1)
+ return f'\n\n$$\n{formula}\n$$\n\n'
+
+ # 使用正则表达式修复块公式格式
+ content = re.sub(r'\$\$\s*\n?(.*?)\n?\s*\$\$', fix_block_math, content, flags=re.DOTALL)
+
html_content = markdown.markdown(
- document.content.decode(),
+ content,
extensions=extensions,
extension_configs=extension_configs
)
@@ -97,17 +106,18 @@ class MD2HTMLExporter(MDExporter):
katexCss=katex_css,
katexJs=katex_js,
autoRender=auto_render,
- markdown=html_content, # 使用新的 html_content
+ markdown=html_content,
renderMathInElement=render_math_in_element,
mermaid=mermaid,
)
return Document.from_bytes(content=render.encode("utf-8"), suffix=".html", stem=document.stem)
+
if __name__ == '__main__':
from pathlib import Path
- # d=Document.from_path(r"C:\Users\jxgm\Desktop\A_Survey_on_Decentralized_Identifiers_and_Verifiable_Credentials_translated.md")
- d=Document.from_path(r"C:\Users\jxgm\Desktop\mcp文件夹\学习笔记\互联网认证授权机制\互联网认证授权机制.md")
- exporter=MD2HTMLExporter()
- d1=exporter.export(d)
- path=Path(r"C:\Users\jxgm\Desktop\a.html")
+
+ d = Document.from_path(r"C:\Users\jxgm\Desktop\a2f9907d-6d49-4e87-9075-126218336b1e_origin_translated.md")
+ exporter = MD2HTMLExporter()
+ d1 = exporter.export(d)
+ path = Path(r"C:\Users\jxgm\Desktop\a.html")
path.write_bytes(d1.content)
\ No newline at end of file