使用python转换markdown to html

起因

有很多编辑器可以直接将markdown转换成html，为什么还要自己写呢？因为我想写完markdown之后，即可以保存在笔记软件中(比如有道)，又可以放到github进行版本管理，还可以发布到博客(比如博客园)。这些如果都操作一遍，是很繁琐的，所以必须交给脚本去做。

原材料

• markdown2 or mistune
• pygments

操作原理

• 首先，我需要一个markdown的词法解析器，然后我需要html转换器。这个可以由markdown2或者mistune来完成。
• 然后，我的笔记中有较多的代码，我需要代码高亮。这首先需要将markdown中的代码块提取出来，然后判断是哪种语言，然后进行着色。这部分可以由pyments完成

代码

使用mistune(源码很有学习价值)。需要自己引入pygments模块渲染代码块，官网有参考例子。

import mistune
import sys
import codecs
from pygments import cnblogs_code 
from pygments.lexers import get_lexer_by_name
from pygments.formatters import html


class HighlightRenderer(mistune.Renderer):
    def block_code(self, code, lang):
        if not lang:
            return '
<pre><code>%s</code></pre>
' % 
                mistune.escape(code)
        lexer = get_lexer_by_name(lang, stripall=True)
        formatter = html.HtmlFormatter()
        return cnblogs_code (code, lexer, formatter)

def main(argv):
    name = argv[0]

    input_file = codecs.open(name, mode='r', encoding='utf-8')

    text = input_file.read()
    renderer = HighlightRenderer()
    markdown = mistune.Markdown(renderer=renderer)
    html = markdown(text)

    html_name = '%s.html' % (name[:-3])
    output_file = codecs.open(
        html_name, 'w', encoding='utf-8', errors='xmlcharrefreplace')

    output_file.write(html)

if __name__ == "__main__":
    main(sys.argv[1:])

上面代码还不能使代码着色，因为没有指定css，还需要在生成的html头中加入css，不同的css文件可以在http://richleland.github.io/pygments-css/找到。

<style type = "text/css">
.cnblogs_code  .hll { background-color: #ffffcc }
.cnblogs_code  .c { color: #60a0b0; font-style: italic } /* Comment */
.cnblogs_code  .err { border: 1px solid #FF0000 } /* Error */
.cnblogs_code  .k { color: #007020; font-weight: bold } /* Keyword */
.cnblogs_code  .o { color: #666666 } /* Operator */
.cnblogs_code  .cm { color: #60a0b0; font-style: italic } /* Comment.Multiline */
.cnblogs_code  .cp { color: #007020 } /* Comment.Preproc */
.cnblogs_code  .c1 { color: #60a0b0; font-style: italic } /* Comment.Single */
.cnblogs_code  .cs { color: #60a0b0; background-color: #fff0f0 } /* Comment.Special */
.cnblogs_code  .gd { color: #A00000 } /* Generic.Deleted */
.cnblogs_code  .ge { font-style: italic } /* Generic.Emph */
.cnblogs_code  .gr { color: #FF0000 } /* Generic.Error */
.cnblogs_code  .gh { color: #000080; font-weight: bold } /* Generic.Heading */
.cnblogs_code  .gi { color: #00A000 } /* Generic.Inserted */
.cnblogs_code  .go { color: #808080 } /* Generic.Output */
.cnblogs_code  .gp { color: #c65d09; font-weight: bold } /* Generic.Prompt */
.cnblogs_code  .gs { font-weight: bold } /* Generic.Strong */
.cnblogs_code  .gu { color: #800080; font-weight: bold } /* Generic.Subheading */
.cnblogs_code  .gt { color: #0040D0 } /* Generic.Traceback */
.cnblogs_code  .kc { color: #007020; font-weight: bold } /* Keyword.Constant */
.cnblogs_code  .kd { color: #007020; font-weight: bold } /* Keyword.Declaration */
.cnblogs_code  .kn { color: #007020; font-weight: bold } /* Keyword.Namespace */
.cnblogs_code  .kp { color: #007020 } /* Keyword.Pseudo */
.cnblogs_code  .kr { color: #007020; font-weight: bold } /* Keyword.Reserved */
.cnblogs_code  .kt { color: #902000 } /* Keyword.Type */
.cnblogs_code  .m { color: #40a070 } /* Literal.Number */
.cnblogs_code  .s { color: #4070a0 } /* Literal.String */
.cnblogs_code  .na { color: #4070a0 } /* Name.Attribute */
.cnblogs_code  .nb { color: #007020 } /* Name.Builtin */
.cnblogs_code  .nc { color: #0e84b5; font-weight: bold } /* Name.Class */
.cnblogs_code  .no { color: #60add5 } /* Name.Constant */
.cnblogs_code  .nd { color: #555555; font-weight: bold } /* Name.Decorator */
.cnblogs_code  .ni { color: #d55537; font-weight: bold } /* Name.Entity */
.cnblogs_code  .ne { color: #007020 } /* Name.Exception */
.cnblogs_code  .nf { color: #06287e } /* Name.Function */
.cnblogs_code  .nl { color: #002070; font-weight: bold } /* Name.Label */
.cnblogs_code  .nn { color: #0e84b5; font-weight: bold } /* Name.Namespace */
.cnblogs_code  .nt { color: #062873; font-weight: bold } /* Name.Tag */
.cnblogs_code  .nv { color: #bb60d5 } /* Name.Variable */
.cnblogs_code  .ow { color: #007020; font-weight: bold } /* Operator.Word */
.cnblogs_code  .w { color: #bbbbbb } /* Text.Whitespace */
.cnblogs_code  .mf { color: #40a070 } /* Literal.Number.Float */
.cnblogs_code  .mh { color: #40a070 } /* Literal.Number.Hex */
.cnblogs_code  .mi { color: #40a070 } /* Literal.Number.Integer */
.cnblogs_code  .mo { color: #40a070 } /* Literal.Number.Oct */
.cnblogs_code  .sb { color: #4070a0 } /* Literal.String.Backtick */
.cnblogs_code  .sc { color: #4070a0 } /* Literal.String.Char */
.cnblogs_code  .sd { color: #4070a0; font-style: italic } /* Literal.String.Doc */
.cnblogs_code  .s2 { color: #4070a0 } /* Literal.String.Double */
.cnblogs_code  .se { color: #4070a0; font-weight: bold } /* Literal.String.Escape */
.cnblogs_code  .sh { color: #4070a0 } /* Literal.String.Heredoc */
.cnblogs_code  .si { color: #70a0d0; font-style: italic } /* Literal.String.Interpol */
.cnblogs_code  .sx { color: #c65d09 } /* Literal.String.Other */
.cnblogs_code  .sr { color: #235388 } /* Literal.String.Regex */
.cnblogs_code  .s1 { color: #4070a0 } /* Literal.String.Single */
.cnblogs_code  .ss { color: #517918 } /* Literal.String.Symbol */
.cnblogs_code  .bp { color: #007020 } /* Name.Builtin.Pseudo */
.cnblogs_code  .vc { color: #bb60d5 } /* Name.Variable.Class */
.cnblogs_code  .vg { color: #bb60d5 } /* Name.Variable.Global */
.cnblogs_code  .vi { color: #bb60d5 } /* Name.Variable.Instance */
.cnblogs_code  .il { color: #40a070 } /* Literal.Number.Integer.Long */
</style>

所以完整的代码应该为：

import mistune
import sys
import codecs
from pygments import cnblogs_code 
from pygments.lexers import get_lexer_by_name
from pygments.formatters import html


class HighlightRenderer(mistune.Renderer):
    def block_code(self, code, lang):
        if not lang:
            return '
<pre><code>%s</code></pre>
' % 
                mistune.escape(code)
        lexer = get_lexer_by_name(lang, stripall=True)
        formatter = html.HtmlFormatter()
        return cnblogs_code (code, lexer, formatter)

def main(argv):
    md_name = argv[0]

    with codecs.open(md_name, mode='r', encoding='utf-8') as mdfile:
        with codecs.open("friendly.css",mode = 'r',encoding = 'utf-8') as cssfile:
            md_text = mdfile.read()
            css_text = cssfile.read()
            renderer = HighlightRenderer()
            markdown = mistune.Markdown(renderer=renderer)
            html_text = markdown(md_text)

            html_name = '%s.html' % (md_name[:-3])
            with codecs.open(html_name, 'w', encoding='utf-8', errors='xmlcharrefreplace') as output_file:
                output_file.write(css_text + html_text)

if __name__ == "__main__":
    if len(sys.argv) == 2:
        main(sys.argv[1:])
    else:
        print("Error:please specify markdown file path")

friendly.css文件中存放之前的css文件。

同样使用markdown2的代码如下：

import markdown2
import codecs
import sys


def main(argv):
    md_name = argv[0]

    with codecs.open(md_name, mode='r', encoding='utf-8') as mdfile:
        with codecs.open("friendly.css", mode='r', encoding='utf-8') as cssfile:
            md_text = mdfile.read()
            css_text = cssfile.read()

            extras = ['code-friendly', 'fenced-code-blocks', 'footnotes']
            html_text = markdown2.markdown(md_text, extras=extras)

            html_name = '%s.html' % (md_name[:-3])
            with codecs.open(html_name, 'w', encoding='utf-8', errors='xmlcharrefreplace') as output_file:
                output_file.write(css_text + html_text)

if __name__ == "__main__":
    if len(sys.argv) == 2:
        main(sys.argv[1:])
    else:
        print("Error:please specify markdown file path")