Ignore this article as it is old!

As pointed out by redtree, I learned that Pelican now uses github style code block notation by default.

So please ignore the content of this article! Pelican is the best!

Introduction

I use Pelican as a blogging tool. By the way, when I was a kid, when I went out to the garden, I suddenly saw a pelican in front of me. In front of me, who was dismayed, the pelican screamed "gar" and flew away.

Even if I told people about it, I was disappointed that no one could believe it. As time goes by and my memory fades, "Is that really a pelican? Isn't it a mistake on a pelican flight?" I sometimes think that.

Well, Pelican blogs can be written using Markdown notation, Especially for those who write a lot of source code, than plain Markdown Github-Flavored-Markdown (GFM) is better I think there are many people who are used to it and find it easy to write.

Unfortunately, you usually can't use GFM.

Python has a library called misaka that can use GFM notation, so let's make it available.

Modify the source code

I'm sorry for those who read the meaninglessly long "Introduction", It's a crazy, pretty defeated fix. I think that it should be realized by plug-ins etc., but it is troublesome.

By the way, I'm trying the Python version with 2.7 recommended by Pelican. The OS is Mac OS 10.9.

installation of misaka

Install with pip.

$ pip install misaka

Source fix

Markdown parsing is done in a file called readers.py, so I will mess with that file directly.

If you don't know where the files are, use pip show to find them.

(pelican)comme@garcons$ pip show pelican                                                                   
---
Name: pelican
Version: 3.3
Location: /Users/comme/.virtualenvs/pelican/lib/python2.7/site-packages
Requires: feedgenerator, jinja2, pygments, docutils, pytz, blinker, unidecode, six

You can find readers.py in the pelican directory of the directory indicated by location.

Corrected here on the 19th line

try:
    from markdown import Markdown
except ImportError:
    Markdown = False  # NOQA

↓ Correct like this

try:
    from markdown import Markdown
    import misaka
    import pygments
except ImportError:
    Markdown = False  # NOQA

Fixed here around line 215

    def read(self, source_path):
        """Parse content and metadata of markdown files"""

        self._md = Markdown(extensions=self.extensions)
        with pelican_open(source_path) as text:
            content = self._md.convert(text)

        metadata = self._parse_metadata(self._md.Meta)
        return content, metadata

↓ Correct like this

    def read(self, source_path):
        """Parse content and metadata of markdown files"""

        self._md = Markdown(extensions=self.extensions)
        with pelican_open(source_path) as text:
            content = self._md.convert(text)

        metadata = self._parse_metadata(self._md.Meta)
   
        renderer = self.BleepRenderer()
        misaka_md = misaka.Markdown(renderer,
            extensions=misaka.EXT_FENCED_CODE | misaka.EXT_NO_INTRA_EMPHASIS)
        with pelican_open(source_path) as text:
            misaka_content = misaka_md.render(text)

        return misaka_content, metadata

    class BleepRenderer(misaka.HtmlRenderer, misaka.SmartyPants):
        def block_code(self, text, lang):
            if not lang:
                return '\n<pre><code>%s</code></pre>\n' % escape(text.strip())
            lexer = pygments.lexers.get_lexer_by_name(lang, stripall=True)
            formatter = pygments.formatters.HtmlFormatter()
            return pygments.highlight(text, lexer, formatter)

That's it.

Now, like kobito and github,

```python
def func():
    print('uuuuuuuuuuwaaaaaaaaaaaaaaaaaaahhhhhhhh')
```

If you write like, and make html, you can get the html you expected.

diff

--- a/readers.py
+++ b/readers.py
@@ -18,6 +18,8 @@ except ImportError:
     docutils = False
 try:
     from markdown import Markdown
+    import misaka
+    import pygments
 except ImportError:
     Markdown = False  # NOQA
 try:
@@ -215,7 +217,22 @@ class MarkdownReader(BaseReader):
             content = self._md.convert(text)

         metadata = self._parse_metadata(self._md.Meta)
-        return content, metadata
+
+        renderer = self.BleepRenderer()
+        misaka_md = misaka.Markdown(renderer,
+            extensions=misaka.EXT_FENCED_CODE | misaka.EXT_NO_INTRA_EMPHASIS)
+        with pelican_open(source_path) as text:
+            misaka_content = misaka_md.render(text)
+
+        return misaka_content, metadata
+
+    class BleepRenderer(misaka.HtmlRenderer, misaka.SmartyPants):
+        def block_code(self, text, lang):
+            if not lang:
+                return '\n<pre><code>%s</code></pre>\n' % escape(text.strip())
+            lexer = pygments.lexers.get_lexer_by_name(lang, stripall=True)
+            formatter = pygments.formatters.HtmlFormatter()
+            return pygments.highlight(text, lexer, formatter)


 class HTMLReader(BaseReader):