Re: [Python-markdown-discuss] Can't manage to use an extension when using documentParse()

[Waylan L. <way...@ic...>]

Thomas,

The fenced_code_block extension uses a preprocessor to parse fenced code 
blocks, which stashes the blocks for later retrieval. They are then 
reinserted into the rendered HTML by a postprocessor. As you are only 
running the blockprocessors, none of the pieces of the extension 
actually ever are run. The only way to use the included fenced code 
block extension is to render to HTML. Then run the output through an 
HTML parser and iterate over that. Of course, it is then imposable to 
tell whether the code block started as a indented block or fenced block.

We do have plans to improve things as outlined in 
https://github.com/waylan/Python-Markdown/issues/53 but I can't say when 
that will happen. Its not particularly high on my priority list as the 
current solution is "good enough" for my needs.

I suppose you could iterate over the stashed blocks. But you would need 
to at least run the preprocessor first. The main concern here is that 
the stash is just the `HtmlStash` instance for raw HTML. There is no 
differentiation between the code blocks and other raw HTML. You might 
need to disable raw HTML parsing (another preprocessor). Although, you 
might be able to just run `FencedCodeExtension.run` directly. Just make 
sure you do whitespace normalization first.

I hope that helps. Let me know if you get stuck on anything and I'll see 
if I can help.

Waylan Limberg

On 12/30/2016 02:18 AM, Thomas Baruchel wrote:
> Hi, I would like to iterate over code-blocks of a Markdown document without
> rendering it to anything. I can do it easely with Python-Markdown, but I can't
> manage to add the fenced block extension. My code is:
>
>
>       import markdown
>       from markdown.extensions.fenced_code import FencedCodeExtension
>       M = markdown.Markdown()
>       FencedCodeExtension().extendMarkdown(M, markdown.__dict__)
>
>       f = open("demo.md", "r")
>       tree = M.parser.parseDocument(f.readlines())
>       f.close()
>       for k in tree.iter():
>           if k.tag == "code":
>               print(k)
>               print(k.text)
>
> It still detects indented code-blocks but not the fenced blocks.
> What am I doing wrong?
>
> Thank you by advance,
>
>     regards,
>