Spike on markdown for Slokas

spike_commonmark.py

# -*- coding: utf-8 -*-
import CommonMark

if __name__ == "__main__":
    with open("trial.md") as f:
        parser = CommonMark.DocParser()
        renderer = CommonMark.HTMLRenderer()
        ast = parser.parse(f.read())
        html = renderer.render(ast)
        json = CommonMark.ASTtoJSON(ast)
        #CommonMark.dumpAST(ast) # pretty print generated AST structure
        print(html) # <p>Hello <em>World</em><p/>
        print(json)

        #html = markdown.markdown(text=f.read(), extensions=[PrajnaExtension()])
    print("--------------------")
    print(html)

spike_markdown.py

# -*- coding: utf-8 -*-

import markdown
import re

from markdown.blockprocessors import BlockProcessor
from markdown.preprocessors import Preprocessor
from markdown.treeprocessors import Treeprocessor
from markdown.postprocessors import Postprocessor


class PrajnaExtension(markdown.Extension):

    def __init__(self, *args, **kwargs):
        print("init")
        pass

    def extendMarkdown(self, md, md_globals):
        print("extendMarkdown")
        md.parser.blockprocessors.add('prajna', MyBlockProcessor(md.parser),
                                      ">indent")
        md.preprocessors.add('prajna', MyPreprocessor(self), "<reference")
        #md.treeprocessors.add('prajna', MyTreeprocessor(self), "_begin")
        md.postprocessors.add('prajna', MyPostprocessor(self), ">amp_substitute")


class MyPreprocessor(Preprocessor):
    MYREGEX = re.compile(r'.*')

    def run(self, lines):
        new_lines = []
        for line in lines:
            m = self.MYREGEX.match(line)
            if m:
                print("pre: line: " + line)
            new_lines.append(line)
        return new_lines


class MyTreeprocessor(Treeprocessor):
    def run(self, root):
        print("tree: ")
        import pdb; pdb.set_trace()
        for node in root:
            print("\t> child: " + node)


class MyPostprocessor(Postprocessor):
    MYREGEX = re.compile(r'.*')
    def run(self, text):
        print("post: " + text)
        return text


class MyBlockProcessor(BlockProcessor):
    SLOKA_RE = re.compile(r'^~~~(\w+)(?:\n|\r\n?)((.|\r\n?)*)(~~~)?', re.MULTILINE | re.DOTALL)
    SLOKA_RE_END = re.compile(r'~~~$')

    block_started = False
    valid_block = True

    def test(self, parent, block):
        if self.SLOKA_RE.search(block) is not None:
            self.valid_block = True
            self.block_started = True
        elif self.SLOKA_RE_END.search(block) is not None:
            self.valid_block = True
            self.block_started = False
        else:
            self.valid_block = False

        print("block: test: " + str(self.valid_block) + ": " + block)
        return self.valid_block

    def run(self, parent, blocks):
        raw_block = blocks.pop(0)
        print("block: parent: " + str(parent.text))
        print("block: raw: " + raw_block)

        if self.block_started is True:
            from markdown.util import etree
            match = self.SLOKA_RE.match(raw_block)
            sloka = etree.SubElement(parent, match.group(1))
            sloka.text = match.group(2)


if __name__ == "__main__":
    with open("trial.md") as f:
        html = markdown.markdown(text=f.read(), extensions=[PrajnaExtension()])
    print("--------------------")
    print(html)

trial.md

SampleKey: SampleValue Metadata1: Metadata1Value

First line of sloka
Second line

and a third line

some additional content

vlah blah

First line split here
Words of second line

and a *3rd* line

first line's information
and that of the second line
and a third line