mdbook 0.1 to 0.2

1to2.py

import os
import re
import sys


LINK_PATTERN = r'(?m)^(\[[^]]+\]: )([^\n]*)()'
LINK_PATTERN2 = r'(\[[^]]+\]\()([^)]*)(\))'


def repl(match, parts, filename):
    link = match.group(2)
    if link.startswith('http'):
        return match.group(0)
    if not link.startswith('#'):
        # Don't fixup links to outside documents.
        if not link.startswith('..'):
            link = link.replace('.html', '.md')
        if parts:
            link = os.path.relpath(link, '/'.join(parts))
    # print('orig=%r new=%r parts=%r' % (match.group(2), link, parts))
    return match.group(1) + link + match.group(3)


def process(path, parts):
    print(path)
    contents = open(path).read()
    filename = os.path.basename(path)
    for pattern in [LINK_PATTERN, LINK_PATTERN2]:
        contents = re.sub(pattern, lambda x: repl(x, parts, filename), contents)
    open(path, 'w').write(contents)


def main(start):
    for (dirpath, dirnames, filenames) in os.walk(start):
        relative = os.path.relpath(dirpath, start)
        if relative == '.':
            parts = []
        else:
            parts = relative.split(os.path.sep)
        for filename in filenames:
            if os.path.splitext(filename)[1] == '.md':
                process(os.path.join(dirpath, filename), parts)


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