BUG: Point in Link breaks PDF-Generation

[alexander-matthiesen]

Hey, the PDF-Generation will break if a link has a point in it.

For example:
(Test)[index.md#4.-New-Section]

This can be avoided by adding the following codechange in transform-py.

# normalize href to #foo/bar/section:id
def transform_href(href: str, rel_url: str):
    head, tail = os.path.split(href)
    num_hashtags = tail.count('#')

    if tail.startswith('#'):
        head, section = os.path.split(rel_url)
        section = os.path.splitext(section)[0]
        id = tail[1:]
    elif num_hashtags is 1:
        section, ext = tuple(os.path.splitext(tail))

        try:
            id = str.split(ext, '#')[1]
        except:
            return href
        
        if head == '..':
            href = normalize_href(href, rel_url)
            return '#{}:{}'.format(href, id)

    elif num_hashtags is 0:
        if not is_doc(href):
            return href

        href = normalize_href(href, rel_url)
        return '#{}:'.format(href)

    if head != '':
        head += '/'

    return '#{}{}:{}'.format(head, section, id)