tankobon

What?

tankobon is a website scraper for comics and mangas. tankobon relies on stores, which define how to parse a website for chapters and chapters for links to the pages themselves. (somewhat like youtube-dl extractors.) Currently, the following websites are supported:

• komi-san.com
• m.mangabat.com
• mangadex.org
• mangakakalot.com

Creating a Store

A store is a regular Python module in the stores/ folder. It should provide a Parser class, which is a subclass of tankobon.manga.Parser. The following methods below must be implemented:

chapters(self) -> Generator[Tuple[str, Dict[str, str]], None, None]

Yields chapter_info which looks like this:

{
    "id": ...,  # chapter number
    "title": ...,  # chapter title
    "url": ...,  # chapter url
    "volume": ...,  # volume, i.e '0'
}

Volume is optional and may be undefined. Example:

def chapters(self):
    # use self.soup to access the title page
    for href in self.soup.find_all("a", href=True):
        # validify href here and parse chapter id
        ...
        yield {"id": ..., "title": href.text, "url": href["href"]}

pages(self, chapter_data: Dict[str, str]) -> List[str]

Return a list of urls to a chapter's pages, given the chapter data yielded from chapters(). The pages must be in order (page 1 is [0], page 2 is [1], etc.) Example:

def pages(self, chapter_data):
    pages = []
    # to get the chapter's html, use self.session.get (requests session)
    # or self.soup (html already parsed by BeautifulSoup).
    chapter_page = self.soup_from_url(chapter_data["url"])

    for href in chapter_page.find_all("a", href=True):
        # validify href here
        ...
        pages.append(href["href"])
    return pages

The following methods below may or may not be implemented: generic implementations are provided.

title(self) -> str

Return the title of the manga. Example:

def title(self):
    return self.soup.title

Index Compatibility

Between version v3.1.0a1 and v3.2.0a0, the location of the index file has moved from site-packages to ~/.tankobon/index.json, specific to each install of tankobon.

Todo

• download pre-parsed indexes from a special Github repo (tankobon-index?)
• create GUI to make downloading easier (like youtube-DLG)

Usage

tankobon download 'https://komi-san.com'  # download all chapters
tankobon store info 'komi_san/https://komi-san.com'  # and then get info on the chapters

Install

python(3) -m pip install tankobon

Build

All my python projects now use flit to build and publish. To build, do flit build.

License

MIT.