Updated support to Kemono.Party

[Nephiro]

I used the Kemono.Party script but lack the author and the download is not perfect because Kemono.Party duplicate the images in the post.
I then updated myself the script fixing these problems.
I'll share it here in case someone wanna use it, it's not perfect but it works.

P.S. I added a variable "SPECIAL_AUTHORS_LIST" where you can insert the author that match the one you find in Kemono.Party with a "fixed" version of the name, mostly because sometimes it doesn't match the real author name from Patreon.

#coding: utf8
#title: Kemono.Party 추가
#title_en: Add Kemono.Party
#comment: https://kemono.party
#author: Kurt Bestor and Nephiro

import downloader
from utils import Downloader, Session, Soup, LazyUrl, urljoin, get_ext, clean_title, get_print, print_error
import clf2
from translator import tr_
from ratelimit import limits, sleep_and_retry


INVALID_EXTENSIONS = ['txt']
SPECIAL_AUTHORS_LIST = [{'found': 'Arisayoshi', 'replace': 'Arisa Yoshi'},
						{'found': 'Melissan', 'replace': 'Melissa N'},
						{'found': 'Meowwithme', 'replace': 'MeowWithMe'},
						{'found': 'Nimbletail', 'replace': 'NimbleTail'},
						{'found': 'Thetransformistress', 'replace': 'TheTransformistress'}
						]

class Image:

    def __init__(self, url, referer, p):
        self._url = url
        self._p = p
        self.url = LazyUrl(referer, self.get, self)

    @sleep_and_retry
    @limits(1, 5)
    def get(self, _):
        ext = get_ext(self._url) or downloader.get_ext(self._url)
        self.filename = '{:04}{}'.format(self._p, ext)
        return self._url


@Downloader.register
class Downloader_kemonoparty(Downloader):
    type = 'kemono.party'
    URLS = ['kemono.party']
    icon = 'base64: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'
    display_name = 'Kemono.Party'
    MAX_CORE = 4

    def read(self):
        self.session = Session()

        info = read_info(self.url, self.session, self.cw)

        type_ = info['type']
        if type_ == 'post':
            for img in info['imgs']:
                self.urls.append(img.url)
            self.title = info['title']
            self.artist = info['artist']
        else:
            raise NotImplementedError(type_)


def read_info(url, session, cw):
    print_ = get_print(cw)
    info = {}
    if '/post/' in url:
        type_ = 'post'
    else:
        raise NotImplementedError('Not a post')
    info['type'] = type_
    res = clf2.solve(url, session=session, cw=cw)

    soup = Soup(res['html'])
    info_ = read_post(url, soup, cw)
    
    info['imgs'] = info_['imgs']
    artist = '{}'.format(clean_title(info_['artist']))
    author = artist.title()
    
    for specialAuthor in SPECIAL_AUTHORS_LIST:
        if specialAuthor['found'] == author:
            author = specialAuthor['replace']
            break
    
    info['title'] = '[{}] {}'.format(author, clean_title(info_['title'], allow_dot=True))
    info['artist'] = artist
    
    return info


def read_post(url, soup, cw):
    print_ = get_print(cw)
    info = {}
    info['title'] = soup.find('h1').findAll('span')[0].text
    info['id'] = soup.find('meta', {'name': 'id'})['content']
    info['artist'] = soup.find('a', {'class': 'post__user-name'}).text
    
    imgs = []
    newImgs = []
    fileNames = []
    
    # Downloads
    for item in soup.findAll('li', class_='post__attachment'):
        href = urljoin(url, item.find('a')['href'])
        img = Image(href, url, len(imgs))
        imgs.append(img)

    # Files
    files = soup.find('div', class_='post__files')
    if files:
        for item in files.findChildren(recursive=False):
            if 'href' in item.attrs:
                a = item
            else:
                a = item.find('a')
            href = urljoin(url, a['href'])

            # Imgur
            if 'imgur.com/' in href:
                print_('Imgur: {}'.format(href))
                try:
                    from extractor import imgur_downloader
                    for img in imgur_downloader.get_imgs(href):
                        img = Image(img, href, len(imgs))
                        imgs.append(img)
                except Exception as e:
                    print_(print_error(e))
                continue

            img = Image(href, url, len(imgs))
            imgs.append(img)

    # Content
    content = soup.find('div', class_='post__content')
    if content:
        for img in content.findAll('img'):
            src = urljoin(url, img['src'])
            img = Image(src, url, len(imgs))
            imgs.append(img)
    
    # Remove duplicates
    for i, img in enumerate(imgs):
        fileName = img._url.rpartition('?f=')[-1]
        ext = fileName.rpartition('.')[-1].lower()
        
        if not ext in INVALID_EXTENSIONS:
            if fileName in fileNames:
                index = fileNames.index(fileName)
                fileNames.pop(index)
                newImgs.pop(index)
            
            fileNames.append(fileName)
            newImgs.append(img)
    
    info['imgs'] = newImgs
    return info


messageBox(u'{}: Kemono.Party'.format(tr_(u'사이트를 추가했습니다')))

[Nephiro]

Oh yes, forgot to say. Fixing the duplicates images you find in Kemono.Party you'll find that the filename at the end a little bit.... wonky. Mean you can end up with: "0007.jpg, 0008.jpg, 0010.jpg" (mean it can also start from 7), but trust me the images will be in the real order and not the strange sort Kemono.Party has.
To fix this I added a personal rename script for the files that works at the end of the process.

[lollos73]

hi
just create a file script and copy/paste your script but don t work for me, can you help me?
need cooky o i miss something?

Restart in 59m 52s: https://kemono.party/fanbox/user/39123643

version: 3.7m (22-06-08 05:17:46 UTC)
platform / locale: Windows-10-10.0.19041-SP0 / it_it
order / group / uid: 1064 / False / f56cc09b1dd540849e1841387e748239
input: https://kemono.party/fanbox/user/39123643
type: kemono.party
single: False
url: https://kemono.party/fanbox/user/39123643
dir:
zip:
artist: None
valid / done: False / True
range / range_p: None / None
time: 1655483641.5002005 (22-06-17 16:34:01 UTC)
tags: []
lock: False
color: invalid
paused: False
format: None
p2f: None
segment: None
admin: True
goodbyedpi: True
ytdl: yt_dlp 2022.05.18
pinned: False
extras: {}
live: False
changed: True

[Gallery]
None

[File Names]

[URLs]

[Messages]
Not a post
stop
Traceback (most recent call last):
File "utils", line 1184, in start
File "utils", line 1254, in start_
File "dynamic_module_0", line 48, in read
File "dynamic_module_0", line 66, in read_info
NotImplementedError: Not a post

Invalid: fail=True
EOT: https://kemono.party/fanbox/user/39123643 (0.5s)

[Nephiro]

Hi, no the problem is simple, the url you copied is not a post but the entire artist collection. Right now it doesn't work coping the entire artist (maybe in the future i'll look up if it's possible to download everything this way), you need to copy every single post url of the artist from Kemono

[lollos73]

i found a program to download entire collection (video psd and zip included) i don't know if this can help or if i can post here (delete message if i can't)
https://github.com/AvocadoFlour/KemonoDownloader