ZdS clean gallery

archive_list_images.py

#!/usr/bin/env python

import json
import re
import sys
from zipfile import ZipFile


img_regex = re.compile(rb'!\[[^\]]*\]\(([^)]+)\)')


def find_images_file(file):
    for line in file:
        if m := img_regex.search(line):
            yield m.group(1).decode()


def find_images_archive(archive):
    for path in archive.namelist():
        if not path.endswith('.md'):
            continue
        with archive.open(path) as file:
            yield from find_images_file(file)


def find_images(*archive_names):
    for archive_name in archive_names:
        with ZipFile(archive_name, 'r') as archive:
            yield from find_images_archive(archive)


if __name__ == '__main__':
    if len(sys.argv) < 2:
        print(f'Usage: {sys.argv[0]} archive [archives]', file=sys.stderr)
        sys.exit(1)

    _, *archive_names = sys.argv
    for img in find_images(*archive_names):
        print(img)

clean_gallery.py

#!/usr/bin/env python

import sys
import tempfile
from urllib.parse import urljoin

import lxml.html
import requests

from archive_list_images import find_images
from list_gallery import filter_images


def get_archive_url(url, doc=None):
    if doc is None:
        req = requests.get(url, cookies=cookies)
        doc = lxml.html.fromstring(req.text)

    archive_url, = doc.xpath('//a[starts-with(normalize-space(), "Archive")]/@href')
    archive_url = urljoin(url, archive_url)
    return archive_url


def get_archives(content_url, doc=None):
    archives = {
        'current.zip': get_archive_url(content_url, doc),
    }

    try:
        beta_url, = doc.xpath('//a[normalize-space()="Voir la version en bêta"]/@href')
    except ValueError:
        pass
    else:
        beta_url = urljoin(content_url, beta_url)
        archives['beta.zip'] = get_archive_url(beta_url)

    try:
        public_url, = doc.xpath('//a[normalize-space()="Voir la version en ligne"]/@href')
    except ValueError:
        pass
    else:
        public_url = urljoin(content_url, public_url)
        archives['public.zip'] = get_archive_url(public_url)

    return archives


def download_archives(archives, dirname):
    for name, url in archives.items():
        req = requests.get(url, allow_redirects=True, cookies=cookies)
        path = f'{dirname }/{name}'
        with open(path, 'wb') as f:
            f.write(req.content)
        yield path


def list_all_public_images(content_url, doc=None):
    archives = get_archives(content_url, doc)

    with tempfile.TemporaryDirectory() as dirname:
        return set(find_images(*download_archives(archives, dirname)))


def delete_image(edit_link, cookies):
    req = requests.get(edit_link, cookies=cookies)
    doc = lxml.html.fromstring(req.text)

    img_name, = doc.xpath('//input[@name="title"]/@value')
    img_src, = doc.xpath('//div[@class="gallery-col-image"]/a[img][1]/@href')
    img_src = urljoin(edit_link, img_src)

    form, = doc.xpath('//form[@id="form-delete-image"]')
    action_url = urljoin(edit_link, form.get('action'))
    image_id, = form.xpath('.//input[@name="image"]/@value')
    csrf, = form.xpath('.//input[@name="csrfmiddlewaretoken"]/@value')

    cookies = cookies | {'csrftoken': csrf}

    req = requests.post(
        action_url,
        data={
            'image': image_id,
            'csrfmiddlewaretoken': csrf,
            'delete': '',
        },
        cookies=cookies,
        headers={'Referer': action_url},
    )
    req.raise_for_status()
    print(f'Image {img_name} ({img_src}) deleted!')


def clean_content_gallery(content_url, cookies):
    req = requests.get(content_url, cookies=cookies)
    doc = lxml.html.fromstring(req.text)

    gallery_url, = doc.xpath('//aside//a[starts-with(@href, "/galerie/")]/@href')
    gallery_url = urljoin(content_url, gallery_url)

    images = list_all_public_images(content_url, doc)
    todelete = set()

    try:
        for img_src, img_name, edit_link in filter_images(gallery_url, cookies, exclude=images):
            print(f'Do you want to delete {img_name} ({img_src})? (y/n)')
            if input('> ').strip().lower() == 'y':
                todelete.add(edit_link)
    except EOFError:
        pass

    for edit_link in todelete:
        delete_image(edit_link, cookies)


if __name__ == '__main__':
    if len(sys.argv) < 3:
        print(f'Usage: {sys.argv[0]} content_url session_id')
        sys.exit(1)

    _, content_url, session_id, *_ = sys.argv

    if not content_url.startswith('https://zestedesavoir.com/contenus/'):
        print(f'Malformatted content URL', file=sys.stderr)
        sys.exit(1)

    cookies = {'sessionid': session_id}

    clean_content_gallery(content_url, cookies)

list_gallery.py

#!/usr/bin/env python

import sys
from urllib.parse import urljoin

import lxml.html
import requests


def find_images_page(page_url, cookies):
    req = requests.get(page_url, cookies=cookies)
    doc = lxml.html.fromstring(req.text)

    img_links = doc.xpath('//div[contains(@class, "gallery-item")]//a')
    images = {}
    for img_link in img_links:
        img_src, = img_link.xpath('.//img/@src')
        img_src, _, _ = img_src.rsplit('.', 2)
        img_src = urljoin(page_url, img_src)
        img_name, = img_link.xpath('.//span[@class="topic-title"]/text()')
        img_edit_link = urljoin(page_url, img_link.get('href'))
        images[img_src] = img_name, img_edit_link

    try:
        rel_next_url, = doc.xpath('(//a[@rel="next"]/@href)[1]')
        next_url = urljoin(page_url, rel_next_url)
    except ValueError:
        next_url = None

    return next_url, images


def iter_gallery_images(gallery_url, cookies):
    next_url = gallery_url
    while next_url is not None:
        next_url, images = find_images_page(next_url, cookies)
        for img_src, (img_name, edit_link) in images.items():
            yield img_src, img_name, edit_link


def filter_images(gallery_url, cookies, exclude):
    for img_src, img_name, edit_link in iter_gallery_images(gallery_url, cookies):
        if img_src not in exclude:
            yield img_src, img_name, edit_link


if __name__ == '__main__':
    if len(sys.argv) < 3:
        print(f'Usage: {sys.argv[0]} gallery_url session_id', file=sys.stderr)
        sys.exit(1)

    _, gallery_url, session_id, *_ = sys.argv

    if not gallery_url.startswith('https://zestedesavoir.com/galerie/'):
        print(f'Malformatted gallery URL', file=sys.stderr)
        sys.exit(1)

    cookies = {'sessionid': session_id}

    for img in iter_gallery_images(gallery_url, cookies):
        print(img)