getjmanga

Retrieve and save images from Japanese web comic sites.

Do not redistribute the downloaded images. Keep them for private use.

Supported sites

See docs/SUPPORTED_SITES.md.

Installation

# mise via github release
mise use -g github:eggplants/getjmanga

# mise via pipx
mise use -g pipx:getjmanga

# pipx
pipx install getjmanga

# pip
pip install getjmanga

Docker

docker pull ghcr.io/eggplants/getjmanga

docker run --rm -v "$PWD:/work" -w /work \
  ghcr.io/eggplants/getjmanga https://takecomic.jp/episodes/74f33031e13cd

CLI

# one episode
jm https://takecomic.jp/episodes/74f33031e13cd https://piccoma.com/web/viewer/8195/1185884

# episodes in bulk: this one and every next one
jm -b https://shonenjumpplus.com/episode/13932016480028799982

# every previous one too: the whole work from one episode
jm -B https://shonenjumpplus.com/episode/13932016480028799982

# login
jm -u you@example.com https://piccoma.com/web/viewer/8195/1185884

# save pages as png (or webp) instead of jpg
jm -F png https://takecomic.jp/episodes/74f33031e13cd

# also pack the saved pages into <series>/_cbz/<episode>.cbz, with a ComicInfo.xml naming the work and its author
jm -C https://takecomic.jp/episodes/74f33031e13cd

# every link on a page that some extractor takes
jm -s https://shonenjumpplus.com/

# numbered pages
# 1 to 3
jm -s "https://comic-ryu.jp/series/list/up/[1-3]"
# from 1 on until a page has nothing to download
jm -s "https://comic-ryu.jp/series/list/up/[1-]"

# remember the work, then download what is new in every remembered work
jm -S -b https://shonenjumpplus.com/episode/13932016480028799982
jm patrol

# one timestamped log line per step and per request, instead of the live display
jm -v https://takecomic.jp/episodes/74f33031e13cd
# nothing but the warnings and the errors
jm -q https://takecomic.jp/episodes/74f33031e13cd

What is going on is shown on one or two lines that come down once a work is done, leaving one line per work behind:

saved: /home/you/manga/shonenjumpplus.com/阿波連さんははかれない (2 episodes, 1 skipped, 3 locked)
skipped: /home/you/manga/takecomic.jp/メイドインアビス (12 episodes)
done.

Configuration

Use jm c.

Default: ~/.config/getjmanga/config.toml

Example: config.example.toml

jm c init

# asks for the username and password
jm c site shonenjumpplus.com
jm c site piccoma

jm c savedir ~/manga
jm c overwrite true
jm c bulk false
jm c both true
jm c format webp
jm c cbz true
jm c metadata true

jm c patrol https://shonenjumpplus.com/episode/13932016480028799982
jm c patrol -s https://shonenjumpplus.com/

# read the file back: a key nothing reads, a savedir that is not a directory,
# a [site] no extractor reads, a patrol url no extractor takes or listed twice
jm c check

Patrol

jm -S adds what it downloads to a list of works to watch for new episodes in the config file.

jm c patrol <url> adds a url to the list but does not download it.

jm p then goes through the list. It skips the episodes that are already saved, so it downloads only what is new.

What jm p does with an entry depends on what the entry is:

• For an episode, it follows the next links to the newest episode. The entry then moves to the first episode that is still locked. As a result, a wait-to-read episode gets one more try next time.
• For a series page, it reads the episode list again.
• For a page stored with -s, it scans the links again.

patrol = [
  { url = "https://shonenjumpplus.com/episode/13932016480028799982", title = "阿波連さんははかれない" },
  { url = "https://takecomic.jp/series/3f846451aff2d/1", title = "メイドインアビス" },
  { url = "https://shonenjumpplus.com/", search = true },
  ...
]

Running every day

jm p saves into savedir from the config file, which defaults to the current directory, so set it first: jm c savedir ~/manga. Both cron and systemd run with a minimal PATH, so use the full path that command -v jm prints.

cron

crontab -e, then:

# every day at 04:00, logging one line per work to a file
0 4 * * * /home/you/.local/bin/jm p >> /home/you/.local/state/getjmanga/patrol.log 2>&1

# log only the skips and the errors
0 4 * * * /home/you/.local/bin/jm p -q >> /home/you/.local/state/getjmanga/patrol.log 2>&1

# log every step, with a timestamp
0 4 * * * /home/you/.local/bin/jm p -v >> /home/you/.local/state/getjmanga/patrol.log 2>&1

systemd timer

~/.config/systemd/user/getjmanga-patrol.service:

[Unit]
Description=Download what is new in every patrolled work

[Service]
Type=oneshot
ExecStart=/home/you/.local/bin/jm p

~/.config/systemd/user/getjmanga-patrol.timer:

[Unit]
Description=Run getjmanga-patrol daily

[Timer]
OnCalendar=daily
Persistent=true

[Install]
WantedBy=timers.target

Persistent=true runs the job at the next boot when the machine was off at the scheduled time.

systemctl --user daemon-reload
systemctl --user enable --now getjmanga-patrol.timer

# keep the user units running after logout
loginctl enable-linger "$USER"

# next run, and the output of the last one
systemctl --user list-timers getjmanga-patrol.timer
journalctl --user -u getjmanga-patrol.service

Library

from getjmanga import Downloader, find_extractor

url = "https://takecomic.jp/episodes/74f33031e13cd"
extractor = find_extractor(url)() # returns `Comici`
result = Downloader(extractor, "out", fmt="png", cbz=True).download(url)

result.status
result.save_dir

result.archive
result.episode.next_url
result.episode.writer
result.episode.publisher
result.episode.published
result.episode.number

An extractor on its own reads the site and writes nothing:

from getjmanga import Comici

comici = Comici()
for url in comici.series_urls("https://takecomic.jp/series/b167ea507d35f"):
    episode = comici.episode(url)
    print(episode.episode_title, len(episode.pages), episode.readable)

Writing an extractor

See docs/ADD_SITE.md.

License

MIT License