This project allows you to automaticaly download opening and ending songs from all of your favorite anime without the need of downloading everything yourself. Since almost every weeb uses MAL to track the anime he's watching, this tool is really useful, as every information you need to give it has been written down already. All you need to do is to enter your MAL or AniList username.
All videos are downloaded from animethemes.moe. If you plan on using this program just for looking at openings, I recommend using themes.moe or their own site instead. This program is made for creating your own playlist and such.
This project was made for batch downloading themes from anime you have watched, but is programmed so it's easily improved, making it possible to add very easily. It's made with both command line usage and with python as a module.
- clone this repository from github.com or download it from pip with
pip install animethemes-dl
- if you cloned, do
pip install -r requirements.txt
to install all required modules - install ffmpeg into the same folder or in PATH
make sure you have ffmpeg and python installed
To run in console use animethemes-dl
if installed with pip. python -m animethemes-dl
if you have cloned the repository.
These commands will be reffered to as
animethemes-dl
in the documentation.
The script should raise errors in case you pass in an improper arg, but sometimes an error won't be raised if the error is not obvious, therefore make sure you read the documentation before running it.
You must set a username and a save folder.
You must set a username. By default usernames are assumed to be a MAL user, you can use a different site with --site
.
--animelist-args
can be:
- url args for MAL ``
query
andvariables
for POST request for AniList
--animelist-args
are passed as a<key>=<value>
pairs, for example:sort1=1,sort2=14
There are filters for minimum score and priority.
--minscore
is the minimum score between 0 and 10.
--minpriority
is the minimum priority. For mal, use Low=0,Medium=1,High=2
--range <start> <end>
only gets a slice of the animelist.
To download only openings or only endings, use --OP
or --ED
By default, you can just use a --smart
filter, that takes out all the dialogue. This works by removing all themes that contains a part of the episode and spoilers at the same time. This works 95% of the time.
Since animethemes can have a single song bound to multiple anime, --no-copy
filters them.
You can set --banned-tags
or --required-tags
. These will take multiple tags, possible tags are:
Tag | Meaning |
---|---|
spoiler | Video contains spoilers. |
nsfw | Video is NSFW. |
nc | No captions/no credits. |
subbed | Video includes English subtitles of dialogue. |
lyrics | Video includes English lyrics as subtitles. |
uncen | Video does not have censorship. |
You can set a --min-resolution
, they show up in 420,720,1080
.
You can set the required --source
, possible sources are:
Source | Meaning |
---|---|
BD | Video is sourced from a Blu-ray disc. |
DVD | Video is sourced from a DVD. |
Video is sourced from a TV release. |
Some themes contain a part of the episode. You can set a --overlap
to show only some overlaps.
Overlap | Meaning |
---|---|
Over | Part of episode is over the video. |
Transition | Part of episode transitions into the video. |
None | No dialogue in video. |
If you're only looking to remove dialogue, transitions are fairly fine. They don't even have dialogue most of the time, I recommend just banning
Over
Downloads are by default disabled for both video and audio.
You can enable it by setting a save folder. Save folders are set with -a
(audio) and -v
(video).
The filename format can be changed with --filename
.
The possible formats are defined in this table:
Format | Meaning |
---|---|
anime_id | Animethemes' id of anime. |
anime_name | Name of Anime. |
anime_slug | Animethemes' slug of anime. |
anime_year | Year the anime came out. |
anime_season | Season the anime came out. |
theme_id | Animethemes' id of theme. |
theme_type | Type of theme (OP/ED). |
theme_sequence | Sequence of theme. |
theme_group | Group of theme (e.g. language). |
theme_slug | Animethemes' slug of theme (type+sequence). |
entry_id | Animethemes' id of entry. |
entry_version | Version of entry ("" or 1+). |
entry_notes | Notes of entry (e.g. SFX version). |
video_id | Animethemes' id of video. |
video_basename | Animethemes' basename of video. |
video_filename | Basename without the filetype. |
video_size | Size of file in bytes. |
video_resolution | Resolution of video. |
video_source | Where the video was sourced from. |
video_overlap | Episode overlap over video. |
song_id | Animethemes' id of song. |
song_title | Title of song. |
video_filetype | Filetype of video. |
anime_filename | Name of anime used in filenames. |
formats should be used as a python format string, meaning that it will be put as
%(format)s
. For example%(anime_filename)s-%(theme_slug)s.%(video_filetype)s
.
Windows and Linux banned characters will be removed by default, to remove those and also unicode characters use
--ascii
You can disable redownloading with -r
. This is highly recommended. If you have downloaded video you can --update
theme, this will check file validity by looking at the filesize. It will also update audio files if the video is downloaded.
You can add a coverart to audio files with --coverart
, --coverart
takes in a resolution, if set, image will be fetched from anilist.co, with high resolutions it's recommended to save them in --coverart-folder
.
Downloader timeout can be changed with --timeout
and max amount of retries with --retries
.
Sometimes when using filters a video that you wanted gets filtered out. you can --force-videos
and keep them this way.
re:zero for example has lots of unique EDs, but they often have an overlay, meaning smart filter will remove them.
Data from animethemes is sending a lot of requests at the same time, so to reduce stress on the servers, the data is saved in a temp folder. You can change it's max age with --max-cache-age
.
You can download anime that you have --on-hold
,--dropped
or --planned
.
Downloaded files can be compressed in case you want to save them.
It will be enabled by setting a directory you want to compress with --compress-dir
, this should be the same directory as you chosen one. The destination file is set with --compress-name
, set it without the extension.
You can choose the --compress-format
, this must be a format allowed by shutils.make_archive
.
Additionally you can set the --compress-base
.
You can set the loglevel with --loglevel
. This will set the logger.setLevel(...)
.
There are quick commands --quiet
(print none) and --verbose
(print all). To restrict download and ffmpeg messages, you MUST use --quiet
.
You can disable color with --no-color
.
In case you haven't added ffmpeg to path, you can set the path with --ffmpeg
.
In case the mp3 tags are not showing, you can specify to --use-id3v23
, that will allow support for older systems.
You can --repair
in case the script made some errors or you picked wrong options. This will delete unexpected files and readd metadata.
You can load options from a file with --options
, the file is in json format.
The default options are:
{
"animelist": {
"username": "",
"site": "MyAnimeList",
"animelist_args": {},
"minpriority": 0,
"minscore": 0,
"range": [0,0]
},
"filter": {
"smart": false,
"no_copy": false,
"type": null,
"spoiler": null,
"nsfw": null,
"resolution": 0,
"nc": null,
"subbed": null,
"lyrics": null,
"uncen": null,
"source": null,
"overlap": null
},
"download": {
"filename": "%(anime_filename)s-%(theme_slug)s.%(video_filetype)s",
"audio_folder": null,
"video_folder": null,
"no_redownload": false,
"update": false,
"ascii": false,
"timeout": 5,
"retries": 3,
"max_cache_age": 10368000,
"force_videos": []
},
"coverart": {
"resolution": 0,
"folder": null
},
"compression": {
"root_dir": null,
"base_name": "animethemes",
"format": "tar",
"base_dir": null
},
"statuses": [1,2],
"quiet": false,
"no_colors": false,
"ffmpeg": "ffmpeg",
"id3v2_version": 4,
"ignore_prompts": false
}
You can generate the options with
python -m animethemes_dl.options
.
The code uses the module models
that contains models of typing.TypedDict
.
Meaning python 3.8 is required.
Module parsers
contains all parsers for MAL, Anilist and themes.moe.
Module tools contains extra tools for animethemes-dl
.
examples:
# parsers module uses API's to get data
import animethemes_dl.parsers as parsers
parsers.fetch_animethemes(username) # fetchess raw data
parsers.get_download_data(username) # gets download data
# models module uses typedDict to help linters
import animethemes_dl.models as models
animelist: AnimeThemeAnime = _myanimefunc()
metadata: Metadata = _mymetadatafunc()
# tools have multiple tools used for several stuff
import animethemes_dl.tools as tools
tools.ffmpeg_convert(webm_file,mp3_file) # converts a webm file
tools.COLORS['progress'] = Fore.CYAN # changes colors
tools.compress_files(base,'zip',root) # compresses a direcotory
tools.update_metadata(
parsers.get_download_data(username),False
) # updates metadata of all audio files
# you can implement your own batch dl
import animethemes_dl
data = parsers.get_download_data(username)
for theme in data:
animethemes_dl.download_theme(theme,True)
# you can directly change options
animethemes_dl.setOptions(options)
# you can make special catchers
import animethemes_dl.errors as errors
try:
animethemes_dl.batch_download(data)
except FfmpegException:
print('oh no')
- parser
- get data from MAL/AniList
- get data from themes.moe
- combine data
- filter out unwanted themes
- create download data
- download
- download video file
- convert video to audio
- convert with ffmpeg
- add mp3 metadata
- optional
- compress files
- code optimizations
- improve code documentation
- make a better README (too complicated rn)
- concurrent downloads, since animethemes disabled multithreaded dl.
- support for aria2c