1.0.2

What's changed

• import-from-url now has the option --add-tags to add tags for all imported media (#56)
• add (configurable) default safety for auto-tagger in case Deepbooru cannot guess the safety (#57)
• update dependencies

Full Changelog: 1.0.1...1.0.2

1.0.1

What's Changed

• update crontab_sample by @shibaobun in #52
• Add option to update tags if it already exists by @reluce in #53

Full Changelog: 1.0.0...1.0.1

1.0.0

Changelog

Breaking

• All previous scripts are now commands of the szuru-toolkit script
  • Basically, just preface szuru-toolkit to the scripts which were available before
  • There are changes on how options are passed to the script. Make sure to check the help of each command (with -h or --help.
• Options for the config file are updated. Make sure to check out the current config_sample.toml

Changes

• All options which you could only set in the config.toml file before are now available as CLI options. These CLI options take priority over the config file.
• The config.toml file is therefore not required anymore
• The config.toml file is now being read from additional sources, see the README for details
• Add docstrings to all of the functions
• Fix an issue where multiple progress bars would appear
• The import-from-booru command is now just a simple wrapper for the import-from-url script
  • This should fix #43 and #47
• Improved startup speed of the script
• Updated dependencies

Full Changelog: 0.9.5...1.0.0

0.9.6

Changelog

• Fix an issue where deprecated imports where causing issues (#51)

0.9.5

Changelog

• Add Fanbox support in import-from-url script
• Try to convert Pixiv/Fanbox tags + Twitter hashtags to Booru compatible tags through the Danbooru wiki
• Add Pixiv Tagging by @nitori-dev in #46
  • New config.toml option: use_pixiv_tags
• Update dependencies
• Small bug fixes

Upcoming 1.0 release

The next release will bring some breaking changes:

• All scripts will be available as subcommands to the szuru-toolkit script:
  • import-from-url -> szuru-toolkit import-from-url
  • auto-tagger -> szuru-toolkit auto-tagger
  • Etc.
• The config.toml is not a requirement anymore to run the script
• All options of config.toml can be specified as command line arguments to the szuru-toolkit script
• The config.toml file will change/move some options

New Contributors

• @nitori-dev made their first contribution in #46

Full Changelog: 0.9.4...0.9.5

0.9.4

Changelog

• Add -v option to import-from-url script which shows the download progress from gallery-dl
• If images are being resized, the original md5 hash of the file will now be used for auto tagging instead of the resized one
• import-from-url will now create a dedicated download directory for each instance in your tmp directory. Running multiple instances of the script should now work better.
• Add a Chrome extension which allows you to import media files from your current tab to your szurubooru via gallery-dl.
  • You have to run the run-webserver.py script in the background (a systemd unit file is recommended)
  • In your Chrome settings, go to your extensions and click on Load unpacked. After that, you can select the chrome_extension folder from this repo.
  • The extension allows you to specify the path to a cookie file and limit the download range

0.9.2

Changelog

• Fix --limit not being respected in import-from-booru script (#43)
• Fix TagNotFoundError when tagging with Deepbooru while searching through implications (#44)

0.9.1

Changelog

• Fix create-relations script not appending tags correctly
• Improved error handling
• Previously, relations within auto-tagger would only be set if SauceNAO was disabled or the limit has been reached. Now, relations will be set with following conditions:
  • If Deepbooru tagging is enabled
  • If the create relation option is enabled
  • If no results were found with SauceNAO and the MD5 hash of the file

Full Changelog: 0.9.0...0.9.1

0.9.0

Breaking

• Require Python 3.11
• Added multiple options to config.toml which have to be present. Compare with config_sample.toml.
• Remove --sankaku-url from auto-tagger script. Sankaku is now supported by the import-from-url script and md5 search feature in auto_tagger.

Changelog

• New script: create-relations: Will iterate through all posts retrieved from the input query and create a relation between character <> parody. Parodies will be added as implications to characters, while characters are added as suggestions to parodies. You can specify the threshold in config.toml how many other posts with this relation should exist before creating the relation.
• Improve Docker by @shibaobun in #39

auto-tagger

• Add support to search with the post's md5 hash on various Boorus (in addition to SauceNAO). Enable/disable in config.toml.
  • Video search by md5 is now also supported because of this addition
• Allow to create relations in auto-tagger script if no result in SauceNAO is found and Deepbooru is enabled. Enable/disable in config.toml.

import-from-url

• Support Deepbooru in import-from-url script
• Add source URL for Kemono, E-Hentai + Twitter
• Added the --cookie flag to import posts from sites where authentication is needed (e.g. Twitter). Check https://github.com/mikf/gallery-dl#cookies for details.
• Allow the Twitter user (name + nickname) to be set as an artist tag (will be created if no alias for the artist is found in Danbooru)

Upcoming

• The import-from-twitter script will be removed in the future as the Twitter API had a massive crackdown on free tier (You can just use import-from-url though)
• All scripts will unify in one script (szuru-toolkit) where each script is a command within szuru-toolkit

Full Changelog: 0.8.1...0.9.0

0.8.1

Changelog

• The auto-tagger script now looks for pixiv results on SauceNao. If only a result on pixiv was found, the script looks up the pixiv artist on Danbooru. If the artist was found on Danbooru (it checks aliases from the artist as well), it will include the artist tag from Danbooru in the tags for that post.
  • You can set set_pixiv_artist in config.toml (check config_sample.toml or the README): In that case, if the artist was not found on Danbooru, it will create the pixiv artist as a tag automatically. Note that the category for that tag is hardcoded to 'artist', so you need to have a tag category 'artist' as well.
  • Other tags on the pixiv post won't be included for now (maybe a character/parody lookup on Danbooru in the future)
• Add sankaku to config sample by @LibertX in #37