This is a list of small scripts I have made to make my life easier and avoid a lot of repetitive tasks. These are intended for my own personal use, so they are not exactly "user-friendly," but feel free to ask me if you need help using them or installing dependencies!
-
DLsiteMetadataDownloader.py: Given a DLsite RJ code, this script will scrap the metadata for the product page and apply it to the mp3 files. This is intended for music or voice dramas.
-
TwitterImageProcessing.py: Given a Tweet URL it will download its images, denoise them using waifu2x-caffe and sort them into the right folder, adding a shortcut file to the tweet in said folder. The
config.json
file is formated as follows:Configuration file
{ "consumer_key": "TWITTER_CONSUMER_KEY_GOES_HERE", "consumer_secret": "TWITTER_CONSUMER_SECRET_GOES_HERE", "access_token": "TWITTER_ACCESS_TOKEN_GOES_HERE", "access_token_secret": "TWITTER_ACCESS_TOKEN_SECRET_GOES_HERE", "download_root": "ABSOLUTE_PATH_TO_WHERE_YOU_WANNA_SORT_YOUR_IMAGES", "waifu_path": "ABSOLUTE_PATH_TO_WAIFU2X-CAFFE_FOLDER" }
- YouTubeDownloadSegment.py: Given a YouTube video URL, a start time and an end time, this script will download the specified segment of the video using youtube-dl and ffmpeg. Useful when the video is 8 hours long and you only need 20 seconds from it.
- set-styles-by-actor.lua: This Aegisub script changes the style of every line based on their actor. If there's a style that exactly matches the actor, then the style will be set to that one. Very useful if, like me, you like doing the translation in a plain text file first.
- hide-original-text.lua: This Aegisub script will put between curly brackets the contents of a line. I use this for the official Holo no Graffiti Spanish subs, that way the Japanese text is conserved but hidden at the same time, making proofreading easier for my translation partner.