Files and folders¶
MiningCat works inside the project folder. Everything it generates goes to output/, which is ignored by git.
mining-cat/
├── sources/ # your input files (ignored by git)
│ ├── ebook/ # the .epub or .txt file(s)
│ ├── audiobook/ # the audio file(s)
│ └── game_ocr.json # the video game window and areas you selected
└── output/
├── chapters_audio/ # one audio file per chapter
├── chapters_text/ # one text file per chapter (chapter_001.txt...)
├── srt/ # generated subtitles (.srt)
├── videos/ # downloaded online videos
├── frequency/ # word frequency (.csv) and character lists (.json)
├── temp/ # intermediate files (chapter metadata, OCR frames...)
└── final/ # The videos you want
library/ # the reader's books, the player's videos, dictionaries and cards (ignored by git)
├── books/ # one folder per book
├── videos/ # one folder per video: the file, its subtitles, your position
└── miningcat.db # dictionaries, word statuses and cards
user/ # your own settings (ignored by git)
├── user-config.json # Settings › User preferences: theme, highlight colors, heatmap color, Discord status
└── secret.key # encrypts the API keys kept in the database (ElevenLabs'): keep it with library/
When you pick files in the GUI, they're copied into sources/ for you. With the CLI, put them there yourself.
Start from a clean slate
Each run overwrites the files of the previous one. Move the videos you want to keep out of output/final/ before processing another book, and run make clean to delete everything generated.
Copyright
sources/ is ignored by git so that copyrighted books and audiobooks never end up in the repository. Don't commit them!