Friendly Usable Zero‐Bullshit Analyzer & Labeler: a keyboard‐driven utility for tagging events in video clips, created for analyzing football matches but likely useful for other kinds of videos.
While functional, this project is not yet production‐ready. While unlikely, it might eat your files or your cake.
## Usage
Open a video. Press `space` to start or stop video playback. Seek with `←` and `→`. Use `,` and `.` to decrease and increase the playback rate, and `=` to reset it.
To add a new event, press the key for the corresponding tag and fill out event details. Custom tags can be defined as a JSON array and loaded at runtime. See `tags.json` for the built‐in example showcasing all supported field types.
Events for `video.mp4` are saved in JSON format to the file `video.mp4.events`. The saved file includes tag definitions, which are loaded automatically when the video is opened. Event timestamps are stored with millisecond precision.
### Filter
Input space‐separated search strings to select a subset of events. Selected events must match each string, either as the tag name or a field value. Tags are matched from the beginning, so `p` will select events tagged `pass` and `penalty`.
Text and enum fields match when they contain the given search string. Prefixing the string with `=` specifies a complete match. The string `2` thus matches values `123`, `2` and `22`, while `=2` only matches `2`. A boolean field matches when it’s value is true and it’s name starts with the search string. Prefixing the string with `!` inverts the match.
Finally, a search string can be prefixed with a (prefix of) field name, as in `play:2`. This selects events with a field named `player` whose value matches `2` as above. This can be combined with the `!` and `=` operators, as in `pass:!success`.