[录制设置] section in config/config.ini. The file is read at startup and re-checked on each monitoring loop, so most changes take effect without restarting.
Annotated config.ini
config/config.ini
Field reference
Language
Interface language. Set to
en to display console output in English.File saving
Absolute or relative path for saving recordings. Leave blank to use the default
downloads/ folder next to main.py.When
是, recordings are placed in a sub-folder named after the streamer. Set to 否 to save all files flat in the root save directory.When
是, a date-based sub-folder (YYYY-MM-DD) is created inside the streamer folder.When
是, a sub-folder named after the stream title is created.When
是, the stream title is appended to the output filename.Strips emoji characters from streamer names and titles when building file and folder names. Recommended to avoid filesystem compatibility issues.
Video format and quality
Output container format. Accepted values:
| Value | Description |
|---|---|
ts | MPEG-TS — recommended; resistant to corruption if recording is interrupted |
mkv | Matroska container |
flv | Flash Video |
mp4 | MPEG-4 |
mp3音频 | Audio only, MP3 |
m4a音频 | Audio only, M4A/AAC |
Global quality preference applied to all rooms. Accepted values:
Individual rooms can override this setting in
| Value | Meaning |
|---|---|
原画 | Original / source quality |
超清 | Ultra HD |
高清 | HD |
标清 | SD |
流畅 | Smooth / lowest |
URL_config.ini. See URL configuration.Threading and loop
Maximum number of concurrent HTTP threads used to query room status and fetch stream URLs. The value is automatically reduced when the error rate rises.
How often (in seconds) the tool re-checks the URL list for rooms that are live. Lower values detect stream starts faster but increase server request frequency. Keep this at 300 seconds or higher for long-running unattended monitoring.
Delay in seconds between reading each URL in the queue on a single loop pass. Increase this to spread requests over time.
Display a countdown timer in the console showing seconds until the next loop.
Print the resolved stream URL to the console and log file. Useful for debugging.
Segment recording
Split recordings into multiple files based on
视频分段时间. Recommended for long streams to keep file sizes manageable.Length of each segment in seconds when segment recording is enabled. Default is 1800 seconds (30 minutes).
Post-processing
After recording ends, automatically remux the file to MP4 using FFmpeg. Stream copying is used by default (no quality loss, fast).
When
是, the video is re-encoded to H.264 during MP4 conversion instead of stream-copied. This produces smaller, more compatible files but is slower and uses more CPU.Delete the original file (e.g. the
.ts file) after remuxing or conversion succeeds.Subtitle and script
Generate an SRT subtitle file alongside the recording. Each subtitle entry contains a real-world timestamp, making it easy to find specific moments in long recordings.
Run a custom script automatically after each recording completes.
The command to execute. Supports Python scripts, shell scripts (bash/bat), and arbitrary shell commands. The following arguments are automatically appended:
--record_name— streamer display name--save_file_path— absolute path to the recorded file--save_type— output format (e.g.TS)--split_video_by_time— whether segment recording was enabled--converts_to_mp4— whether MP4 conversion was enabled
python /home/user/scripts/upload.pyDisk space
Minimum free disk space in GB. Recording stops automatically if available space on the save drive drops below this threshold.
HTTPS
Force all stream URLs to use HTTPS. Enable this if recordings fail due to mixed-content or redirect issues.
