Skip to content

⚙️ Client Local Configuration ​

Configuration Reading ​

The default configuration file path is .mdc/mdc.ini under the system user directory.

If a file named mdc.ini also exists in the program directory, the program will read that file first. This allows programs in different locations to use independent configuration files.

Directory Config ​

When one of the Scraping Directory entries is the same as the Output Directory, scraping and organizing can be done directly under the scraping directory. In this mode, only 1 scraping directory can be configured.

Scraping Directory ​

Defaults to the folder where the program is located. You can add multiple folders manually.

Output Directory ​

Defaults to the newly created output folder under the program runtime directory.

Subtitle Config ​

If the subtitle file prefix in the Scraping Directory matches the video file prefix, the subtitle will be automatically moved to the same directory as the video during subsequent processing. The Subtitle tag will also be added to the NFO metadata file, and a watermark will be added to the cover (this can be turned on or off on the web cloud configuration page).

General Settings ​

Debug Mode ​

  • Keep this disabled unless a BUG occurs, otherwise it may increase the probability of program crashes

When a BUG occurs, click Stop, enable Debug Mode, click Run again, then save the output log and related screenshots and send them according to Report BUG.

Check for Update on start ​

  • Recommended to keep enabled to receive the latest update reminders

Chrome Simulator ​

It is recommended to disable this feature. Enabling it will allow you to request sites with strong anti-scraping measures, but it will increase the scraping time. Chrome browser must be installed.

Backend Hosting Mode ​

When checked, the scraping will not automatically stop after each cycle. It will automatically scrape newly added video files at intervals based on the currently selected cloud configuration settings.

Custom URL Scraping ​

  • Click+to add video files
  • Fill in the specific video information URL, such as https://www.themoviedb.org/movie/13189-a-christmas-carol
  • Supported video information URLs include general websites in cloud configuration and video type websites in naming rules
  • During processing, if a video has a custom URL, the custom URL will be used

Actors ​

Add avatars for media server

TIP

Only Emby and Jellyfin are supported.

  • Enter the server address
  • Enter the API key, or open the Key Creation Page, create a new API key, and copy it into the input box
  • Click Save and test
  • Click Start completing avatars

Custom Field Mapping Table ​

  • After the program runs for the first time, a mapping_tables directory will be generated under .mdc in the system user directory. It contains two XML files and can be used with Field Conversion Settings in cloud configuration
  • Actor mapping table: mapping_actor.xml
  • Other field mapping table: mapping_info.xml
  • Simplified Chinese to Traditional Chinese corresponds to zh_tw, Traditional Chinese to Simplified Chinese corresponds to zh_cn, and English en only supports content in the custom mapping table