Dear all,
when importing workouts from ErgDB, I see some issues in Golden Cheetah:
- The names coming from ErgDB are HTML-encoded, making many of them hard to read in the Train-Views sidebar
- Some names contain slashes (/), making it impossible to import these files (both Linux and Windows)
- Some workouts have duplicate names, allowing only one of them to be stored in Golden Cheetah
I have implemented something I consider a solution to these issues:
- The titles are HTML-decoded both in ErgDBDownloadDialog and ErgFile (here also copied to the field Name)
- Filesystem: All ErgDB-workouts are stored as ergb-<ergdbID>.erg2
- TrainDB: The decoded title is stored in the column "workouts.filename" instead of the real filename
- Recorded CSV: An optional metadata section above the payload was added (format: "# <Key>:<Value>") to allow the Route-Tag to be set correctly when importing the data
Drawback of my solution:
- I had to manually remove all "old" ErgDB-workouts from the TrainDB and the filesystem to prevent duplication
If there is interest in my changes, I would be happy to provide a pull request.
Regards
Joachim