Closed brianredbeard closed 3 years ago
Thanks for an amazing issue report!
This is interesting - also something that I wouldn't have caught myself, since I only use beets with write: on
. Just had a brief look at the code and I think I see the culprit
self.add_additional_data(item, write=True)
this line hardcodes write
to True
which is obviously incorrect. That function also currently writes regardless of whether the additional fields are excluded.
I will try to address both of these over the next couple of days.
This should now be fixed with 0.9.2.
While troubleshooting the incorrect population of musicbrainz metadata, I discovered that beets-bandcamp (neé "bandcamp") is ignoring global configurations and mutating source material.
In my workflow I start from "vanilla" files that then go through a number of processing steps before ending up in a curated location.
Related to this, I have a beets configuration solely for handling content from Bandcamp, thus allowing me to rule out conflicts with plugins, misconfigurations, etc.
This is the rendered
config.yaml
:To demonstrate, I'm going to use a recent release: (Tales From The Ultra Tribe by Steve Roach / Byron Metcalf)
The original metadata in the FLAC files appears as follows:
After running
beet import
with the configuration above but thebandcamp
plugin disabled, the metadata on the new file remains the same:However, enabling the
bandcamp
plugin and runningbeet import
the metadata is modified on the original file:as well as populating it on the new file: