Archival for Conda is currently in development:
Software Heritage currently has a lister and a loader for Archlinux packages, but they load
binary packages (
.tar.gz); and need to be modified to load source packages instead
For every configured channel (
conda-forge, …) and every architecture
win-64, …), the Conda lister downloads
from which it extracts a list of package names. Then, from each of these package names,
it triggers a load for the origin
with the list of tarballs of that package.
There is a
_anaconda_depends package; what do we and should we do with it?
Source code from Conda is currently only archived on Software Heritage’s staging infrastructure. Metadata from Conda is currently not collected or indexed at all.