You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
the libbtrfsutil python bindings are pretty cumbersome. they're not in pypi, and they really can't be, because they need to be compiled against your kernel. this means they need to be installed from a system package manager, and virtualenvs need to be created with --system-site-packages, and can only be created with the system package manager's chosen python version. at minimum this creates speed bumps, at worst this could make installs impossible.
we have a dependency on the btrfs-progs CLI program directly, because this is currently the only way to access btrfs send.
it would be nice if we could just similarly use btrfs --format=json subvolume show instead of btrfsutil.subvolume_info().
the difficulty is that it seems support for btrfs --format=json subvolume show was only introduced in v6.5, which is pretty recent. a dependency v6.5 would restrict installations even more than the current setup. we could parse the output of --format=text, I suppose.
The text was updated successfully, but these errors were encountered:
the libbtrfsutil python bindings are pretty cumbersome. they're not in pypi, and they really can't be, because they need to be compiled against your kernel. this means they need to be installed from a system package manager, and virtualenvs need to be created with
--system-site-packages
, and can only be created with the system package manager's chosen python version. at minimum this creates speed bumps, at worst this could make installs impossible.we have a dependency on the
btrfs-progs
CLI program directly, because this is currently the only way to accessbtrfs send
.it would be nice if we could just similarly use
btrfs --format=json subvolume show
instead ofbtrfsutil.subvolume_info()
.the difficulty is that it seems support for
btrfs --format=json subvolume show
was only introduced in v6.5, which is pretty recent. a dependency v6.5 would restrict installations even more than the current setup. we could parse the output of--format=text
, I suppose.The text was updated successfully, but these errors were encountered: