On Mon, Nov 16, 2020 at 03:26:42PM -0500, Phillip Susi wrote: > > Brian C. Lane writes: > > > Thanks for tracking that down. > > > > I'd like to figure out a nice way to deprecate them, I'm not sure if > > that's been done with a parted API before. > > > > As far as a timeframe goes, given the glacial pace of parted > > releases, I'm thinking a 4.0 release next December would be a good time > > to remove them. This leaves plenty of time to adjust. > > Ever since it was removed and restored last time I have thought that the > proper place for it to go is in dosfstools along with dosfsck etc, but > that package did not appear to be maintained either. Since it is > needed and there is no alternative, and we don't seem to have a bunch of > bug reports about it, then it can just continue to live in parted even > though nobody has been working on it. As the saying goes, if it ain't > broke, don't fix it.
Normally I'd agree. But after looking at how the hfs code works I have no idea if it is broken or not. We have some minimal testing of the code, but given how gnarly the hfs internals are I'm sure we don't test all the corner cases. The thing is, parted doesn't need it. The projects that do need it should take over maintaining it so that it gets the right set of eyes watching it and making sure that it doesn't rot. Brian -- Brian C. Lane (PST8PDT) - weldr.io - lorax - parted - pykickstart
